A gnotobiotic growth assay for Arabidopsis root microbiota reconstitution under iron limitation

Summary We present a gnotobiotic system for microbiota reconstitution on Arabidopsis thaliana under contrasting iron availability. This system induces iron starvation in plants by providing an unavailable form, mimicking conditions in alkaline soils. Inoculation of taxonomically diverse bacteria reconstitutes plants with a synthetic microbiota, allowing observation of nutrient-dependent interactions with commensals. Experimental optimization, including media composition and preparation of seedlings and bacteria, is discussed. This system provides a framework that can be adapted to study plant-microbiota interactions in further nutritional contexts. For complete details on the use and execution of this protocol, please refer to Harbort et al. (2020).


SUMMARY
We present a gnotobiotic system for microbiota reconstitution on Arabidopsis thaliana under contrasting iron availability. This system induces iron starvation in plants by providing an unavailable form, mimicking conditions in alkaline soils. Inoculation of taxonomically diverse bacteria reconstitutes plants with a synthetic microbiota, allowing observation of nutrient-dependent interactions with commensals. Experimental optimization, including media composition and preparation of seedlings and bacteria, is discussed. This system provides a framework that can be adapted to study plant-microbiota interactions in further nutritional contexts. For complete details on the use and execution of this protocol, please refer to Harbort et al. (2020).

BEFORE YOU BEGIN
In this protocol, we will assess Arabidopsis thaliana accession Col-0 performance when grown axenically or in the presence of a synthetic community (SynCom) of bacteria on two media with contrasting iron availability: a replete, ''available iron'' condition and an iron-limiting ''unavailable iron'' condition. In this system, the concentration of iron is identical in both media (100 mM); the availability is varied by providing forms of iron that differ in their solubility at high pH. By buffering the media strongly at pH 7.4 with 10 mM HEPES, we mimic iron-limiting calcareous soils, where iron is present but unavailable to plants (Kobayashi and Nishizawa, 2012). In the ''unavailable iron'' condition, iron(III) (supplied as FeCl 3 ) is sparingly soluble and is unavailable to plants. This is critical for assessing nutritionally beneficial host-microbiota interactions, which are precluded in artificial systems where iron is removed altogether (Harbort et al., 2020). As a control ''available iron'' medium with the same iron content and pH, the media is supplied with the same concentration of a chelated form of iron (FeEDTA, Ferric sodium ethylenediaminetetraacetate), which has increased solubility even in these alkaline conditions. To assess the impact of bacterial commensals on host performance in these conditions, the media are inoculated with a SynCom of rhizobacterial strains originally isolated from Arabidopsis roots grown in soil (Bai et al., 2015). Rhizobacterial culture collections of different origins can also be employed and compared. The complexity and diversity of the SynCom can be adjusted to suit the experimental question, or reduced to single strains in mono-associations.
In this protocol, surface-sterilized seedlings are pre-germinated for 5-6 days on axenic growth medium containing vitamins and sucrose, and then transferred to experimental conditions with controlled iron availability and commensals. Pre-germination and seedling transfer improve seedling germination rate and synchrony, which can vary between seed batches and genotypes, reducing variation within each condition.

Prepare stock solutions
Timing: $2 h a. Weigh out 119.1 g HEPES powder (4-(2-hydroxyethyl)piperazine-1-ethanesulfonic acid, N-(2-Hydroxyethyl)piperazine-N 0 -(2-ethanesulfonic acid)) and $18 g KOH tablets. b. Slowly add HEPES to 100 mL ddH 2 O, intermittently adding KOH tablets. Wait until the last addition is mostly dissolved before adding more to prevent clumping. The KOH tablets release heat during dissolution, which helps the HEPES dissolve while bringing the solution to a basic pH. . c. When everything has dissolved, adjust the volume to near 250 mL with ddH 2 O and bring the pH to 8.1-8.2 using a well-calibrated pH electrode. Adjust pH as needed with high concentration HCl or KOH. It is important that the HEPES is completely dissolved before proceeding, otherwise the pH will continue to change as it dissolves. d. Adjust final volume to 250 mL, sterile filter, and store at 4 C protected from light. 3. Prepare 10 mM MgCl 2 by dissolving 2 g of MgCl 2 $6H 2 O in 1 L of ddH 2 O and autoclaving. Prepare seedling germination plates Timing: $5 h 7. Prepare agar media for seed germination (1/2 MS with vitamins, 0.5% sucrose, 1 g/L MES pH 5.7). a. Per 1 L of ddH 2 O, dissolve 2.2 g of powdered MS medium containing vitamins, 5 g of sucrose, and 1 g MES. Once dissolved, adjust pH to 5.7. b. Add 10 g of low-impurity bacteriological agar and autoclave media. Allow autoclaved media to cool to a temperature that can be handled without solidifying ($45 C). c. In a laminar flow hood, pour media into 120 mm 3 120 mm square petri dishes by measuring $45 mL per plate in a 50-mL conical tube, then pouring into plates. Allow plates to dry with lids open for $30 min. d. Plates can be prepared in bulk and stored at 4 C for months.
Note: For seed germination plates, we use a traditional 1/2 MS medium from powder complete with vitamins and sucrose to increase the germination rate and obtain synchronously grown seedlings. Media for experimental conditions are prepared using individual stock solutions to control iron availability.

Prepare axenic seedlings
Timing: 7 days 8. Surface-sterilize A. thaliana seeds a. Measure out the desired number of seeds and transfer to a 2-mL microcentrifuge tube. b. Add 1.5 mL of 70% ethanol and incubate for 15 min with rotation or agitation. c. Working under sterile conditions, allow seeds to settle to the bottom of the tube and remove as much supernatant as possible. d. Wash seeds twice more with 70% ethanol, then once with 96% (v/v) ethanol. e. Remove as much ethanol as possible without aspirating seeds. f. Wash 5 times with sterile ddH 2 O to remove all traces of ethanol. Seeds can be pelleted by briefly centrifuging to facilitate water removal without removing seeds. g. Resuspend seeds in 1 mL ddH 2 O. 9. Transfer seeds onto germination plates ( Figure 1).

OPEN ACCESS
STAR Protocols 1, 100226, December 18, 2020 a. In a laminar flow hood, transfer seeds into rows on to plates containing germination agar media. Using a 20-mL pipette with a filter tip, aspirate seeds with some water, then deposit individual seeds in rows ($25 seeds per row, 5-6 rows per plate). Having a small space between individuals makes transferring to experimental plates quicker and easier. Rows should be 2.5-3 cm apart to allow for root growth. b. Leave plates horizontal with lids open for a few minutes to allow residual water to dry or absorb into the agar matrix. c. Close plates, seal with micropore tape, and store for 48 h at 4 C in the dark to stratify seeds. d. Transfer plates to a light chamber standing vertically with the program: 10 h light, 60-70 PPFD (mmol m À2 s À1 ) at 21 C/14 h dark, 19 C. These plant growth conditions are used for the entire protocol. e. Allow seeds to germinate and grow for 5-6 days. Troubleshooting 1

Grow bacterial cultures
Timing: 1-7 days 10. Inoculate bacterial cultures a. Timing for this step will depend on the bacteria to be used and the growth conditions. Different taxa of bacteria grow at different rates. It also makes a difference if bacteria are cultured in 96-well format deep-well plates (for large number of strains, but grow more slowly), or test tubes (for fewer strains, but allows for faster growth). Allow bacteria to grow to late exponential growth phase. b. For microbiota reconstitution with a taxonomically diverse synthetic community (SynCom), inoculate cultures in deep-well plates in appropriate nutrient medium (e.g., 1/2 TSB medium) and seal with micropore film to allow aeration. Culturing in 96-well format enables faster experiment setup when using many strains. c. Grow SynCom strains for 5 days at 25 C with shaking, until even slow-growing strains have significant growth. d. One day before experiment setup, resuspend strains by pipetting, dilute 1:100 into fresh medium, and grow 15-18 h at 25 C. Keep the original cultures growing as well. This ensures that both fast-and slowly growing strains are present in exponential growth phase. e. Alternatively, if growing fewer strains in test tubes, cultures can be subcultured 15-18 h, or for a few hours before experiment setup.

Prepare experimental media
Timing: $3 h

STEP-BY-STEP METHOD DETAILS Prepare SynCom for inoculation
Timing: 1-2 h Wash bacteria to remove antibiotics and other metabolites potentially in the culture media, and then combine into a SynCom for inoculation into MS agar media.
1. Wash and resuspend bacterial cultures. a. Centrifuge bacteria at 4,000 3 g for 15 min. b. Carefully remove supernatant without disturbing bacterial pellet using a pipette. c. Resuspend pellets in an equal volume of 10 mM MgCl 2 . d. For SynCom experiments: combine strains in a conical tube, centrifuge again at 4,000 3 g and resuspend well in 10 mM MgCl 2 . 2. Resuspend bacterial cultures to desired OD 600 .
a. Measure OD 600 of bacterial suspension using a photometer. Be sure to measure a dilution of the suspension within the linear range of your photometer. b. Dilute SynCom mixture to OD 600 = 0.1 in 10 mM MgCl 2 . This stock solution will be used to inoculate growth media. Troubleshooting 3 3. For a heat-killed negative control, incubate an aliquot of bacterial inoculant in a heat block at 99 C for 20 min.
CRITICAL: For complex SynComs with many strains growing in 96-well plates, both the original culture plate and the subcultured plate are combined to capture both slow and fast-growing strains in growth phase. Ensure that all bacteria are pelleted during the centrifugation step. Depending on the bacteria being used, you may need to adjust the centrifugation time. For complex SynComs it is generally not feasible to adjust the OD 600 of each strain individually to control the ratios of strains in the input. When working with fewer strains, as in mono-associations, it may be desirable to adjust the OD 600 of each strain before combining. Bacteria are resuspended at OD 600 =0.1, to be used as 1,0003 dilution in media (Final OD 600 =0.0001, $10 4 -10 5 CFU/mL). This may need to be adjusted for individual bacteria or experimental purposes.  1,0003). c. Continuing to stir, add necessary volume of HEPES buffer (stock is 2003). d. Allow media to cool in water bath to 42 C-45 C. Troubleshooting 3 5. Pour experimental plates a. For each plate, 45 mL of media will be poured into a square 120 3 120 mm petri dish. b. Add 50 mL of bacterial solution, or heat-killed SynCom negative control, to the bottom of a 50-mL conical tube (for a 1:1,000 dilution). Pour 45 mL of prepared and cooled media into the tube gently to prevent bubbles. Close lid and invert several times to mix. c. Pour media into petri dish and allow to dry with lid open for 20 min.

Pour experimental media plates
Note: Once HEPES buffer is added, media should be intermittently stirred on a magnetic stir plate to ensure equal distribution of iron without precipitation.
Pause Point: Experiment can be paused after drying plates for up to several hours before continuing to seedling transfer. After drying, cover plates with lids and leave until ready to continue.

Transfer plant seedlings
Timing: 1-3 h Axenic seedlings are transferred from germination plates to experimental media plates (Figure 2). 6. Visually check germination plates for signs of contamination, which should be apparent by day 5. 7. To transfer, gently remove seedlings from the plate by lifting under cotyledons with a sterile pipette tip or forceps. Avoid damaging the seedlings by overhandling or pinching. 8. Place seedlings (6-8 per plate) in a row near the top of the experimental plates, leaving enough room for shoot growth. 9. Seal plates with micropore tape and place vertically in a light chamber with the program: 10 h light, 60-70 PPFD (mmol m À2 s À1 ) at 21 C/14 h dark, 19 C. 10. Grow for 2 weeks, rotating the position of the plates within the growth chamber every 2-3 days to prevent location effects.
Note: Discard tips used for seedling transfer frequently to avoid contaminations. Always use a new tip when transferring to sterile control plates. Alternatively, if using forceps, sterilize between plates with 70% ethanol.

Evaluate plant performance
Timing: 3-5 h After 2 weeks of growth, evaluate plant performance by shoot fresh weight and chlorophyll measurement.
11. Using a scalpel, cut through micropore tape seal and open plates. 12. To measure shoot fresh weight (SFW), cut seedlings at the shoot-root junction, being careful not to cut off any leaves. Remove plant shoot with forceps and blot on paper towels to remove from the leaves any water condensation. Record weight of individual plant shoots using a precision scale. Dead plants are recorded as 0.

OPEN ACCESS
13. Measure chlorophyll content of a pooled leaf sample from each plate. a. After measuring the fresh weight of all plants from a single plate, collect a single leaf from each plant into a 2-mL microcentrifuge tube. Measure tube weight before and after adding sample to record input weight. This should be 20-30 mg of tissue. b. Freeze chlorophyll samples in liquid nitrogen and store at À80 C until further analysis. c. Extract chlorophyll by adding 1 mL DMSO to sample tubes and incubating with shaking at 65 C for 45-60 min, until leaf samples are clear and extracted. If the input sample weight was R30 mg, increase the amount of DMSO added to compensate. d. Measure the absorbance at 652 nm of the DMSO extract using a spectrophotometer. e. Calculate the chlorophyll concentration in the extract with the formula (Hiscox and Israelstam, 1979): -Chlorophyll mg/L = Abs 652 * (1000/34.5) f. Multiply the chlorophyll concentration in the extract by the volume of DMSO added (in mL), and divide by the input sample weight (in g) to obtain the plant chlorophyll content (mg/g of fresh weight).

EXPECTED OUTCOMES
After 2 weeks on experimental plates, there should be clear differences in axenic plant growth between available and unavailable iron conditions (Figure 3). Plants grown on unavailable iron conditions should be smaller, with shorter roots, and noticeably chlorotic (yellow) leaves compared to plants on available iron. Bacteria-inoculated plants will have variable phenotypes, depending on the bacteria and plant genotype. Often a small proportion of plants will not survive the experiment, especially in iron-limiting media. These plants should be included in the results and recorded as zero values.

LIMITATIONS
This assay provides a model for studying the impact of bacterial commensals and iron deficiency on plant performance. Especially relevant is the mechanism of iron limitation, which mimics conditions in iron-limiting alkaline soils where iron is present but unavailable to plants. However, as with all experimental models, some limitations should be considered during the interpretation of results.
In this protocol, plants are grown for 5-6 days axenically before being transferred to media containing microbes. This does not reflect conditions in nature, where colonization of commensals would begin immediately upon germination. The importance of interactions with commensals during this early phase of root microbiota establishment may be significant. Similarly, plants are exposed to iron-limiting growth media after pre-germination on media in which iron is readily available instead of immediately upon germination. Iron-limiting conditions may also impact the process of seed germination significantly, which is overlooked in this assay. Furthermore, as plants are grown in petri dishes on an agar matrix, the full life cycle and reproductive success of the plant cannot be assessed due to space limitations and high humidity. Results in this reductionist system should be cautiously extrapolated to conditions in soils, as it does not reproduce the full chemical and biological diversity of soil or soil physical parameters. When coupled with experiments in nutritionally characterized soils for phenotype confirmation, this assay provides a robust and amenable method to studying plant-microbiota interactions in the context of iron deficiency.

Potential solution
The pH-induced iron limitation may be too severe. The buffer concentration or final pH of the media may be too high for plants to overcome. Measure the final pH of the agar media by cutting out a section of agar, shaking in twice the volume of ddH 2 O for 30 min, and measuring the pH. Ensure that the pH is not above 7.4. Adjust the pH of the HEPES stock solution if necessary, and again check the pH of final media. Alternatively, slightly reduce the amount of HEPES stock added if your stock is too concentrated or alkaline.
Problem 3: large variation in plant growth Large variation in plant growth within plates or among experimental groups

Potential solution
This could arise from multiple potential sources.
Seedling transfer Be sure to transfer only germinated and growing seedlings to experimental plates. It is normal that a small portion of seeds do not germinate or seedling growth ceases after germination. These should be discarded and not transferred to experimental plates. Be careful not to injure seedlings during transfer, as this will impact plant growth. If injury of a seedling is suspected during transfer, discard it.

Insufficient bacterial inoculation
Ensure that the media is cooled to below 45 C before adding bacteria. If not properly cooled, the heat may kill the bacteria. Check the final OD 600 of your bacterial inoculant is 0.1. If this problem persists, the dose of inoculum can be increased.

RESOURCE AVAILABILITY
Lead contact Further information and requests for resources and reagents should be directed to and will be fulfilled by the Lead Contact, Paul Schulze-Lefert (schlef@mpipz.mpg.de).

Materials availability
This study did not generate new unique reagents. All bacterial strains and A. thaliana lines used in this study were previously described and are publicly available.