Plant material consisted of 52 wild potato accessions grown from true potato seeds obtained from the U.S. National Plant Germplasm System (NPGS) in Wisconsin, USA. The introductions belong to the Solanum sect. Petota diversity panel (Supplementary Table 1). This panel represents the germplasm of tuber-bearing Solanum species exhibiting morphological variations (Supplementary Figure 1) (Hardigan et al., 2015). Solanum tuberosum L. var. Atlantic (chip processing market class) was used as a susceptible control in all experiments and was initially propagated from certified disease-free seed tubers. For in vitro germination and propagation, the Solanum botanical seeds were pretreated with 10% (w/v) gibberellic acid (GA3) overnight in a 1.5 ml microcentrifuge tube to break dormancy and enhance seed germination. Furthermore, seeds were surface sterilized in 70% (v/v) ethanol for 3 min and then with 10% (v/v) bleach and 2% (v/v) Tween^® 20 (Sigma-Aldrich, St. Louis, MO, USA) for 10 min, followed by rinsing four times with sterile water. Sterilized seeds were placed on a sterile wetted Whatman™ paper inside a sterile 100 × 25 mm Petri dish and kept at 22°C in the dark for 1 week. Germinated seeds were grown on Murashige and Skoog (Murashige and Skoog, 1962) solid media supplemented with pre-made MS vitamins (Caisson Labs, North Logan, Utah, USA), 2% sucrose (w/v), and 2 g/L of Gelrite™ (Research Products International, Mt. Prospect, IL, USA), pH 5.8. When plants were 4 weeks old (about 8 cm tall), micropropagation was done using internode cutting as explants; sub-culturing every 4 weeks was necessary to increase plant material for each wild potato accession for screening against ZC. All explants were maintained in a temperature-controlled growth chamber at 22°C under 14-h light/10-h dark photoperiod.

Potato psyllid (Bactericera cockerelli Šulc.) colonies consisted of CLso-free (CLso⁻) and CLso-positive (CLso⁺) haplotype B, which were obtained and reared for several generations at Texas A&M AgriLife Research and Extension Center in Weslaco in 60 × 60 × 60 cm nylon mesh cages (Bugdorm, BioQuip Products Rancho Dominguez, CA, USA). Psyllid colonies were reared and maintained on potato plants (Atlantic) and periodically diagnosed for the presence of CLso (>90% CLso⁺) and the absence of CLso (CLso⁻) by polymerase chain reaction (PCR) using genomic DNA isolated from adult psyllids (Sengoda et al., 2014) and primers specific to the 16s rDNA of CLso (OA2-F and OI2c-R) (Supplementary Table 2) (French-Monar et al., 2010). PCR conditions were given as follows: one denaturing cycle at 95°C for 30 s; 35 cycles each at 95°C for 30 s, 68°C for 30 s, and 68°C for 2 min; and a final extension cycle at 68°C for 5 min. PCR amplicons were separated by electrophoresis on a 1.0% (w/v) agarose gel stained with ethidium bromide (0.5 μg/ml).

For primary screening, 52 wild potato accessions were grown in a professional growth mix (Berger BM6 All-Purpose mix, pH 5.4–6.2) in 2.84-l pots at 22°C under 14-h light/10-h dark photoperiod and 50% relative humidity in controlled growth chambers. No choice feeding was performed on 4-week-old plants by inoculating each plant with 20 CLso⁺ psyllids enclosed into two separate organza drawstring bags (10 psyllids/bag), which were tied to the second and third fully expanded leaves (Supplementary Figure 2B), with four replicates of inoculated plants per accession. The bags and psyllids were removed from plants 1 week after placement. Phenotypic evaluations of foliar ZC disease symptoms were done at 14, 21, and 28 days post-inoculation (dpis) in an environment-controlled growth chamber. Foliar symptoms were evaluated and rated based on a scale of 1–3, where 1 = being resistant (no symptoms), 2 = moderately susceptible (some symptoms), and 3 = highly susceptible (severe symptoms and morbidity) (Supplementary Figures 2C, 3). Chip frying and ZC scoring were performed as described in other studies (Henne et al., 2010; Harrison et al., 2019).

A second screening of the ZC resistant and moderately susceptible accessions chosen from the primary screening and a screening of the single seed lines from true potato seeds were done using 1-month-old plants. To increase quantity, all plants were vegetatively propagated by tissue culture and were planted in modified enclosed transparent 32 oz. plastic cups. No choice feeding was performed by releasing five CLso⁺ psyllids (presence of CLso was confirmed by PCR diagnostics) into each enclosed cup containing two plants in a professional growth mix Berger BM6 All-Purpose mix, pH 5.4–6.2, with two cups per accession. CLso⁻ psyllid (absence of CLso was confirmed by PCR diagnostics) challenged and non-challenged plants were used as negative controls (Supplementary Figure 2A). Phenotypic evaluations of inoculated plants consisted of assessing foliar symptoms of ZC at 14, 21, and 28 dpi. At 7 dpi, the soil was drenched with Admire Pro (Bayer Crop Science, Germany) to kill all psyllids in enclosed cups and prevent further plant damage. Leaf tissue was collected at 21 and 28 dpi, with 3–4 leaves from the top, middle, and bottom of each plant, cut into pieces with a single edge blade, and pooled to produce four replicates before freezing in liquid nitrogen (VWR Reinforced 2 ml Bead Mill Tubes, Radnor, PA, USA). Tissues were lyophilized before genomic DNA extraction for PCR and quantitative PCR (qPCR) analyses.

Genomic DNA was extracted from leaf tissue using a modified protocol as previously published (Edwards et al., 1991). For the detection of CLso in tissues collected from wild potato accessions and Atlantic control, conventional PCR was performed on a ProFlex™ PCR System (Applied Biosystems, Life Technologies, Carlsbad, CA) in a total reaction volume of 20 μl, using 150 ng of DNA, 0.5 μM of each target-specific primer, 10 μl of AccuStart Tough Mix (Quantabio, Beverly, MA, USA), 0.5 μl of 50 × loading dye, and 6.6 μl of nuclease-free water (Ambion, Life Technologies, Austin, TX, USA). For initial conventional (gel) PCR analysis, a primer pair, OI2C-F and OA2-R, was used to specifically amplify CLso 16S rDNA (Levy et al., 2011). The potato Ribosomal Protein L2(RPL2)-specific primers RPL2-F and RPL2-R were used to amplify an endogenous reference gene (Supplementary Table 2) (Avila et al., 2012). PCR conditions were given as follows: one denaturing cycle at 95°C for 30 s; 28 cycles each at 95°C for 30 s, 50°C for 30 s, and 68°C for 2 min; and a final extension cycle at 68°C for 5 min. PCR amplicons were separated by electrophoresis on a 1.0% (w/v) agarose gel stained with ethidium bromide (0.5 μg/ml).

The CLso relative titer was quantified in tissues collected from wild potato accessions and Atlantic control using quantitative PCR (qPCR) in a CFX384™ Real-Time System (Bio-Rad Laboratories, Inc., Hercules, CA, USA) in a total reaction of 10 μl, using 50 ng of DNA, 0.4 μM of each target-specific primer, and 5 μl of iTaq™ Universal SYBR Green Supermix (Bio-Rad Laboratories, Inc., Hercules, CA, USA). Three biological replicates were used with two technical replicates. Lso-F and HLB-R primers (Supplementary Table 2) were used to specifically amplify CLso 16S rDNA, while Sotu-RPL2-F and Sotu-RPL2-R primers (Supplementary Table 2) were used for the amplification of an endogenous reference gene, RPL2 (Levy et al., 2011; Irigoyen et al., 2020). PCR conditions were given as follows: one denaturing cycle at 95°C for 3 min; 40 two-step cycles each at 95°C for 15 s and at 55°C for 30 s; and a final melting curve of 65–95°C for 55 s. The results were analyzed and recorded as C_T (threshold cycle) values, which were normalized against the C_T values of the potato RPL2 (Supplementary Table 2) for quantification using the comparative C_T method (2T-ΔΔC) (Zuñiga et al., 2020). The Student's t-test was used to determine statistically significant (p ≤ 0.05 or 0.01) differences between the controls and treatments.

The olfactory host-preference assays of B. cockerelli adults (CLso⁻), reared and maintained on Atlantic potato plants, were conducted using 2-month-old Sb-PI310927 and Atlantic plants and a Y-tube olfactometer (35 cm long × 2.5 cm diameter). Twenty B. cockerelli adults per set of plants (n = 10 sets of plants, Sb-PI310927 and Atlantic plants) were collected into plastic vials for 16 h before initiating the behavioral assays to promote insect host choice. The Y-tube was connected to a 2-port Humidified Air Delivery system (ARS, Gainesville, FL, USA). The air was filtered through an activated carbon filter (16 cm, ARS). Airflow of each arm (bifurcated at a 45 angle) was set up at a constant rate of 10 L/min, leading into two separated sealed cylindrical glass chambers (15 cm diameter, 35 cm high) that contained individual plants, Sb-PI310927 or Atlantic, as odor sources. The Y-tube was placed at a 30° angle from the surface and adult psyllids were released at the end of the tube. To examine the insect olfactory behavior, observations were recorded by counting the number of psyllids at each arm's terminal end (set threshold = 4 cm) every 15 min for a 60-min period. All others were considered as a non-responding group. All measurements were conducted between 2 and 4 p.m. Central Standard Time (CST) at ambient temperature (21°C) and under constant light (~250 μmol/m² s).

The survival of B. cockerelli adults (CLso⁻) on 2-month-old Sb-PI310927 and Atlantic plants (n = 10 plants; 10 psyllids/plant) was evaluated to record the surviving insects every day for 7 days. To examine the oviposition of B. cockerelli on Sb-PI310927 and Atlantic plants, a non-choice assay was performed on insect-proof mesh cages (30 × 30 × 30 cm). Before the assay was initiated, couples (female and male psyllids) were allowed to mate for 4 h in a 1.7-ml tube and subsequently transferred onto plants (n = 10 plants; 10 psyllids/plant). After 7 days, adult psyllids were removed, and the total number of eggs was recorded.

One-month-old in vitro culture plants of ZC-resistant single seed line 10 of Sb-PI310927 accession and Atlantic control were hardened for 1 month in the growth mix. Three 2-mm diameter disks were excised from one leaflet of the second and third fully expanded leaves from two plants of each of Sb-PI310927-resistant line 10 and Atlantic control, using a 2-mm sampling tool (Electron Microscopy Sciences, Hatfield, PA, USA), and observed at 60 × on an Olympus BX51 stereomicroscope (Olympus Corporation, Tokyo, Japan). Leaflet disks were evaluated for leaf trichome shape and density based on two biological samplings and 12 technical replications. Trichome density was calculated as the number of trichomes per mm² leaf area on both the adaxial and abaxial leaflet surfaces.

The data analysis, statistics, and graphs were displayed as described in the figure legends using the Microsoft Excel software (version 2009). The Student's t-test was used to determine statistically significant (p ≤ 0.05, 0.01, or 0.001) differences in CLso titer and oviposition between control and treatment. The Chi-square (χ²) test was used to compare the number of potato psyllids responding to plant odor sources using a Y-tube olfactometer. The survival of potato psyllids was estimated using the Kaplan–Meier survival analysis using the RStudio environment (Allaire, 2011).

Primary screening of 52 wild potato accessions was performed from the Solanum sect. Petota collection (Supplementary Table 1) to identify potential lines with resistance and tolerance to CLso or the psyllid. Multiple in vitro propagated plants of each accession were hardened in pots for 2 months, followed by exposure to CLso⁺ psyllids. Foliar symptoms of ZC were then observed every 7 days for up to 4 weeks (Supplemental Figure 2A, B) and rated based on a scale of 1–3, where 1 = being resistant (no symptoms), 2 = moderately susceptible (some symptoms), and 3 = highly susceptible (severe symptoms and morbidity) (Supplementary Figure 2C). Several of the 52 accessions were susceptible and moderately susceptible, showing some upward leaf rolling, chlorosis, and plant stunting, but they grew new leaves at 21–28 dpi. Interestingly, one diploid accession Solanum berthaultii Hawkes (Sb-PI310927), sourced initially from Bolivia, seemed resistant with no visible ZC symptoms. The nine accessions (one resistant and eight moderately susceptible) were selected, and the challenges with CLso⁺ psyllids were repeated using 1-month-old plants grown in terrariums (enclosed cups). In addition to phenotypic observation for ZC symptoms, the relative CLso titers were estimated at 28 dpi when CLso could be sufficiently detected. The single resistant accession, Sb-PI310927, reproducibly showed no visible symptoms and continued to grow after 28 dpi with a severity index of 1, compared with the susceptible Atlantic control that displayed severe symptoms and died at 28 dpi (severity index 3). Three accessions, namely, Sk-PI498359, So-PI498130, and Sr-PI310953, again showed reproducible ZC symptoms with a severity index scale of 2 (Figure 1A). The remaining five accessions displayed variable symptoms and died at 28 dpi, with a severity index of 3. In this experiment, both conventional and quantitative PCR confirmed the presence or absence of CLso (Figures 1B,C). The resistant accession Sb-PI310927 and the three moderately susceptible accessions (i.e., Sk-PI498359, So-PI498130, and Sr-PI310953) displayed CLso titers of 6.05% and 52.44–119.36%, respectively, relative to that of the Atlantic control (set to 100%) (Figure 1C). Further comparative studies were carried on with the resistant (Sb-PI310927) and highly susceptible control (Atlantic) plants.

The ZC resistance trait of Sb-PI310927 was confirmed by repeating the ZC challenges, using 2-month-old plants under greenhouse conditions, and evaluating the ZC tuber symptoms. Typically, CLso-psyllid feeding for the brief period of the challenge (~7 days) causes the inoculated lower leaves to wilt, while the newly emerging (upper) non-inoculated leaves remain asymptomatic. In contrast, CLso+ psyllids induce symptoms associated with ZC on new upper non-inoculated leaves ranging from leaf wilting, upward curling, and chlorosis/necrosis. They can also lead to severe stunting of the plants (Supplementary Figure 4). Sb-PI310927 plants consistently showed no visual ZC foliar symptoms with a severity index of 1 when challenged with CLso⁺ or CLso⁻ psyllids. However, the susceptible Atlantic variety displayed typical ZC foliar symptoms with a severity index of 3 when infected with CLso⁺ psyllids at 28 dpi (Figure 2B). Freshly cut tubers and fried chips of CLso-infected Sb-PI310927 accession showed no brown discoloration, whereas Atlantic displayed characteristic ZC tuber symptoms (Figure 2C). Furthermore, molecular diagnostic analyses by qPCR and conventional PCR indicated that Sb-PI310927 had a very low relative CLso titer of 4.87% compared with Atlantic plants (CLso titer set to 100%) (Figure 2D,E).

All screening was conducted using in vitro micro-propagated plants from a true seed sourced from the USDA-ARS germplasm bank. As with most wild germplasm collections, Sb-PI310927 true seeds could be heterozygous and pooled from multiple plants during germplasm maintenance, thus having the potential for segregating alleles (Bamberg and Del Rio, 2020). To determine if there would be any segregation of the ZC-resistant trait of Sb-PI310927 among true seed progeny and to recover a stable resistant line for further studies, 10 true botanical seeds of Sb-PI310927 were germinated and multiplied individually by micropropagation. The ZC resistance trait of the 10 single seed lines was then evaluated following no-choice challenges with CLso⁺, CLso⁻, or no psyllids under growth chamber conditions (Figure 3A). At 21 dpi, the susceptible Atlantic control showed characteristic ZC symptoms with high titers of CLso transmission, whereas almost all Sb-PI310927 lines did not show any ZC symptoms. However, conventional PCR revealed that only ~50% of Sb-PI310927 single seed lines were negative for CLso (Figure 3B). Further confirmation by quantitative PCR showed that six Sb-PI310927 single seed lines (#3, 4, 5, 7, 8, and 10) were CLso⁻ or had undetectable levels of CLso. Three lines (#2, 6, and 9) showed low to moderate levels of CLso compared with Atlantic control (CLso titer set to 100%). We also noted some segregation in plant height among the 10 lines, which can be observed even among unchallenged (healthy) plants. Together, these results suggest some degree of segregating alleles among the true seed pools of these wild accessions possible for multiple traits (Bamberg and Del Rio, 2020).

The following two possible mechanisms explain the ZC resistance trait of Sb-PI310927: (i) resistance to the psyllid either by antixenosis or antibiosis effects and (ii) immune gene-mediated resistance to the bacteria or the insect. Although not mutually exclusive, we first explored if there were antixenosis or antibiosis effects. For this, we evaluated whether there was a host preference between Sb-PI310927 line#10 and Atlantic plants equidistantly placed in an olfactometer-based choice assay (n = 10 plants; 20 psyllids/plant). Of the total psyllids analyzed in the experiment (20 psyllids × 10 plants = 200 psyllids), the majority (76%) of them responded within 60 min after the assay was initiated, i.e., reached a 4-cm set threshold distance in the arms toward the plant odor sources (Figure 4A). However, no significant differences in host-selection were found between Sb-PI310927 and Atlantic odor sources at the sampling points tested (Figure 4B; 15 min: χ² = 0.053, df = 1, P = 0.82; 30 min: χ² = 0.644, df = 1, P = 0.42; 45 min: χ² = 0.65, df = 1, P = 0.42; and 60 min: χ² = 0.118, df = 1, P = 0.73).

As there were no differences in host preference, we next evaluated if there were any adverse effects on oviposition, hatching, and survival of psyllids using no-choice assays with 2-month-old Sb-PI310927 line#10 and Atlantic plants (n = 10 plants; 10 psyllids/plant). The results showed a significant reduction (F = 113.68, df = 1, P < 0.001) in the number of oviposited eggs after 7 days on Sb-PI310927 when compared with Atlantic plants (Figure 4C). The few oviposited eggs on Sb-PI310927 were also defective in hatching. Furthermore, survival analysis showed that adult psyllids on Sb-PI310927 had a significantly (Kaplan Meier analysis, P < 0.0001) lowered probability of survival beyond day 3 when compared with Atlantic plants (Figure 4D).

Lastly, in our experiments, we observed that the leaves of S. berthaultii were “sticky,” and the movement of psyllids appeared to be impaired (Figure 5C). Hence, under a microscope, we examined the adaxial and abaxial surfaces of 1-month-old Sb-PI310927 line #10. Sb-PI310927 leaves had a more significant number of trichomes, and they appear longer than those of Atlantic leaves (Figures 5A,B). The number of trichomes was significantly (P < 0.01) higher in Sb-PI310927 than in Atlantic leaves on both the adaxial and abaxial surfaces (Figures 5A,B). It is possible that the observed reduced fecundity and survival of psyllids effects on Sb-PI310927 could be associated with the increased trichome density and the presence of glandular trichomes (Figures 5C,D).