Metagenomic Next-Generation Sequencing of Cerebrospinal Fluid for the Detection of Central Nervous System Pathogens

This is a research-based protocol for Metagenomic Next generation sequencing of cerebral spinal fluid samples to allow for the detection of potential pathogens within the sample. Existing MNGS protocols do not account for the low yield of RNA within CSF. This protocol optimizes several steps to account for the low biomass within these samples.

To begin thaw the CSF samples on ice, transfer 100 microliters to one milliliter of each sample to a 1.5 milliliter tube, and centrifuge at 16, 000 G for 10 minutes at four degrees Celsius. Then carefully pipette off the supernatant, leaving behind the pellet and 100 microliters of supernatant. Add 100 microliters of DNA RNA shield to the remaining 100 microliters of the supernatant and pellet and mix thoroughly.

Then add 600 microliters of DNA RNA lysis buffer and mix by pipetting until the solution is homogeneous. Perform the DNA and RNA extraction protocol as per the manufacturer's instructions with the following important modifications. Perform all centrifugation steps for one minute at maximum speed, except the final DNA RNA wash, which should be centrifuged for three minutes.

To elute the sample, add 22 microliters of nuclease free water directly onto the column matrix. Then incubate for three minutes at room temperature. After ellucian pipette the flow through back onto the column matrix, centrifuge again for one minute at maximum speed before transferring the sample to a new 1.5 milliliter storage tube.

Pipette the components required for RNA fragmentation and priming in a PCR tube. Incubate the tube in a thermocycler with the heated lid set to 105 degrees Celsius. Pipette mix the fragmented and primed RNA with eight microliters of nuclease free water and two microliters of first strand synthesis enzyme mix.

Incubate the tube in a thermocycler with the heated lid set to 105 degrees Celsius. Then mix the first strand synthesized DNA with 48 microliters of nuclease free water. Eight microliters of second strand synthesis reaction buffer, and four microliters of second strand synthesis enzyme mix.

Place the tube in a thermocycler and incubate for one hour at 16 degrees Celsius with the heated lid turned off. Next, perform magnetic bead purification using a 1.8 times ratio of solid phase reversible immobilization magnetic beads. Elute the sample into 53 microliters of nuclease free water.

Transfer 50 microliters of the final supernatant containing purified double stranded CDNA to a clean nuclease free PCR tube. At this point, the samples can be frozen at minus 20 degrees Celsius overnight. If the sample was stored at minus 20 degrees Celsius overnight, thaw on ice before restarting.

Pipette mix 50 microliters of purified double stranded CDNA with seven microliters of end preparation reaction buffer, and three microliters of end preparation enzyme mix. Place the samples in a thermocycler and incubate at 20 degrees Celsius for 30 minutes and 65 degrees Celsius for 30 minutes with the heated lid turned off. Next, perform the adapter ligation step on ice by first creating a one to 100 dilution of adapter in nuclease free water.

Pipette the end preparation reaction mixture into a tube containing 30 microliters of ligation master mix one microliter of ligation enhancer and 2.5 microliters of the diluted adapter. Place the samples in a thermocycler and incubate at 20 degrees Celsius for 15 minutes with the heated lid turned off. After that, proceed immediately to magnetic bead purification using a 0.9 times ratio of solid phase reversible immobilization magnetic beads.

Then elute into 17 microliters of nuclease free water. Transfer 15 microliters of the final supernatant to a clean nuclease free PCR tube. Now pipette purified adapter ligated CDNA into a tube with three microliters of user enzyme, 25 microliters of Q five master mix, and 10 microliters of unique barcoded primers.

Place the tube in a thermocycler with the heated lid set to 105 degrees Celsius and incubate at the presented conditions. Perform a final magnetic bead purification using a 0.8 times ratio of magnetic beads. Then elute the sample into 23 microliters of nuclease free water.

Transfer 20 microliters of the final supernatant to a clean nuclease free PCR tube. Quantify the concentration of the sample libraries using a DNA quantification kit and the corresponding fluorimeter. Verify that water controls have substantially lower or unquantifiable concentrations compared to the other samples.

Assess the sample library sizes by running the samples on an automated capillary electrophoresis machine. Use the open source web-based platform, Chan Zuckerberg ID to perform metagenomic analysis for pathogen detection. Log into Chan Zuckerberg ID and click upload.

Select metagenomics as the analysis type and choose the input FASTQ files from the sequencing run. Enter the required sample information including sample name, sample type CSF and nucleotide type RNA or DNA. Once processing is complete, click the project folder containing the samples and navigate to the summary dashboard.

Scroll down to examine the number of reads per sample. Review the percentage of reads passing quality control filtering, and the duplicate compression ratio to assess for biased PCR over amplification. To determine any foreign DNA contamination within the project folder, select each water control sample and click background model to generate the model.

During analysis, apply the model and use threshold filters to set an NTZ score greater than one to remove likely contaminants. Then analyze the data using appropriate statistical software. The capillary electrophoresis and electropherogram data demonstrated a CDNA sample library that was over fragmented and contained adapter dimers not for sequencing.

The CDNA library without significant adapter dimer contamination was ready for pooling and sequencing. CSF metagenomic Next generation sequencing detected diverse pathogens in cohort A and B with at least one pathogen in 73.5%of cohort A samples and 51.2%of cohort B samples. A key challenge is precise pipetting due to low volumes along with time sensitive B drying where under or over drying can compromise results.

While this specific protocol is focused on the pathogen detection side, the host RNA transcripts that are generated by this protocol can be used for additional downstream analyses. So it's important for future studies to compare the performance of this particular library preparation and sequencing protocol to emerging methods that are designed for sparse samples as well, to see how it matches up, not only in terms of pathogen detection, but also for transcriptomic analyses.