Field experiments were carried out in the summer of 2020 using the potato cultivar Kuras. A total of six bacterial isolates that displayed antagonistic activity against Phytophthora infestans under lab conditions were assessed for their plant growth promotion activity. These strains include Pseudomonas fluorescens SLU99, Serratia rubidaea EV23 and AV10, S. plymuthica S412 and AS13, and S. proteamaculans S4. A single colony of each bacterial strain was cultured in Luria-Bertani (LB) liquid for 16 h at 28°C. Then the cultures were centrifuged at 3000 × g for 10 min and the bacterial cells were resuspended in 5 L sterile water to a final OD₆₀₀ 0.2. The bacterial suspension was transferred to a 5 L pressure sprayer for field application, and water treatment was used as a control. The testing site was Mosslunda, near the city of Kristianstad, Sweden (55°58’00.3”N 14°07’03.0”E) where the average daily temperature ranged between 12 and 18°C and average monthly precipitation was 42-64 mm, with an average daylength was 17.5 hours. The experiment was carried out in a plot with four rows of 10 meters in width and 15 meters in length. The experimental design consisted of randomized block design with four blocks. Each block contained all eight treatments distributed randomly, with 10 plants per treatment. The potato plants were sprayed with six biocontrol bacteria (individual treatments) once every two weeks for a period of eight weeks. Statistical analysis of plant height was performed using JMPpro software.

To evaluate the effects of selected rhizobacteria on tomato and potato plant growth, a pot experiment was conducted for each crop in two separate plant cultivation chambers (biotron). Nine and ten biological replicates were used for tomato and potato, respectively. The tomato seeds were surface sterilized using 3% sodium hypochlorite (v/v) followed by washing with sterile water three times before sowing in a 105-hole seed tray filled with nursery soil. At 25 days after sowing, uniform four-leaf stage seedlings were transplanted into 1 L pots containing 325 g of nursery soil (1 seedling/pot). For bacterial inoculations, strains SLU99, EV23, AV10 and S412 were cultured as described in the previous section. Thirty-five milliliters of bacterial suspensions at OD₆₀₀ 0.2 prepared with 1 × phosphate-buffered saline (PBS) (Thermo Scientific, Waltham, MA, USA) or only 1 × PBS (control treatment) was added to the roots of 25-day-old tomato seedlings. For potato, all seed tubers were first cleaned with water, treated with 100 ppm gibberellic acid, and dried for eight days to promote bud initiation. Prior to planting in 2 L pots (containing 650 g nursery soil), the seed tubers were inoculated with bacteria suspension at OD₆₀₀ 0.2 prepared with 1 × PBS or with only 1 × PBS for 15 min.

The inoculation was repeated twice, once at 10 and 30 days after transplanting (DAT) for tomato or days after planting (DAP) for potato with 35 ml bacterial suspension (OD₆₀₀ 0.2) per plant in both crops to ensure the growth of bacteria in the soil. The liquid was poured 2-3 cm from each plant’s base and 5 cm deep in the soil. For the duration of the experiments, the plants were grown in two separate growth chambers in 16 h light/8 h dark photoperiods supplemented with 250 µmol m-2 s-1 light, temperatures of 20-23°C and 60% humidity for 60 days. The plant positions within the growth chambers were randomly rearranged at least twice a week.

To determine the effects of selected rhizobacteria on plant growth, parameters of plant height, total plant dry weight and yield were measured. Plant height was measured from the soil surface to the tip of the plant every ten days from planting until harvest. The yield of tomato and potato were measured by total fruit or tuber number and total fresh weight. Firstly, fruit or tubers were harvested, counted, and weighed. Then, the shoot and root systems were separated. Leaves were separated from the plant by cutting the leaf petiole, and the stems were cut. The number of leaves was counted and recorded. The intact roots were gently shaken to remove soil, briefly washed with water, and gently blotted dry using paper towels. Then, immediately the roots, leaves, stems, and fruits/tubers were separately placed in paper bags and oven-dried at 65°C until they reached constant weight. Finally, the dry weight was measured using a precision scale and recorded.

The chlorophyll content of tomato and potato plants grown in the growth chamber were measured at harvest using the MC-100 Chlorophyll Concentration Meter (Apogee Instruments Inc., Logan, UT, USA). The MC-100 was calibrated to measure chlorophyll concentration with units of chlorophyll content index (CCI) and zeroed before commencing measurement according to the manufacturer’s instructions (Apogee Instruments, 2018). For tomato plants, the measurements were made by clipping the sensor onto the second terminal leaflet on the fifth fully expanded leaf from the top of each plant (Matsuda et al., 2014). For potato plants, the measurements were made on the top point of the top leaflet of the 4th compound leaf (Li et al., 2012).

To evaluate the effects of rhizobacteria applications on soil nutrient content, rhizosphere soils were collected and pooled from all pots of each bacterial treatment for analysis. For the control treatment, soils were collected from eight randomly selected non-inoculated plants. 30g of dried soil samples were sent to the LMI AB testing laboratory (Helsingborg, Sweden) for analysis. The soils sampled before planting and at harvest were analyzed for pH, soil organic matter (SOM), electrical conductivity (EC), total nitrogen, available phosphorus (P), and available potassium (K).

Tomato seeds or potato explants were grown in half-strength Murashige and Skoog (MS) culture medium (M0221; Duchefa Biochemie B.V., Haarlem, The Netherlands), pH 5.8, supplemented with 0.25% (w/v) Phytagel (Sigma-Aldrich Co., St. Louis, MO, United States), 0.05% (w/v) 2-(N-morpholino)ethanesulfonic acid (MES) and 1% sucrose (w/v) (Duchefa Biochemie B.V., Haarlem, The Netherlands). For SLU99 treatment, bacteria were cultured overnight in LB until an OD₆₀₀ of 2 was reached. Cell-free supernatant was obtained by centrifuging at 3000 × g for 10 min, filtered through a 0.22-μm filter and mixed with ½ MS medium at 1:10 ratio before plating. For mock treatment, 0.22-μm sterile filtered liquid LB media was mixed with ½ MS media at 1:10 ratio before plating seeds or explants. Tomato var Moneymaker seeds were obtained from Plantagen AB and were surface sterilized with 3% (v/v) sodium hypochlorite for 5 min and washed three times with sterile water. Seeds were then plated either on control or SLU99 supernatant supplemented ½ MS plates and stratified for 48 h at 4°C. For potato plants, the explants from sterile potato plants were grown either on mock-treated or SLU99 supernatant supplemented ½ MS media in plant tissue culture boxes. The plants were grown in a growth chamber with 16 h light/8 h dark photoperiods and temperatures of 20-22°C.

For RNA-seq analysis, leaf or root samples were collected for each treatment and frozen in liquid nitrogen. Total RNA was extracted from three biological replicates using the RNeasy Mini Kit (Qiagen). Each biological replicate represented a pool of leaves or roots from three individual plants. The quality and integrity of the RNA was measured using an Agilent Bioanalyzer 2100 (Agilent, California, USA). Library preparation was carried out using a TruSeq RNA poly-A selection kit (Illumina, Inc.). Sequencing was performed at National Genomics Infrastructure (NGI), Stockholm on an Illumina NovaSeq6000 S4 platform. Adapter sequences and poor quality reads (<Q30) were removed using BBduk (Bushnell et al., 2017). Next, cleaned data were fed into STAR (Dobin et al., 2013) for alignment against the reference genome. FeatureCounts (Liao et al., 2014) was used for the quantification of aligned reads. Finally, R package DEseq2 (Love et al., 2014) was used for the analysis of differential gene expression and normalization. For volcano plots, R package ggplot2 (Ito and Murphy, 2013) was used. Pathway and enrichment analysis of differentially expressed genes was carried out with the R package clusterProfiler (Wu et al., 2021).

Total RNA (1 μg) extracted for the purpose of transcriptome analysis was used for reverse transcription using qScript cDNA synthesis kit (Quantabio, Beverly, MA) according to the manufacturer’s instructions. 4 μL of 10-fold diluted cDNA was used as a template for the real-time PCR. Quantitative PCR was performed using DyNAmo Flash SYBR green kit (Thermo Scientific, Waltham, MA) according to the manufacturer’s instructions with three biological replicates and three technical replicates for each biological replicate. Relative expression levels of each of the target genes were normalized to that of tomato tubulin (SlTub) gene for quantification using the ΔΔCT method (Livak and Schmittgen, 2001).

We investigated six bacterial strains with biological control activity, P. fluorescens SLU99, Serratia rubidaea EV23, AV10, Serratia plymuthica AS13, S412, and Serratia proteamaculans S4. To evaluate the growth promotion activity of these bacteria, potato plants grown in field conditions were subjected to treatment with these bacteria. Four out of the six tested strains, SLU99, EV23, AV10 and S412 resulted in significantly increased plant height, compared to the water treated control ( Figure 1 ).

To examine the effectiveness of plant growth promotion, potato and tomato plants grown under greenhouse conditions were inoculated with the four selected strains. Inoculation with P. fluorescens SLU99 significantly increased plant height, total dry weight, and yield of tomato plants by 13.3%, 34.5%, and 183.9%, respectively, compared to non-inoculated plants. Meanwhile, S. rubidaea AV10-treated tomato plants had 12.3% greater total dry weight than non-inoculated plants. On the other hand, inoculation with S. rubidaea EV23 and S. plymuthica S412 had no significant effect on all morphological parameters tested on tomato plants ( Table 1 ; Figures 2A–F ). Furthermore, inoculation with both SLU99 and EV23, individually, significantly increased potato plant height, yield, and total plant dry weight ( Table 1 ). Inoculation with AV10 promoted plant growth as total plant dry weight but not plant height, and vice versa for S412 on potato. Nevertheless, all inoculated plants had significantly higher tuber yields compared to the controls. The highest increase in yield for potato was recorded for EV23 (15%) and SLU99 (13.7%), followed by AV10 and S412 treated plants with 11.3%, and 10.7%, respectively ( Table 1 ).

Plant height increased rapidly during the vegetative stage until fruit set (40 DAT), or until the start of tuber initiation (30 DAP) for tomato and potato plants, respectively, and was not significantly different among treatments. Thereafter, during the fruit growth stage (50-60 DAT), a significant effect of bacteria inoculation on tomato plant height was observed (P=0.0431 and P=0.0131 at 50 and 60 DAT, respectively) ( Table 1 ; Supplementary Figure 1B ). In contrast, a 4.2% reduction in tomato plant height was observed when plants were treated with EV23 ( Table 1 ). Applications of SLU99 resulted in increased tomato leaf ( Figure 2E ) and fruit number ( Figure 2F ) compared to other treatments. Consequently, SLU99-treated tomato plants had 39.7% and 191.2% more leaf and fruit dry weights, respectively, compared to non-inoculated plants.

For potato plants at 30 DAP, swelling stolon tips were observed in all treatments suggesting tuber initiation had started. The effect of bacteria inoculation on potato plant height was only significant during the tuber bulking stage (40-60 DAP) ( Table 1 ; Supplementary Figure 1 ). In addition, all bacteria except S412 significantly increased tuber dry weight over control plants (P=0.0003) ( Figure 2J ). However, the dry weight of other plant parts was not enhanced by bacteria treatments ( Figures 2G–I, K, L ).

Relative chlorophyll content was significantly higher after SLU99 application in tomato (P=0.0188) and potato plants (P=0.0287) but was not elevated after EV23, AV10 and S412 treatments compared to the control ( Figure 3 ).

Additionally, at harvest, inoculation with SLU99 resulted in 3.4% higher total nitrogen (TN) in the tomato grown soil, while inoculation of EV23, AV10 and S412 resulted in 10.7%, 7.7% and 24.3% lower TN over control treatment, respectively ( Supplementary Table 1 ). The elements P and K were slightly increased in tomato grown soil after SLU99 treatment, whereas EC and SOM were decreased. In potato grown soil, all bacterial treatments increased available K over the control by 22.2%-66.7%. TN in potato grown soil decreased in a small range after SLU99, EV23 and AV10 treatments (2.1%-6.2%). Meanwhile, the available P, pH, EC, and SOM at harvest of soils grown with potato were not affected by bacteria treatments ( Supplementary Table 1 ). In summary, inoculation of SLU99 significantly promoted the growth of tomato and potato plants, and EV23 promoted the growth of potato plants grown in a controlled environment as reflected by high chlorophyll content, total plant biomass and yield.

To gain insights into the mechanisms behind growth promotion, tomato and potato plants were grown in MS media supplemented with the sterile-filtered bacterial culture supernatant. When compared to the mock treatment, tomato seeds grown in the SLU99-supernatant supplemented media showed a higher germination rate, increased root growth and shoot height, and enhanced number of lateral roots ( Figures 4A, B ). Potato explants grown in the media supplemented with SLU99 supernatant resulted in increased shoot height by 24.6% (p = 0.02) compared to the control ( Figures 4C, D ). SLU99 treatment also increased root growth and resulted in the formation of adventitious roots and secondary adventitious roots (SAR), which emerge from aerial parts ( Figure 4D ).

To gain insights into the mechanisms of SLU99-mediated growth promotion in tomato and potato plants, RNA extracted from the roots and leaves grown in the media supplemented with the supernatant of SLU99 was subjected to transcriptome profiling. A total of 1193 and 2226 genes were differentially expressed (log2FC >1.5, p < 0.05) in the roots of tomato and potato, respectively, compared to the control samples ( Table 2 ; q-value or padj value information can be found in Supplementary Excel files 1–4 ). Among them, 1076 and 996 genes were upregulated, and 117 and 1230 genes were downregulated in the respective samples ( Table 2 ; Figures 5A, B ). In tomato leaves, a total of 1732 DEGs were detected (log2FC >1.5, p < 0.05) with 1322 genes upregulated and 410 genes downregulated ( Figure 5C ). For potato leaves, we identified 959 (log2FC >1.5, p < 0.05) when treated with SLU99 compared to the control samples. Of the identified DEGs, 206 were upregulated, and 753 genes were downregulated ( Table 2 ; Figure 5D ).

We performed KEGG pathway enrichment analysis to gain a better understanding of the functional categories of the DEGs in response to the PGPR culture supernatant treatment. In the treated root samples, a total of 300 and 753 genes were grouped into enriched pathways in tomato and potato, respectively, and a total of 382 and 259 genes were grouped into enriched pathways in the treated leaves of tomato and potato, respectively. Treatment resulted in enrichment for genes involved in plant hormone signal transduction pathway (37 and 33 genes, respectively), zeatin biosynthesis (15 and 13 genes), plant-pathogen interactions (37 and 25 genes) and MAPK signaling pathway (30 and 19 genes) in both tomato and potato leaves. Additionally, a total of 40, 76, 44 and 25 genes related to the phenylpropanoid biosynthesis pathway were enriched in the roots and leaves of tomato and potato upon treatment with SLU99 culture supernatant ( Figure 6 ).

To compare the effects of the tomato and potato samples, we examined the overlap of the DEG products among the roots and leaves. The analysis showed an overlap of 164 gene products in the root samples, and 95 in the leaf samples upon SLU99 treatment in potato and tomato plants ( Figures 7A, B ). Within the species, there was an overlap of 515 gene products between root and leaf samples of tomato, while 265 similar gene products were responsive in the root and leaf samples of potato ( Figures 7C, D ).

One of the major objectives of this study was to elucidate the changes in the host transcriptomes that are important for PGP activity. Phytohormones, as a result of their complex interaction and crosstalk, regulate various cellular processes involved in plant growth and development. The gibberellins (GAs) play an important role in several developmental processes including seed germination. Upon SLU99 treatment in tomato root, gibberellin 3-beta-dioxygenase 1-like (GA3OX1), a gene involved in the biosynthesis of GA was upregulated (log2FC 1.9, p<0.05). On the other hand, in tomato leaves, gibberellin 2-beta-dioxygenase 8 (GA2ox8), a gene involved in the deactivation of GA, was upregulated (log2FC 1.6, p<0.01) ( Figure 8 ).

The hormones auxin and CK are major players in the regulation of signaling pathways underlying plant growth and development. In tomato, tryptophan synthase beta subunit 1 (TSB1)-like, responsible for increased in tryptophan biosynthesis, and YUCCA8, responsible for the last enzymatic step from indole-3-pyruvic acid (IPA) to IAA were upregulated 16 and 3-fold (log2FC 4 and 1.6) in leaf and root samples, respectively. Additionally, in potato roots, SLU99 treatment resulted in increased expression (log2FC 3.3) of L-tryptophan-pyruvate aminotransferase 1 (TAA1)-like, which encodes an enzyme that converts L-tryptophan to IPA ( Figure 8 ).

The activation of auxin biosynthesis genes and production of IAA by SLU99 (Kelbessa et al., 2022) prompted us to analyze the differences in the expression of genes involved in auxin signaling pathways. Major auxin-responsive genes such as Gretchen Hagen 3 (GH3) and small auxin up RNA (SAUR) are transcriptionally regulated by auxin at early stages of signal transduction. Following treatment with SLU99 culture supernatant, expression of the IAA-amido synthetase GH3.1 (GH3.1), important for maintaining auxin homeostasis increased significantly (log2FC 6.8, 4.6 and 1.6 in tomato root, tomato leaf and potato root, respectively) in comparison to the mock treatment. Within the SAUR family, five and seven genes were differentially expressed in the tomato and potato plants, respectively. Additionally, PIN2, a root-specific auxin transporter and PIN8, a constitutively active auxin transporter were also upregulated in the potato plants ( Figure 8 ).

The gene encoding CK hydroxylase, an enzyme that catalyzes the biosynthesis of trans-zeatin (tZ), a biologically active CK, was strongly upregulated (log2FC 7.5) upon treatment with SLU99 in tomato roots. On the other hand, in tomato leaf CK dehydroxygenase 3 (CKX3), involved in the degradation of CK was upregulated 4-fold (log2FC 2) compared to the mock treatment. Furthermore, zeatin O-glycosyltransferase (ZOG1) and zeatin O-xylosyltransferase (ZOX1) were upregulated in tomato leaves (log2FC 5.4 and 3.5), while only ZOG1 was upregulated in the tomato and potato roots (log2FC 2.1 and 1.6). ZOG1 and ZOX1 are involved in the conversion of tZ to a stable and reversible O-glycosylzeatin and O-xylosylzeatin, respectively.

The PGP activity of bacteria is also attributed to the changes in the plant hormone ethylene (Poupin et al., 2016). Since SLU99 displayed ACC deaminase activity, we examined changes in the expression of genes encoding the enzymes of the ethylene biosynthesis pathway. 1-aminocyclopropane-1-carboxylate_synthase-like (ACS) involved in the synthesis of ACC, a direct precursor of ethylene, and 1-aminocyclopropane-1-carboxylate_oxidase (ACO) subsequently oxidizes ACC to ethylene. In tomato, ACS and ACO1 were upregulated in roots (Log2FC 5.1 and 1.8, respectively), while ACO1 and ACO4 were upregulated in leaves (Log2FC 3.2 and 2.6, respectively). In potato, ACS2 was upregulated in roots (log2FC 4.1), while ACS4 was upregulated in leaves (log2FC 2.2). ACO homologs were upregulated both in the roots and leaves (log2FC 2.9 and 1.7, respectively), while ACO1, ACO5 and ACO11 were upregulated in leaves (log2FC 2.1, 2.1 and 3.3, respectively). Lysine histidine transporters (LHT) are associated with the transportation of ACC. When treated with SLU99 supernatant, LHT8 was upregulated in both tomato and potato roots and tomato leaves by 3.6, 2.5 and 2.5-fold (log2FC), respectively, while LHT1 was upregulated only in tomato leaves (log2FC 1.6). To validate the observed gene expression differences in the NGS data we employed RT-qPCR. We analyzed three key genes involved in the ethylene pathway, SlACS, SlACO1, and SlLHT8 that showed increased gene expression following SLU99 treatment ( Supplementary Figure 2 ), in agreement with the findings from RNA-seq. Taken together, these results indicate that treatment with the culture supernatant from SLU99 stimulates the accumulation of phytohormones auxin, CK and ethylene.

The discovery of several DEGs involved in signal transduction pathways in response to SLU99 treatment suggests strong regulation of host transcription factor networks. To investigate the interaction of SLU99 with the plant transcriptional network, DEGs encoding TFs were identified and assigned to families using the plant transcription factor database. In total, DEGs belonging to 24 different TF families were identified with the greatest number of DEGs identified in the root samples compared to the leaf samples ( Supplementary Table 2 ). In the tomato and potato roots, a total of 112 and 182 TF encoding DEGs were found, whereas 62 to 92 DEGs were found in leaf samples, implying significant reprogramming of the host transcriptional network. The HD-ZIP TF genes were exclusively upregulated in tomato plants (11 upregulated and 2 downregulated), while 14 HD-ZIPs were downregulated in potato. Notably, the strongest upregulation was found for ATHB12 (log2FC 9.6) followed by ATHB40-like (log2FC 6.4), and ATHB7-like (log2FC 3.4) in tomato leaves. ATHB12 and ATHB7, considered paralogs, belong to the class-I HD-ZIP TF family and were also upregulated in root samples (log2FC 2.1 and 2.0, respectively). ATHB40, on the other hand, is a class-II HD-ZIP TF that was upregulated in the root (log2FC 3.7). ATHB12 and ATHB7 are associated with root elongation and leaf development, albeit at different stages (Ré et al., 2014; Hur et al., 2015). ATHB40 is a negative regulator of primary root development (Mora et al., 2022).

In addition to the PGP-related TF encoding DEGs, we also identified several DEGs encoding TFs involved in defence-related pathways, namely, WRKY, MYB, MYC, HSF, and NAC TFs. WRKY TFs are global regulators of host responses to phytopathogens. Treatment with SLU99 culture supernatant triggered differential expression of 14 WRKY genes in tomato and 11 genes in potato. Two of the WRKY TF encoding genes, WRKY30 and WRKY45, involved in defence against biotic and abiotic stresses, were upregulated in common between tomato and potato. WRKY6, WRKY55, and WRKY71 were uniquely upregulated in tomato leaf and, along with WRKY33B and WRKY75, are involved in defence against necrotrophic fungal pathogens. WRKY40 and WRKY41, genes involved in abiotic stress tolerance, were strongly upregulated in tomato leaves (log2FC 9.0 and 4.5, respectively), and were downregulated in potato leaves upon SLU99 treatment (log2FC -2.1 to -2.7). The MYC2 TF upregulated (log2FC 5.1) in tomato roots upon SLU99 treatment indicates a role in JA-mediated induced systemic resistance (ISR). Furthermore, treatment with SLU99 induced expression of several R-gene orthologs such as Cf-2,2-like (log2FC 2.2), Cf-9 (log2FC 3.4), RPM1 (log2FC 1.7), RPS5 (log2FC 4.1), CSA1 (log2FC 4.8) and Alternaria stem canker resistance (log2FC 1.6) in tomato leaves, LEAF RUST 10 DISEASE-RESISTANCE LOCUS RECEPTOR-LIKE PROTEIN KINASE-like 1.1 (LRK10) (log2FC 2.9 and 5.3 in root and leaf, respectively), TMV resistance protein N gene (log2FC 6.0) in tomato root and putative late blight resistance protein homolog R1A-10 (log2FC 1.6) and R1A-3 (log2FC 1.5) in potato leaves.