Lettuce root bacterial culture collection

To isolate a collection of culturable root-associated bacteria, 3 genotypes of L. sativa (CGN, Wageningen, Netherlands) were grown in greenhouse conditions at LandLab S.r.l. in Quinto Vicentino, Vicenza (Italy). The genotypes TKI-24, TKI-38 and TKI-100, were sown in seed trays and  germinated in a growth chamber for approximately two weeks. During this initial growth phase, the plants were fertigated with a nutritional solution containing NPK 18/18/18 to support their development. To simulate phosphate starvation, plants transferred in sand as a medium and supplemented with triple superphosphate (TSP) at a rate of 32.61 kg/ha (or 0.33 g per plant), equivalent to 15 kg/ha of P₂O₅. The procedure involved filling each pot (15 cm x 15 cm x 20 cm) with 3 L of non-sterilized soil-sand mixture, adding the pre-weighed TSP, and then topping it up with an additional 1 L of soil-sand to complete the pot-filling process. A mesh pad (30 g/m²) with holes was placed at the base of each pot to prevent water spillage. 

After the transplantation, plants were treated with 50 mg of microbial inoculum, applied as a 50 ml aqueous solution directly to the plant rootlet, at a concentration comparable to that used under field conditions. The synthetic microbial community used was formulated by Sacco System, located in Cadorago, Como (Italy). Further information available in Capparotto et al., 2025, bioRxiv1. To support plant growth, periodic fertigation was implemented at a concentration of N-K2O 250, equivalent to 76.92 kg/h.

The collection was established according to Zhang et al., 20212. The plants were harvested 3 weeks after the transplanting in the pot and one whole root from each genotype was taken for a total of 750 mg. The roots were washed in phosphate buffered saline solution, cut, mixed and aliquots of 23 mg each were made. 

From the pre-test result, dilution 1:54,000 was chosen as Optimal Dilution Concentration (ODC), corresponding to 19 µL of root suspension in 1 L of Tryptic Soy Broth (TSB) at 10% concentration (Supplementary Figure 1). 160 µL of the medium were dispensed in sixty 96-well plates (Greiner Bio-One, ref: 650101) with the VIAFLO96 (Integra). The same procedure was repeated for the 1/10, 1/3 , 3x and 10x of the Optimal Dilution Concentration. All the plates were placed in the dark at 25 °C for 14 days to let the bacteria grow. The wells with detected growth were counted (Supplementary Figure 2). The plates from 1/10, ⅓ and ODC were selected and 130 µL of bacterial suspension were transferred, resulting in twenty 96-well plates (1’920 samples, controls included). From each well, 10 µL were transferred to 96-PCR plate for bacterial identification. In the remaining culture, 120 µL of glycerol 80% (v/v) were added in each well, to allow storage at - 80 °C.

The library for the bacterial identification was performed following Zhang et al., 2021. The pipeline used to analyze the raw data is based on the ‘Culturome’ pipeline, with some important modifications: a filter step was added after the primer trimming to remove reads with polyN (>10 bp); to identify amplicon sequence variants (ASVs) was used DADA2 (mamba-0.18.2); to reduce redundant ASVs with the same biological meaning, a 99% similarity threshold was used to cluster the ASVs in meta-ASVs before the taxonomic assignation, which was performed with the SILVA classifier (138-99-nb).

On R (version 4.5.1), for each ASVs, the maximum relative abundance among all samples was considered. The ASVs with the maximum relative abundance less than 1% were discarded. The collection is made by the remaining 105 ASVs described in Collection_info.csv

1 Capparotto et al., 2025, bioRxiv doi: https://doi.org/10.1101/2025.04.02.646582

2 Zhang, J., Liu, Y. X., Guo, X., Qin, Y., Garrido-Oter, R., Schulze-Lefert, P., & Bai, Y. (2021). High-throughput cultivation and identification of bacteria from the plant root microbiota. Nature Protocols, 16(2), 988-1012.