High-throughput cultivation and identification of soil bacteria

Media: This is a big topic and determines the outcome of what is being isolated. General laboratory broth media contain high amounts of nutrients and isn't necessarily reflective of the soil environment. It is recommended that a dilute medium be used for isolation to mimic slightly more natural conditions. We typically use a 1/10 dilution of tryptic soy broth for general isolation, and more specialized medium like soil infusion or root exudates for more specific microbes.

Calculations: Cultivation can occur in any number of multi-well microtiter plates. We generally use 96-well round bottom plates for ease of growth observation. Depending on the volume of each well, calculate the total amount of media for each plate. For a 96-well plate, we add 150µL of media to each well, so you will need 14.4mL of media total. We use 20mL per plate to be safe.

Dilutions : This protocol uses a dilution to extinction method to try and isolate single cells into each well of a microtiter plate. Since each sample can contain drastically different amounts of living cells, each experiment needs to test the appropriate dilution to achieve 1/3 to 1/4 of the plate being colonized. This is to minimize the chance of multiple organisms growing within any single well.

Collect the soil (or sample) of interest. It is best to do the cultivation immediately after collection, but materials can be stored in the refrigerator overnight. Longer storage can affect the outcome if you are using this method for quantification.

Prepare the collected soil. Any materials in the soil that is not the target of interest (e.g. roots, detritus) should be removed.

Homogenize the sample, including breaking of aggregates into fine particles.

Weigh out 1g of the homogenized sample.

Add the homogenized sample into 5mL of sterile medium.

Vortex for 0h 1m 0s to suspend the soil particles in the medium.

Allow the heavier particles to sink for 0h 0m 15s.

Pipet 1mL of the suspension into 1L of medium. Do not include any large particles. Shake well to mix.

Allow heavier particles to sink for 0h 0m 15s. Observe carefully to make sure that this is happening. No large particles should be visible.

Add the appropriate amount of the sample suspension prepared above to the growth medium to make a total of 20mL per dilution series. These 4 dilutions series should be enough for most soil samples.

• 10^-1or 1/10 dilution (2000µL into 18000µL medium)
• 10^-2 or 1/100 dilution (200µL into 19800µL medium)
• 10^-3 or 1/1000 dilution (20µL into 19980µL medium)
• 10^-4 or 1/10000 dilution (2µL into 19998µL medium)

Troubleshoot : If you did not get any growth on the 10^-1 dilution, use the undiluted (10⁰) sample, dilute the original 1g of soil into 100mL medium, or increase the soil sample to 10g.

Pour a dilution into a sterile pipetting reservoir. Use a multichannel pipet to dispense 150µL into each well of a 96-well plate.

Repeat for all 4 dilutions.

Wrap at least two layers of parafilms around the plate, and secure the lid with lab tape. Place the plate into a ziplock bag to minimize moisture loss. Make sure your plates are labeled properly.

Incubate at the desired temperature and check periodically. For soil bacteria, we incubate them for at least 1 week, sometimes up to 2. However, you risk evaporation of your medium for longer or warmer incubations. For warm and/or longer term incubations, place the ziplock bag into a moist chamber inside the incubator.

For each soil sample, determine the dilution plates that have visible growth in less than 30% of the wells. Plates with higher percentages of growth tend to have more mixed-culture wells. Discard plates that have higher number of wells with growth.

Tip: To inspect for growth in your plate, set up an diffused light panel and place the plate on top. This will allow you to see which well has growth and which doesn't. Mark wells on the lid of the plate as appropriate. These lights can commonly be found in the lighting section of hardware stores or specialty lighting stores.

Transfer all the contents of any wells with growth into a new 96-well plate.

It is generally recommended that you refresh the cells so that you don't lose any cultures. Use a 96-well replicating tool to inoculate a new 96-well plate with fresh medium. Incubate using the same parameters as isolation.

Storage: Add enough glycerol (or preferred cryopreservative) to each well so that they contain ~30% (v/v) glycerol. Freeze at -80°C.

Post-identification purification: After identification and determination of purity of a sample, wells with mixed cultures of interest should be streaked out, isolated, and re-identified. Since working with 96 well-plates increases the chance of cross-contamination, any cultures from a well used in future experiments should be streaked out to ensure purity of isolate.

This protocol allows high-throughput identification of soil bacteria in

96-well plate format. It starts with an alkaline lysis of cells, followed by a

2-step PCR method that amplifies and barcodes the samples, then Illumina MiSeq

sequencing of a targeted variable region(s) of the rRNA gene, and a

bioinformatic workflow that cleans up the sequences and identifies the samples.

In our preliminary proof of concept trial, we amplified and sequenced >900

isolates within one month working non-continuously. The final estimated cost

(without technician time) for each sample is $3.50 and the cost will drop with

more samples added to a single MiSeq run. We expect that a single run could

handle 1500 isolates.

Transfer 100µLof cell cultures grown in liquid medium to a fresh 96-well PCR plate.

Spin down at 4000x g,0h 0m 0s for 0h 3m 0s.

Pipet out as much of the growth medium as possible. We find that it’s difficult to pipet out more than 75µLusing a multichannel pipet and not disrupt the cell pellet. A bit of medium left over should not interfere with PCR.

Add 75µL of Extraction Solution (ES) to each well. Seal the plate with a sealing film and gently vortex to mix the cells into solution. Spin down quickly but do not pellet the cells.

Add 75µL of Dilution Solution (DS) so that the final ES:ED solution ratio is 1:1.

Note: The recipes for Extraction and Dilution Solutions is available elsewhere.

Pellet the cells by centrifuging at 4000x g,0h 0m 0s for 3 minutes. The final DNA concentration is ~25 ng/µl. Use 1-2 µl of the clear solution for PCR (avoid the precipitate). For bacteria and yeast samples, we typically use 2µL template in a 15µL PCR reaction.

For PCR, we use a two-step dual barcoding method to amplify and barcode each sample. For detailed methods, please see our lab's Illumina high-throughput sequencing protocol.

Perform the 1^st PCR using the gene primer with overhangs for the second primer. Samples should amplify well, and there should be little to no primer dimers left over. Check all samples for amplification using gel electrophoresis.

Total reaction volume: 15µL

Mastermix: Any good quality mastermix with relatively high-fidelity polymerase.

Template: 2µL of crude extract

Primers: 0.3micromolar (µM) (0.15micromolar (µM) each) final concentration of premixed pairs of primers.

Perform the second PCR using index primer mixes. If you have products in the first PCR, there is a very high chance that it will amplify in the second PCR. We do random spot checks (8 wells) across the plate to make sure that there is good amplification across the plate.

Total reaction volume: 10µL

Mastermix: Any good quality mastermix with relatively high-fidelity polymerase.

Template: 1µLof first PCR

Primers: 0.2micromolar (µM) (0.1micromolar (µM) each) final concentration of premixed pairs of primers. See the Nguyen Lab Index Primer setup.

PCR parameters : Same as the first PCR, except primer annealing is at 52 °C.

Combine all samples in equal volume. Here we assume that all samples have been amplified equally, and in our experience, amplification from the 2^nd PCR tends to be uniform. We typically only use 2 µl per sample in this step.

Perform SPRI bead cleanup on the final sample. Depending on how many samples you have, the final volume will exceed the limits of the tubes. For this step, we usually clean up 500µLof the combined library. The cheapest way to do this is to buy a small, rare-earth magnet and set it up in such a way that it can interact with your tube in a rack. We simply tape it to the rack when needed.

Submit the sample to your favorite sequencing facility to be sequenced using Illumina MiSeq 300PE. Generally, we find that 1500 samples isn't enough for a run so we combine them with other microbiome amplicon libraries.

Use your favorite bioinformatics pipeline to identify the resulting cultures. Interpret your results carefully. Consider these points:

• Illumina sequencing can produce a good bit of noise, so an isolate may appear to be mixed but it isn't. In that case, you will need to determine what portion of the sequence is actual data and what portion is noise.
• You will find that certain isolates have many copies of the 16 rRNA variants. This is not noise, but the true biology of the organism.