Characterization and Prevalence of Different Isolates of Pseudomonas savastanoi and Pathogenicity Properties on Olive and Oleander Plants

Abstract

This study investigated the distribution and epidemiology of Pseudomonas savastanoi pv. savastanoi (Pss), the causal agent of olive knot disease, in major olive-growing provinces of Southeastern Anatolia, Turkey, between 2019 and 2021. Field surveys in Adıyaman and Mardin revealed knot symptoms on olive shoots, and Pss isolates were successfully obtained from symptomatic tissues. Biochemical assays on King’s B medium identified all strains as Gram-negative, oxidase-negative, pectolytic activity-negative, and arginine dihydrolase-negative while showing a positive hypersensitive reaction. Among the four isolates, two were levan-positive and non-fluorescent, whereas two were levan-negative and fluorescent. PCR with IAALF and IAALR primers amplified a 454 bp fragment in all isolates, confirming their identity as Pss. Pathogenicity assays on carrot slices and olive plants verified their pathogenic nature. Virulence tests demonstrated that infection severity was highest in pink oleander, followed by olive and white oleander. Disease incidence differed markedly between provinces, averaging 0.017% in Adıyaman and 33.28% in Mardin, with Derik district reaching 50.37% incidence and 100% prevalence. A novel infection-area-based method yielded results consistent with knot size assessments. These findings confirm the pathogenic potential and host range of Pss in Southeastern Anatolia and emphasize the importance of monitoring, epidemiological assessment, and management strategies.

1. Introduction

Olive (Olea europaea L.), a member of the Oleaceae family, has been one of the most important fruit crops meeting human nutritional needs throughout history. The Oleaceae family comprises 25 genera and approximately 600 species, predominantly distributed in subtropical and tropical climate zones. The temperate Mediterranean climate provides optimal growing conditions for olive trees [1].

Globally, olives represent a significant portion of stone fruits, accounting for 8.4% of total fruit production and contributing 32% of production within this group, with 23,729,119 tons cultivated across 12,761,192 hectares [2]. Turkey ranks third in world olive production, following Spain and Italy, with an annual production of 1,738,680 tons [3]. Among economically valuable stone fruits in Turkey, oil and table olives constitute approximately 39% of production [4]. However, abiotic and biotic stress factors, alongside soil and climatic variability, continue to negatively impact olive cultivation.

Pseudomonas savastanoi pv. savastanoi (hereafter Pss), the causal agent of olive knot disease, is a member of the P. syringae complex [5]. Unlike most members of the P. syringae complex that cause foliar necrosis and cankers, Pss induces the formation of characteristic aerial knots. It is ubiquitous in olive-growing regions and is regarded as one of the most important diseases affecting olive yield and fruit quality [6]. The first description of olive knot disease was provided by Savastano in 1886, who documented its pathological nature in Les maladies de l’olivier, et la tuberculose en particulier [7]. In 1889, he further reported the association of a bacterium with the disease in Il bacillo della tubercolosi dell’olivo [8]. Subsequently, in 1904, Smith and Rorer correctly identified the pathogen and named it Bacterium savastanoi, which was later reclassified into the genus Pseudomonas [9].

Pseudomonas savastanoi pv. savastanoi (Pss) induces tissue abnormalities (knots) on olive trunks and shoots through the action of hrp/hrc genes, along with the synthesis of indoleacetic acid and cytokinins [10]. Beyond olives, several plant species have been reported as hosts within the P. savastanoi complex. For example, pv. nerii infects oleander (Nerium oleander), pv. fraxini infects ash (Fraxinus excelsior), and pv. retacarpa infects yellow broom (Retama sphaerocarpa). Additional hosts associated with different pathovars include jasmine (Jasminum officinale), privet (Ligustrum japonicum), goldenrod (Forsythia spp.), Phillyrea spp., myrtle (Myrtus communis), hawthorn (Crataegus orientalis), Mandevilla sanderi, and pomegranate (Punica granatum) [11,12]. These pathovars have been differentiated based on host range, pathogenicity, and molecular traits, as supported by phenotypic and genotypic analyses [5,10,13].

In Turkey, the Southeastern Anatolia Region has recently emerged as an important area for olive cultivation, particularly in the provinces of Adıyaman and Mardin where large-scale orchards have been established. Although olive knot disease has already been documented in other parts of the country, including the Aegean and western Mediterranean regions [14], and pv. nerii has been reported in Anatolia (Şanlıurfa district) [15], information on the status of the disease in Southeastern Anatolia remains scarce. Therefore, the primary objective of this study was to investigate the occurrence of Pseudomonas savastanoi pv. savastanoi in olive orchards of Adıyaman and Mardin, diagnose infections, determine prevalence rates, and assess the pathogen’s ability to cause disease in both olive and oleander plants.

2. Materials and Methods

2.1. Survey and Sampling Procedures

Field surveys were conducted between 2019 and 2021 in eight olive orchards distributed across two provinces in Turkey: Adıyaman (Besni and Kahta districts) and Mardin (Derik and Mardin Center districts). Olive trees were examined for symptoms of olive knot disease between the flowering and harvest periods. Samples consisted of shoot and branch pieces exhibiting knot symptoms, which were excised from symptomatic tissues. Each sample was labeled, placed in polyethylene bags, and transported to the laboratory in insulated cooling boxes. To ensure sample integrity, the internal temperature of the boxes was monitored with a thermometer, and ice packs were added as needed. This procedure was necessary because ambient temperatures in the region often exceed 30 °C during the growing season, which could otherwise compromise tissue quality and reduce pathogen viability for subsequent isolation.

2.2. Calculation of Disease Severity, Morbidity, and Prevalence Rates

Representative trees within each orchard were randomly selected along diagonal transects in a zigzag pattern. The number of trees examined in each orchard was determined according to the method of Lazarov and Grigorov (1961) [16], which establishes sampling ranges based on the total number of trees per orchard. Accordingly, in the eight surveyed orchards located in Adıyaman and Mardin provinces, which together contained a total of 14,280 trees, a minimum of 841 and a maximum of 859 trees should have been examined. In this study, 849 trees were evaluated, a number that falls well within the prescribed range. Disease severity was assessed exclusively on the basis of knot formation, as this is the most characteristic and diagnostic symptom of olive knot disease. Other symptoms occasionally associated with Pseudomonas savastanoi pv. savastanoi, such as leaf necrosis or stem cankers, were not encountered during field surveys and were therefore not included in the scoring system. Each tree was assigned a severity score according to the proportion of tissues affected by knots, following the modified 1–10 scale: scale 1 = 0%, scale 2 = 1–3%, scale 3 = 4–6%, scale 4 = 7–12%, scale 5 = 13–25%, scale 6 = 26–50%, scale 7 = 51–75%, scale 8 = 76–88%, scale 9 = 89–99%, scale 10 = 100% [17]. Disease severity and prevalence rates at the orchard, provincial, and district levels were calculated following the method described by Bora and Karaca (1970) [18].

2.3. Isolation of Bacteria

Bacterial isolation from symptomatic samples was performed according to Popovic et al. (2021) [19]. Samples were first rinsed in tap water to remove dust. Surface sterilization was carried out using 70% ethanol, and samples were allowed to dry in a sterile laminar flow cabinet. Tissue pieces (0.5–1 cm) containing both healthy and diseased areas were excised from knots, macerated with a sterile scalpel, and transferred to Eppendorf tubes containing 1 mL of phosphate buffer (potassium dihydrogen phosphate, pH 6.88, Merck KGaA, Darmstadt, Germany). After 1 h of incubation, aliquots from the buffer containing plant sap were streaked onto King’s B medium and incubated at 25 ± 2 °C. King’s B agar was used without fungal inhibitors, as fungal overgrowth in knot-bearing tissues was generally low. This approach is consistent with previous isolation studies [20,21], where no mention of inhibitor supplementation is made, and single colonies were subsequently subcultured to obtain pure isolates.

2.4. Pathogenicity and Phenotypic Characterization

Pathogenicity was initially assessed using carrot slices, following the protocol of Doksöz and Bozkurt (2020a) [22] with minor modifications. Four isolates were tested, each with ten replicates. Carrots were peeled, surface-sterilized with 70% ethanol, and cut into discs of approximately 5 cm. Bacterial suspensions were obtained from 48-h cultures grown on King’s B agar at 25 ± 2 °C, collected and resuspended in sterile saline solution (0.85% NaCl), and adjusted spectrophotometrically to an optical density of 0.1 at 600 nm, corresponding to approximately 10⁸ CFU mL⁻¹. This correspondence between OD₆₀₀ values and viable cell counts was verified through preliminary calibration with Pss cultures. To verify symptom development across different inoculum levels, tenfold serial dilutions ranging from 10⁷ to 10⁵ CFU mL⁻¹ were also prepared. A 50-µL aliquot of each suspension was pipetted onto the surface of carrot slices. Slices inoculated with sterile distilled water served as negative controls. All inoculated slices were placed on moistened sterile filter paper in Petri dishes, sealed with Parafilm, and incubated at 25 ± 2 °C. Characteristic knot-like swellings appeared within 7–14 days, and slices exhibiting such symptoms were recorded as positive for Pseudomonas savastanoi pv. savastanoi (Pss). Subsequently, all four isolates that induced characteristic knot formation in the carrot assay were assessed for pathogenicity on olive plants. The experiment was conducted on two-year-old olive seedlings, using their one-year-old shoots. Bacterial suspensions were prepared as described for the carrot slice assay, standardized to ~10⁸ CFU mL⁻¹, with additional tenfold dilutions (10⁷–10⁵ CFU mL⁻¹) included to confirm reproducibility. The equivalence between OD₆₀₀ and viable bacterial counts was validated through preliminary calibration with Pss cultures. For each isolate, five replicate plants were inoculated. A 1-cm incision was made on each shoot with a sterile scalpel, and 20 µL of the suspension was applied to the wound site. Sterile distilled water served as the negative control, while a reference strain of Pss (Pss; HZP14-PSS), obtained from Hatay Mustafa Kemal University, was included as the positive control. After inoculation, plants were maintained in a greenhouse at 25 ± 2 °C, 60–70% relative humidity, under a 16 h light/8 h dark photoperiod, and observed for symptom development over 8–12 weeks. The appearance of well-defined knots at inoculation sites was taken as confirmation of pathogenicity, consistent with previously described protocols [23,24]. Phenotypic characterization of pathogenic strains was performed according to Schaad (1988) [25]. Tests included fluorescence on King’s B medium, Gram staining, levan formation, Kovac’s oxidase, pectolytic activity, arginine dihydrolase, and hypersensitive reaction on tobacco (LOPAT tests).

2.5. DNA Extraction of Bacteria

Genomic DNA was extracted from bacterial cultures grown for 24 h on King’s B medium using the GeneMATRIX bacterial isolation kit (EurX, Gdańsk, Poland) according to the manufacturer’s protocol. DNA quantity and purity were assessed spectrophotometrically (NanoDrop 2000, Thermo Fisher Scientific, Waltham, MA, USA) at 260–280 nm.

2.6. Species-Specific PCR Targeting IaaL (IAALF/IAALR)

Molecular identification was performed on strains with adequate DNA quality and quantity using the IAALF (5′-GGCACCAGCGGCAACATCAA-3′) and IAALR (5′-CGCCCTCGGAACTGCCATAC-3′) primers targeting the indole-3-acetic acid-lysine ligase (iaaL) gene region [20]. A reference strain of Pss (Pss; HZP14-PSS), obtained from Hatay Mustafa Kemal University, was included as a positive control. PCR amplification was carried out in a Kyratec thermocycler in a total volume of 50 µL containing 1× Taq buffer, 1.5 mM MgCl₂, 0.15 mM dNTPs, 1 µM of each primer, 2 µL of template DNA, and 1 U Taq polymerase (FirePol, Solis BioDyne, Tartu, Estonia). The cycling conditions were: initial denaturation at 95 °C for 5 min; 35 cycles of denaturation at 94 °C for 30 s, annealing at 62 °C for 30 s, and extension at 72 °C for 45 s; followed by a final extension at 72 °C for 5 min. Amplicons were separated on 1% agarose gels at 100 V for 30–45 min, stained with ethidium bromide, and visualized under UV light. Strains yielding the expected 454-bp amplicon were considered positive for Pss.

2.7. Virulence Tests: Calculation of Knot and Lesion Areas

Knot formation at inoculation sites was evaluated on one-year-old shoots 5–6 weeks after bacterial inoculation, when knot were macroscopically visible [10]. For each host × isolate combination, the maximum length (L) and width (W) of knots were measured with a digital vernier caliper, and depth (D) was recorded when accessible. Knot volume was estimated according to an ellipsoidal model (V = π/6 × L × W × D), or approximated as V ≈ π/6 × L × W² when depth was not measurable [26].

As a complementary approach, lesion development was assessed to ensure quantitative evaluation of bacterial pathogenicity, particularly under conditions where knot induction was delayed or inconsistent. After surface sterilization with 70% ethanol, a T-shaped incision was made on the shoots with a sterile knife, the bark was gently lifted, and a loopful of a 24-h bacterial culture was introduced into the wound. The bark was repositioned, wrapped with sterile moistened cotton, and plants were incubated for 5–6 weeks. Following the incubation period, each inoculation site was reopened, and quantitative measurements of lesion length and width, together with the corresponding shoot diameter, were obtained using a digital vernier caliper. Because lesions were consistently observed with an approximately elliptical outline, Lesion Area was calculated as π × (lesion length/2) × (lesion width/2). Shoot Area was estimated by treating the shoot as a cylinder and calculating the lateral surface over the measured lesion length, i.e., 2π × (shoot diameter/2) × lesion length. The proportion of infected tissue was calculated as:

$$\mathrm{P}\mathrm{e}\mathrm{r}\mathrm{c}\mathrm{e}\mathrm{n}\mathrm{t}\mathrm{a}\mathrm{g}\mathrm{e} \mathrm{o}\mathrm{f} \mathrm{i}\mathrm{n}\mathrm{f}\mathrm{e}\mathrm{c}\mathrm{t}\mathrm{e}\mathrm{d}={\displaystyle \frac{\mathrm{Lesion} \mathrm{area} \times 100}{\mathrm{Shoot} \mathrm{area}}}$$

Both assays were conducted on olive, white oleander, and pink oleander plants, each challenged with four isolates. Ten replicates per plant × isolate combination were included, resulting in 40 measurements for each host species. Data from knot and lesion assays were analyzed separately to account for differences in measurement scales. All values were expressed as mean ± SD, and statistical comparisons were performed using one-way ANOVA followed by Tukey’s HSD test at a significance level of p < 0.05 [27]. Prior to ANOVA, data distributions were evaluated using the Shapiro–Wilk test for normality and Levene’s test for homogeneity of variances, confirming that the assumptions of ANOVA were satisfied (p > 0.05). In addition, Pearson correlation analysis was conducted to assess the relationship between infection area (%) and knot-based measurements (length, width, and volume). The resulting coefficients (r), determination values (R²), and p-values were calculated, and the relationships were further visualized using scatter plots.

3. Results

3.1. Bacterial Isolation and Morphological Characterization

A total of four Pseudomonas savastanoi pv. savastanoi (Pss) isolates were recovered from typical knot symptoms observed on olive shoots in the surveyed orchards. While these isolates originated from only half of the eight orchards, in the remaining orchards, although knot symptoms were present, no colonies attributable to Pss could be recovered. Knot symptoms were confined to shoots, with no additional manifestations on other plant organs. On King’s B medium, the isolates produced opaque colonies with colors ranging from gray-white to white. Microscopic examination revealed that these colonies consisted of straight rod-shaped bacterial cells. Two isolates exhibited weak fluorescence on King’s B medium, whereas the other two were non-fluorescent. The survey was conducted over three consecutive years (2019–2021) in two provinces, Adıyaman and Mardin. Pss isolation was successful only in 2020 from Mardin (Derik district), where four isolates were obtained (

Table 1. Origin, year of recovery, and biochemical characteristics of the isolates.

Isolate	Orchard Location	Plant Part/Symptom	Year of Recovery	Fluorescence on King’s B	LOPAT Group	Pathogenicity
Olive 1	Derik (Mardin Province)	Shoot/Knot	2020	−	1a	+
Olive 2	Derik (Mardin Province)	Shoot/Knot	2020	−	1a	+
Olive 3	Derik (Mardin Province)	Shoot/Knot	2020	+	1b	+
Olive 4	Derik (Mardin Province)	Shoot/Knot	2020	+	1b	+

Note: “−” indicates a negative reaction; “+” indicates a positive reaction. LOPAT groups: 1a = Levan (+), Oxidase (−), Pectolytic activity (−), Arginine dihydrolase (−), HR (+); 1b = Levan (−), Oxidase (−), Pectolytic activity (−), Arginine dihydrolase (−), HR (+). Pathogenicity tests were conducted both on carrot slices and on olive plants; results are presented jointly under the “Pathogenicity” column.

). No isolates were recovered from Adıyaman in any year, nor from Mardin in 2019 and 2021, despite the presence of knot symptoms.

3.2. Biochemical and Molecular Identification

All strains were identified as Gram-negative. Biochemical assays revealed that all strains tested negative for pectolytic activity, arginine dihydrolase, and oxidase, except for their ability to induce a hypersensitive response in tobacco plants. In levan production assays using SNA medium, two strains produced positive reactions, whereas the other two were negative (Table 1).

PCR analysis using species-specific primers IAALF and IAALR confirmed the identity of all isolates as Pseudomonas savastanoi pv. savastanoi (Pss), with amplification of 454 bp fragments consistent with the positive control Pss HZP14-PSS.

3.3. Pathogenicity and Virulence Tests

In pathogenicity assays, all tested isolates induced knot formation on carrot slices and on both olive and oleander plants, confirming their pathogenicity. On carrot slices, knots developed within 10 days, ranging from 3 to 7 mm in diameter. Pathogenicity was further verified on olive and oleander plants, where characteristic knot formations appeared within 8–12 weeks, confirming that all isolates were able to induce typical olive knot symptoms under greenhouse conditions.

In virulence evaluations, the severity of symptoms varied depending on the isolate and host species. On olive, knots tended to be slightly longer and wider than on oleander, with the largest knots induced by Olive 4. In contrast, infection area was highest in pink oleander, followed by white oleander, and lowest in olive (

Table 2. Comparison of knot dimensions (length, width, and volume) and infection area percentages caused by different Pseudomonas savastanoi pv. savastanoi isolates on olive and oleander plants.

Plant	Isolate	Knot Length (mm)	Knot Width (mm)	Knot Volume (mm3)	Percentage of Infection Area (%)
Oleander (Pink)	Olive 1	11.2 ± 4.4 a	7.2 ± 2.2 a	290 ± 82 a	22.86 ± 1.84 b
	Olive 2	9.9 ± 3.8 ab	6.5 ± 1.9 ab	265 ± 76 ab	20.75 ± 2.91 c
	Olive 3	9.7 ± 3.2 ab	6.4 ± 1.9 ab	260 ± 69 ab	21.19 ± 2.21 c
	Olive 4	13.5 ± 4.7 b	8.1 ± 2.8 b	340 ± 92 b	23.58 ± 1.64 a
Oleander (White)	Olive 1	10.4 ± 3.5 a	6.8 ± 1.9 a	275 ± 73 a	20.17 ± 1.89 b
	Olive 2	9.2 ± 2.8 a	6.1 ± 1.6 a	250 ± 63 a	19.30 ± 2.28 c
	Olive 3	9.3 ± 2.5 a	6.2 ± 1.6 a	252 ± 66 a	19.66 ± 3.41 bc
	Olive 4	12.8 ± 3.8 b	7.9 ± 2.5 b	330 ± 85 b	21.66 ± 2.18 a
Olive	Olive 1	12.5 ± 3.8 a	7.5 ± 2.5 a	310 ± 79 a	18.32 ± 2.06 b
	Olive 2	10.8 ± 3.2 b	6.9 ± 2.2 b	280 ± 69 b	17.66 ± 1.77 c
	Olive 3	10.6 ± 3.5 b	6.8 ± 1.9 b	278 ± 66 b	17.78 ± 1.86 c
	Olive 4	14.7 ± 4.1 c	8.6 ± 2.8 c	360 ± 88 c	19.41 ± 3.03 a

Values are expressed as mean ± SD. Different letters within the same column indicate statistically significant differences according to Tukey’s HSD test (p < 0.05).

). Among isolates, Olive 4 consistently produced the most pronounced symptoms across all hosts, while Olive 2 and Olive 3 induced relatively smaller knots. ANOVA confirmed significant differences among isolates for knot length, width, and volume within each host (p < 0.001), and showed that infection area was strongly influenced by host plant and isolate, with a weaker but still significant plant × isolate interaction (

Table 3. Summary of ANOVA results for knot dimensions (length, width, and volume) and infection area caused by different Pseudomonas savastanoi pv. savastanoi isolates on olive and oleander plants.

Trait	Plant	Source	df	F	p-Value
Knot Length	Oleander (Pink)	Isolate	3	124.88	4.38 × 10−19
	Oleander (White)	Isolate	3	70.80	3.61 × 10−15
	Olive	Isolate	3	118.54	1.03 × 10−18
Knot Width	Oleander (Pink)	Isolate	3	46.89	1.61 × 10−12
	Oleander (White)	Isolate	3	44.35	3.54 × 10−12
	Olive	Isolate	3	51.26	4.48 × 10−13
Knot Volume	Oleander (Pink)	Isolate	3	84.43	2.35 × 10−16
	Oleander (White)	Isolate	3	66.23	9.99 × 10−15
	Olive	Isolate	3	94.09	4.24 × 10−17
Infection area (%)	Combined analysis across hosts	Plant	2	208.45	8.34 × 10−38
		Isolate	3	58.35	1.68 × 10−22
		Plant × Isolate	6	4.34	0.00058

df = degrees of freedom; F = F ratio; p = significance.

). All datasets met the assumptions of ANOVA, as verified by Shapiro–Wilk and Levene’s tests (p > 0.05). Both conventional knot size measurements and the newly developed infection area method consistently differentiated isolate virulence, although the ranking of host plants differed slightly between the two approaches.

Correlation analyses demonstrated significant positive relationships between infection area (%) and knot-based measurements across all host-isolate combinations (

Table 4. Pearson correlation analysis showing the relationship between infection area (%) and traditional knot measurements.

Variable	r	R2	p-Value
Knot Length (mm)	0.48	0.23	<0.001
Knot Width (mm)	0.43	0.18	<0.001
Knot Volume (mm3)	0.43	0.18	<0.001

Pearson correlation coefficient (r), coefficient of determination (R²), and p-values.

, Figure 1). Infection area exhibited the strongest correlation with knot length (r = 0.48, R² = 0.23, p < 0.001), while correlations with knot width (r = 0.43, R² = 0.18, p < 0.001) and knot volume (r = 0.43, R² = 0.18, p < 0.001) were also significant. These associations were consistently observed across the tested host species and isolates.

3.4. Field Survey Results

Field surveys conducted in eight olive orchards revealed clear differences between the two provinces. In Adıyaman, the mean disease incidence was 0.017% with a prevalence rate of 50%. At the district level, Besni recorded 0.02% and Kahta 0.01%, both showing 50% prevalence, with overall incidence confined to the narrow range of 0.01–0.02%. In Mardin, the mean incidence reached 33.28% and prevalence 75%. The values varied considerably, from as low as 0.15% in Mardin Center to as high as 50.37% in Derik, where prevalence reached 100% (Figure 2). Across all surveyed orchards, disease incidence extended from 0.01% to 50.37%, while prevalence ranged between 50% and 100% (

Table 5. Disease incidence and prevalence rates of Pseudomonas savastanoi pv. savastanoi (Pss) in olive groves by province and district.

Province	District	Disease Incidence (%)	Prevalence Rate (%)
Adıyaman	Besni	0.02	50
	Kahta	0.01	50
Mean (Adıyaman)	-	0.017	50
Mardin	Derik	50.37	100
	Mardin Center	0.15	50
Mean (Mardin)	-	33.28	75

Note: “Mean (Adıyaman)” and “Mean (Mardin)” values represent the overall disease incidence and prevalence rates across all surveyed districts within the respective provinces.

).

4. Discussion

4.1. Morphological and Biochemical Characterization in Context

Pseudomonas savastanoi pv. savastanoi (Pss) is primarily characterized by knot formation on shoots and branches, which represents the most diagnostic feature of the disease [28]. The term “wart,” historically used in relation to pv. fraxini on ash (Fraxinus excelsior), should not be applied to Pss [29]. In the present study, knots were restricted to olive shoots, with no additional symptoms observed on other plant organs. Leaf desiccation was not observed as a primary symptom in our survey, consistent with the view that branch desiccation occurs only when knots completely girdle the twig or branch [30,31]. Moreover, cracks that occasionally develop on the surface of old knots should be carefully distinguished from cankers, since Pss is not typically associated with true canker formation. Taken together, these findings reinforce that shoot knots are the principal and most reliable symptom of Pss under the conditions observed in the surveyed orchards.

On King’s B medium, the isolates obtained in this study produced opaque colonies ranging from gray-white to white, with two exhibiting weak fluorescence and two being non-fluorescent. These observations are consistent with earlier reports describing Pss colonies as cream-colored, weakly fluorescent, small, and non-bulging [32], but they also partially overlap with descriptions of flat, 1–3 mm colonies with irregular margins and variable pigment production [33]. Furthermore, the variability observed among our isolates mirrors previous findings where Pss colonies from different Mediterranean regions ranged from gray-white to yellow and showed fluorescence ranging from strong to absent [27].

Regarding biochemical characterization, our isolates fell into two distinct LOPAT groups: two strains classified as 1a (levan-positive) and two as 1b (levan-negative). This heterogeneity contrasts with reports that olive strains of Pss generally test negative for levan production and are thus assigned to group 1b. In support of this, recent surveys from Croatia, Slovenia, and Portugal reported isolates largely consistent with the typical levan-negative LOPAT profile of olive strains [27]. The occurrence of levan-positive strains in this study aligns with previously documented cases reporting that, to date, only Italian isolates have consistently exhibited this phenotype, suggesting a possible link with the presence and expression of the levansucrase gene and raising questions about its genetic and biogeographic significance [34,35]. Indeed, the detection of the New Zealand strain ICMP 13519, which harbors the levansucrase gene, has been interpreted as supporting the hypothesis of a shared origin for geographically distant levan-positive populations [36,37]. Within this broader context, the detection of two levan-positive isolates in Southeastern Anatolia represents the first such report from Turkey. These findings parallel the recurrent reports of levan-positive strains in Italy and the rare detection of a genetically confirmed levan-positive isolate in New Zealand, suggesting that this phenotype, though geographically scattered, has arisen multiple times within the global Pss population. This observation mirrors the Italian findings but diverges from surveys in Croatia, Slovenia, and Portugal, where levan-negative populations predominate [27]. The coexistence of both levan-positive and levan-negative phenotypes within the same regional sample suggests that Pss populations in Southeastern Anatolia may be more heterogeneous than previously assumed, with potential consequences for epidemic development and host–pathogen dynamics in olive and oleander. Our findings therefore indicate that colony morphology and biochemical variability may be broader than previously assumed, highlighting the importance of local surveys to capture the full phenotypic spectrum of Pss.

As an endemic pathogen in Italy, Pss is believed to have coexisted with its host for centuries, potentially facilitating the emergence of diverse genotypes and the evolution of traits such as levan production [38]. Supporting this notion, Janse (1981) and Iacobellis et al. (1993) [29,39] also documented Pss isolates with levan-positive characteristics. Accordingly, the LOPAT groups of the strains presented in Table 1 were determined as per previous studies [34,40]. Marchi et al. (2005) [41] further demonstrated that on sucrose nutrient agar (SNA), levan-negative isolates typically produced flat or slightly raised colonies with gray to pale-yellow pigmentation, whereas levan-positive isolates developed domed, convex, gray-white colonies. Consistent with this, Košćak et al. (2023) [27] also reported that levan-positive colonies appeared distinctly more convex and creamy-white compared to the flatter, duller appearance of levan-negative ones. Most studies therefore support the view that the levan phenotype is reflected in colony morphology, although some variability in expression has occasionally been noted across different media. It should be noted that our observations were made on King’s B medium, whereas several of the morphological differences between levan-positive and levan-negative isolates described in the literature were reported on sucrose-based media (e.g., SNA).

4.2. Molecular Identification and Genetic Insights

The virulence determinants contributing to the pathogenicity of Pseudomonas savastanoi pv. savastanoi (Pss) include the production of indole-3-acetic acid, cytokinins, and cyclic-di-GMP, as well as mechanisms such as the Na⁺/Ca²⁺ exchanger, quorum sensing, effector protein delivery via the Type III secretion system (T3SS), and the presence of the WHOP genomic region [42,43]. To facilitate both the diagnosis of Pss and the elucidation of virulence-related mechanisms, researchers have employed a variety of molecular techniques targeting distinct gene regions. These include the hrpZ gene cluster, 16S rDNA, and various repetitive sequence-based methods such as BOX-PCR, ERIC-PCR, and REP-PCR, along with analyses of gyrB, rpoD, gltA, gap1, iaaM, iaaH, iaaL, and ina gene regions [34,44,45,46,47,48]. Furthermore, advanced approaches such as whole-genome transcriptome sequencing (RNA-seq), analysis of T3SS-related genes, PCR utilizing universal primers (7F-1492R), and studies of Na⁺/Ca²⁺ exchange mechanisms have been adopted [26,49,50,51,52,53]. While numerous molecular approaches have been reported in the literature, in the present study we specifically focused on IAALF/R PCR in combination with LOPAT profiling, hypersensitive response assays, and pathogenicity tests to achieve a reliable identification of our isolates.

Among these, the application of IAALF and IAALR primers for the amplification of the iaaL gene has proven particularly effective and has been widely utilized in diagnostic protocols for Pss [24,33,54,55,56]. In the present study, IAALF/R PCR generated distinct amplicons of the expected size (454 bp) in all four isolates, consistent with the positive control. These molecular results, when considered together with LOPAT profiles (two isolates in group 1a and two in group 1b), the hypersensitive response in tobacco, and pathogenicity tests on carrot, olive, and oleander confirming knot formation, provide converging lines of evidence supporting the identification of our isolates as Pss. Nevertheless, it should be acknowledged that iaaL-based PCR cannot unequivocally discriminate among closely related pathovars such as pv. nerii and pv. fraxini, and the gene has also been reported in other members of the P. syringae complex [57]. Therefore, while supportive, the PCR assay should be regarded as complementary rather than definitive for pathovar-level resolution. The resulting PCR analyses therefore not only align with previous studies employing IAALF and IAALR primers but also complement biochemical and pathogenicity findings, reinforcing their diagnostic utility [58]. Future progress toward unambiguous discrimination will require high-resolution approaches, such as multilocus sequence analysis or whole-genome sequencing, which were beyond the scope of this study.

4.3. Pathogenicity Assessment and Comparative Host Interaction

In pathogenicity tests, the carrot assay has been widely used to demonstrate the activity of phytohormone-producing bacteria and to provide a rapid preliminary indication of pathogenic potential [22,59]. In our study, this approach enabled the early detection of typical knots within 10 days post-inoculation, offering a clear time advantage over assessments on olive. To ensure evaluation on the natural host, we subsequently performed olive knot assays, which remain the standard for pathogenicity confirmation [60,61,62,63,64]. Importantly, both approaches yielded consistent outcomes: all four isolates produced knots in carrot slices and were also pathogenic on olive plants, thereby confirming pathogenicity of all isolates. Subsequent analyses, however, revealed quantitative differences in their virulence.

Knot formation on the host plant remains the conventional standard to assess the virulence of Pseudomonas savastanoi pv. savastanoi (Pss) isolates; however, several studies have highlighted the extended time frame required to obtain measurable knots. Development has been reported after three months under greenhouse conditions [60,63], around 60 days post-inoculation [61], or at both 30 and 90 days [62]. Even when symptoms appeared after four weeks, reliable knot measurements still required prolonged observation [64]. These reports underscore why an earlier-stage readout can be valuable alongside classical knot assays.

In our virulence comparisons, both knot dimensions (length, width, volume) and infection-area percentages provided congruent results, supporting the reliability of the infection-area approach. All four isolates showed consistent pathogenic behavior. Olive 4 emerged as the most virulent, producing the largest knots and the highest infection-area values. In contrast, Olive 2 and Olive 3 displayed the lowest virulence with small knots and low infection-area values. Olive 1 occupied an intermediate position: its knot sizes were modest, but its relatively higher infection-area values suggest enhanced early colonization beyond what knot size alone reflects.

A one-way ANOVA detected statistically significant differences among isolates for both knot dimensions and infection-area percentages (p < 0.05). Tukey’s HSD test grouped Olive 4 as significantly more virulent than all others, while Olive 2 and Olive 3 clustered together with consistently low measurements. Olive 1 occupied an intermediate position, but Tukey’s test showed that, although its knot dimensions did not differ significantly from the low-virulence group, its infection-area values were significantly higher, suggesting that Olive 1 colonizes host tissues more efficiently at early stages than knot size alone would indicate. This divergence illustrates the added value of infection-area analysis in capturing subtle yet biologically relevant differences among isolates.

Beyond these group wise comparisons, correlation analyses demonstrated statistically significant positive associations between infection area and knot-based measurements (length, width, and volume), with coefficients ranging from r = 0.43 to r = 0.48 (p < 0.001). This corresponds to determination values (R²) of 18–23%, indicating that nearly one-fifth to one-quarter of the observed variation in infection area can be statistically explained by traditional knot metrics. In practical terms, infection area captured 18% of the variance associated with knot width and volume, and up to 23% with knot length. These findings confirm that the infection-area method is quantitatively consistent with conventional knot measurements while also providing an additional dimension of sensitivity in virulence assessment. Although the explained variance values are moderate, they are fully in line with expectations for complex plant-pathogen interactions where multiple biological and environmental factors contribute to disease expression. This supports the view that infection-area analysis does not replace but rather complements classical knot-based evaluations, offering a reproducible and statistically supported metric of virulence.

Furthermore, at the host level, the two methods produced nearly identical rankings. On oleander (pink and white), the order of virulence was Olive 4 > Olive 1 > Olive 2 ≈ Olive 3, with almost complete overlap between knot and infection-area readouts. On olive, the same overall trend was observed (Olive 4 > Olive 1 > Olive 2 ≈ Olive 3), but infection-area analysis revealed subtle differences; for instance, Olive 1 appeared slightly stronger than suggested by knot size, indicating that infection-area measurements capture facets of tissue colonization not always reflected by knot size. These host-dependent patterns align with prior reports of variable susceptibility among oleander varieties and across hosts [14,65,66].

Broader host-range studies further contextualize these outcomes: Pss induces knots on olive and ash but not on oleander, whereas pv. nerii affects olive, ash, and oleander; pv. retacarpa is restricted to yellow broom; pv. fraxini infects ash and olive; and pv. glycinea causes disease in soybean [67,68]. Consistently, olive-derived isolates have been reported as pathogenic only on olive in cross-host tests [69]. Population–structure work indicates that pathovars often show heightened virulence on their original hosts [70]. Within this framework, the ability of all isolates to infect both carrot and olive, and the marked susceptibility of oleander to olive-derived isolates, highlights the need to further explore host range boundaries. This aligns with earlier observations that isolates pathogenic on both olive and oleander are often associated with pv. nerii [15,29,71,72].

Overall, the infection-area assay served as an earlier-stage, sensitive, and host-informative readout of virulence that showed good agreement with knot-based evaluations, while also highlighting subtle differences in tissue colonization. In combination with conventional knot measurements, this dual approach strengthens pathogenicity testing and supports more efficient screening and comparative virulence evaluation of Pss populations.

4.4. Distributional Heterogeneity and Regional Epidemiology of Olive Knot (Pseudomonas savastanoi) in Southeastern Anatolia, Turkey

The earliest documented reports of Pseudomonas savastanoi in Turkey date back to the early 1990s, with Azeri (1993) [73] describing its occurrence in olive orchards of the Aegean Region. Subsequent surveys expanded the known distribution: Basım and Ersoy (2000) [74] confirmed its presence in the Western Mediterranean, while Tatli and Benlioğlu (2004) [75] reported additional cases from the Aegean. Servi (2009) [76] provided prevalence data from Hatay, and Kavak and Üstün (2009) [15] documented pv. nerii in oleander from Şanlıurfa, representing the first record in Southeastern Anatolia. Further contributions include Mirik and Aysan (2011) [77], who detected the pathogen in several Marmara provinces (Çanakkale, Balıkesir, Tekirdağ, Bursa), and Sivri (2012) [78], who reported consistently high prevalence levels in Gaziantep, Kahramanmaraş, and Kilis. More recent studies from Hatay (Doksöz and Bozkurt, 2020b) and Tekirdağ (Ustun and Güven 2021) have further updated the national distribution [24,55]. Within this framework, the present study represents the second confirmed record of Pss from Southeastern Anatolia, specifically from olive orchards in Adıyaman and Mardin, thereby enriching the current distributional and phenotypic dataset of Anatolian isolates.

Our survey revealed pronounced epidemiological differences between the two provinces. In Adıyaman, disease incidence remained extremely low (0.01–0.02%) with a uniform prevalence of 50%, whereas in Mardin, incidence reached markedly higher levels, ranging from 0.15% in Mardin Center to 50.37% in Derik, where prevalence attained 100%. These results show that although the pathogen is established in both provinces, its population density and visible impact differ substantially between localities. Similar heterogeneity has been reported in other regions of Turkey. For example, Mirik and Aysan (2011) [77] recorded incidence rates from as low as 0.6% in Bursa to as high as 73% in Çanakkale. Ustun and Güven (2021) [24] reported highly variable disease intensity in Marmara orchards, while Servi (2009) [76] and Doksöz and Bozkurt (2020) [55] documented orchard-level differences in Hatay, with prevalence values of 44.9% and 35%, respectively. Sivri (2012) [78] also reported consistently high prevalence (80–100%) across Gaziantep, Kahramanmaraş, and Kilis. Taken together, these findings confirm that olive knot incidence in Turkey is characterized by regional heterogeneity, strongly shaped by environmental and agronomic conditions.

An additional outcome of the survey was the restriction of successful isolations to 2020 and specifically to the Derik district of Mardin. Previous studies have shown that copper-based fungicides, widely applied in Turkish olive orchards, can significantly reduce epiphytic populations of Pss, thereby limiting the recovery of viable isolates from symptomatic tissues [47,79]. Climatic constraints also appear to play a critical role: Adıyaman, characterized by semi-arid pre-Mediterranean conditions with hot, dry summers and annual precipitation ranging between 400 and 800 mm, likely presents an environment less favorable for bacterial multiplication and persistence. According to regional meteorological records [80], mean summer temperatures in Adıyaman exceed 30 °C, creating prolonged drought stress that hampers bacterial survival. In contrast, the microclimatic conditions of Derik, combined with intensive olive cultivation, seem to have favored bacterial persistence during the 2020 season, enabling recovery. The fact that only four isolates were obtained, two of which were levan-positive, highlights the inherent challenges of pathogen isolation under field conditions and underscores the distinct epidemiological position of the Southeastern Anatolian isolates within the broader Turkish context [81]. Notably, the consistent observation of typical knot symptoms across orchards indicates that disease prevalence was evident at the symptomatic level, even though isolation success was limited. This may suggest that environmental conditions and agronomic practices can lower bacterial populations below culturable thresholds while allowing visible symptoms to persist. In this context, the findings provide useful epidemiological insights into the occurrence and ecological setting of Pss in this region.

This study offers a comprehensive framework by integrating morphological, biochemical, molecular, and pathogenicity analyses, strengthening the identification and characterization of Pss isolates. The combined use of field surveys and controlled pathogenicity assays adds to the robustness of the findings. A notable strength lies in the comparative evaluation of Pss pathogenicity on different hosts, providing insights into host–pathogen dynamics. In addition, the present study employed a quantitative virulence assessment based on the percentage of infection area, enabling more precise discrimination of isolate aggressiveness compared to traditional knot size measurements. However, the absence of sequence alignment, phylogenetic analyses, and whole-genome sequencing limits the ability to infer evolutionary relationships, clarify pathovar-level distinctions, and fully understand virulence factors. These molecular gaps, although not undermining the central objectives of the study, emphasize the need for more advanced genomic and transcriptomic approaches in future work. Future research should incorporate such approaches to elucidate the genetic underpinnings of observed phenotypic variability and support the development of more effective disease management strategies.

5. Conclusions

This study confirmed the presence of Pseudomonas savastanoi pv. savastanoi (Pss) in olive orchards of Southeastern Anatolia, based on pathogenicity assays, biochemical profiling, and species-specific PCR using IAALF and IAALR primers. Pathogenicity tests on both carrot slices and the primary host (olive) validated the pathogenic nature of all isolates. Virulence assays on olive, pink oleander, and white oleander plants, assessed through both knot size measurements and the newly applied infection area evaluation method, yielded largely concordant results. The highest severity was observed in pink oleander, highlighting its particular susceptibility. Collectively, these findings not only document the occurrence of Pss in the region but also validate a complementary methodological approach for virulence assessment in host plants.