Steady-state pre-rRNA Analysis to Investigate the Functional Microbiome

Abstract

The gut microbiome is recognized as a critical regulator of human diseases. Constituents of the microbiota and their individual activities can affect a broad range of disease states related to autoimmunity, cancer, infection, metabolism, mental health, and toxicant exposure. A substantial number of microbiome species are not culturable, limiting their study in vitro. Sequencing methods have allowed quantification of the composition of the microbiome, but methods to characterize the physiological status of bacterial species remain limited. Ribosomal RNA precursors (pre-rRNA) are species-specific intermediates in bacterial ribosomal synthesis, and their levels are highly responsive to environmental changes. Immediately before and during active growth, pre-rRNA levels are high, whereas in non-dividing cells, copy numbers are orders of magnitude lower. These dynamics are conserved in all bacterial species and occur exclusively in viable cells, allowing the specific characterization of living and functional bacteria in their native states. Pre-rRNA analysis has been shown to yield valuable real-time information on the physiology of individual bacterial species within complex samples, beyond what traditional qPCR and sequencing methods can offer. Herein we describe a PCR-based protocol to interrogate and quantify the in situ growth status of bacterial species of interest within a complex microbiome. We also describe an in vitro protocol to characterize the pre-rRNA/growth relationship for a given bacterial species, to provide greater context for values obtained from natural samples. Improved understanding of microbial physiological responses to exposures could reveal novel toxicological mechanisms, biomarkers, and potential treatments.

INTRODUCTION

The gut microbiome is recognized as a critical component in many diseases, including environmental chemical induced toxicities (Claus et al., 2016; Mazidi et al., 2016; Parekh et al., 2015). However, existing tools to elucidate the role of microbiomes in various disease processes are limited. In vitro studies can only be conducted on bacteria that are culturable, whereas many microbiome species remain non-culturable. Furthermore, even when in vitro experiments are possible, their application to various complex in vivo situations can be difficult. Ribosomal RNA (rRNA) gene (e.g. 16S) sequencing allows researchers to inexpensively quantify members within a microbiota, but often lacks resolution due to rRNA sequence conservation and provides no information regarding bacterial viability or functional activity. Moreover, rRNA gene sequencing only reveals relative abundances, making an increase in one population sometimes difficult to distinguish from a decrease in another. Full transcriptome RNA sequencing (RNA-seq), on the other hand, can provide functional information about individual species’ activities, but can be costly (~$200 USD/sample) and requires significant read depth (at greater expense) to characterize lowly-abundant species. While extremely valuable, RNA-seq data also lacks a conserved biomarker that can be used to compare activities across all bacterial species.

Pre-rRNAs are intermediates in rRNA synthesis with leader and tail fragments that are enzymatically removed to yield mature rRNA (Figure 1). In growing bacteria, pre-rRNAs account for ≥25% of total cellular rRNA (Oerther et al., 2000). Due to the abundance and stability of pre-rRNA, it is orders of magnitude easier to detect than even the most strongly-expressed mRNA. When growth slows down, pre-rRNA synthesis declines but maturation continues, resulting in active drainage of pre-rRNA pools (Cangelosi & Brabant, 1997). Pre-rRNA is rapidly replenished when growth-limited cells encounter a more favorable environment (Figure 2) (Mackow & Chang, 1985; Srivastava & Schlessinger, 1990). These changes occur consistently in viable cells of all bacterial species, but are not seen in dead cells or with free nucleic acids. Therefore, this method has the potential to exclusively quantify living and functional bacteria within a complex microbiome community such as the gut microbiome. Another advantage of pre-rRNA analysis is that compared to mature rRNA sequences, which usually share high homology among different bacteria, pre-rRNA sequences are more hypervariable and species-specific. Therefore, pre-rRNA-targeted PCR can detect and assess the physiology of individual species in complex samples (Cangelosi & Brabant, 1997; Cangelosi et al., 2010; Cangelosi & Meschke, 2014; Do et al., 2014; Oerther et al., 2000; Spooner et al., 2016; Stroot & Oerther, 2003; Weigel et al., 2013).

Figure 1: Bacterial rRNA synthesis and maturation.

RNA polymerase (RNAP) transcribes the 30S rRNA operon, which contains three rRNA subunits, external transcribed spacers (ETS), and internal transcribed spacers (ITS). The transcript is rapidly converted by endonucleases to pre-RNA subunits with leader and tail sequences. The leaders and tails are then trimmed in exonucleolytic processes closely tied to ribosome assembly and the initiation of protein synthesis. RT-qPCRs can be designed to target the pre-rRNA exclusively, or to straddle the pre-rRNA-mature rRNA junction as shown (blue arrows), such that only intact pre-rRNA is amplified. For added specificity, hydrolysis probes (red) can target any portion of sequence between the two pre-rRNA primers. The 5′ leader region (ETS1) is typically used because of its species specificity and relative abundance when transcription is active; however, other pre-rRNA sequences (e.g. ITS1) can also be targeted by RT-qPCR.

Figure 2: Pre-rRNA levels during growth phases.

F. alocis was cultured in vitro as described in Support Protocol 1. Carbon-starved cells were provided fresh nutrients at time 0, prompting pre-rRNA synthesis and eventual proliferation. Cell density was determined by optical density (OD₆₀₀) (red line). As growth later slowed (8-72h), pre-rRNA levels declined to baseline. Pre-rRNA was quantified by RT-qPCR and normalized to genomic DNA (P:G ratios, blue bars). Plotting OD₆₀₀ and pre-rRNA levels together allows the determination of a threshold P:G value (black dashed line), differentiating actively growing cells (above threshold) from non-growing cells (below). In this example, a 2-step RT-qPCR method was used, with Improm II RT kit (Promega Corporation, Madison, Wisconsin, # A3800) and Power SYBR Green (i.e. no hydrolysis probe). As mentioned in the text, we now favor a different strategy; see Strategic Planning.

Here, we describe protocols that use bacterial ribosomal RNA precursor (pre-rRNA) levels as a proxy for bacterial physiological status. Steady-state pre-rRNA analysis (SSP) uses RT-qPCR methods to measure species-specific pre-rRNA copy numbers in complex samples. In SSP, pre-rRNA levels are typically normalized to those of genomic DNA (gDNA) from the same species (P:G ratio), providing information on ongoing bacterial replication within natural samples (high P:G ratios correlate with ongoing growth activity) (Cangelosi & Brabant, 1997; Cangelosi & Meschke, 2014; Oerther et al., 2000).

Basic Protocol 1 describes a targeted SSP approach with which the growth status of specific species can be characterized in natural samples. In this protocol, pre-rRNA is quantified by RT-qPCR from extracted RNA and normalized to the more stable mature rRNA or genomic DNA of the same target species. These normalized pre-rRNA values, pre-rRNA:genomic DNA (P:G), directly relate to the bacteria’s growth activity at the time of sample collection. Throughout the protocol we include detailed parameters for characterizing pre-rRNA levels in opportunistic oral pathogens from periodontal paper point samples, while also including notes for applying the protocol to other species and sample types. To maximize the utility of pre-rRNA quantification, experiments characterizing the species-specific relationship between growth response and pre-rRNA levels are recommended, as described in Support Protocol 1. In that protocol, we specify conditions for characterizing pre-rRNA/growth levels of one oral pathogen (Filifactor alocis), with notes about adaptations for other species. Correlating a particular species’ normalized pre-rRNA levels to its various growth states can help provide context to the values obtained from experimental samples using Basic Protocol 1. However, where it is not possible to complete Support Protocol 1 (e.g. targeting non-culturable bacteria), relative comparisons can still be made from samples collected under different conditions.

We have used SSP to examine the in situ growth activities of two oral anaerobes, Porphyromonas gingivalis and Filifactor alocis (formerly Fusobacterium), in periodontal pocket samples collected before and after non-surgical therapy for periodontitis. Importantly, such an approach allows us to obtain a “snapshot” of the growth physiology of these organisms within the sample, without requiring isolation and culture of the organisms from the sample. The pre-rRNA method detected significantly reduced growth activity of P. gingivalis, but not F. alocis, after therapy (Spooner et al., 2016). Specific methods and data derived from this periodontal study are used as examples in both Basic Protocol 1 and Support Protocol 1.

The application of SSP to complex microbial environments (e.g. oral or gut microbiomes) could result in the discovery of unforeseen microbial mechanisms underlying toxicant responses (and other complex disease states) to potentially enable novel intervention strategies.

STRATEGIC PLANNING

The extraction, processing, and handling of RNA requires adherence to measures designed to limit its degradation, including the use of RNase-free reagents and supplies, and the avoidance of unnecessary exposures to extreme pH or elevated temperatures (see (Precautions for Handling of RNA, n.d.)). The recommended kits for extracting RNA and performing reverse transcription will supply RNase-free reagents and other materials (e.g. microcentrifuge tubes). Designated workspaces and frequent usage of RNase-inactivating cleaners (e.g. RNaseZap and ELIMINase) on work surfaces, micropipettors, and gloves (before and during steps 5-7 of both protocols) can help limit contamination from environmental RNases.

The protocols described here require RT-qPCR assays and, therefore, knowledge of the targeted species’ genome (specifically, 16S rRNA and leader sequence[s]). Sequences can be searched for and downloaded from NCBI (Home - Genome - NCBI, n.d.). Some guidance on assay development is provided in these protocols, but an inexperienced researcher would benefit from familiarizing oneself beforehand to some basic resources on this topic (QPCR Assay Design and Optimization ∣ LSR ∣ Bio-Rad, n.d.; Rodríguez et al., 2015). The protocols reference two recommended tools to aid in primer and probe design (Primer3 Input (Version 0.4.0), n.d.; PrimerQuest - Design QPCR Assays ∣ IDT, n.d.). Coverage and specificity of candidate primers and probes can be evaluated in silico using NCBI’s BLAST tool (Nucleotide BLAST, n.d.). Due to anticipated higher copy numbers, we suggest that pre-rRNA primers straddle the ETS/16S junction. However ITS junctions can also be targeted (see Figure 1).

Throughout both protocols, specific examples (e.g. bacteria in paper point samples of periodontal pockets) are provided, along with the detailed methods needed to generate the example data shown in Figures 2 and 3. In these examples, we sometimes refer to manufacturer’s protocols for certain kits and reagents (Protocol: Power SYBR Green PCR Master Mix and RT-PCR, n.d.; MasterPure Complete DNA and RNA Purification Kit, n.d.; ImProm-II Reverse Transcription System, n.d.). However, the illustrative examples do not necessarily reflect our current recommendations for all assays. SYBR Green-based qPCR chemistries have a lower upfront cost (no probe needed), but exhibit lower specificity than probe-based assays (e.g. no probe, primer dimer contributes to signal). This could complicate analysis (e.g. melt curves) and assay validation. We generally advocate for the use of hydrolysis probes (TaqMan) and 1-step RT-qPCR kits (see recommendations in Materials List) due to their increased specificity and simplicity. It should be noted though that 2-step RT-qPCR methods have the advantage of allowing both gDNA and the cDNA generated from RNA to be run on the same qPCR plate with a shared standard curve. If the user chooses a SYBR Green (intercalating dye) method, we suggest using a gene-specific primer (GSP) during RT that is unique from the qPCR primers, for increased specificity. If targeting a large number of taxa with 2-step RT-qPCR methods, we suggest the user to employ a “universal” 16S primer to generate amplifiable cDNA from most species’ pre-rRNA (e.g. 16S-uni-338R: GCTGCCTCCCGTAGGAGT or 16S-uni-519R: GTATTACCGCGGCKGCTG).

Figure 3: Example analyses from growth status (pre-rRNA:gDNA, P:G ratios) data generated in Basic Protocol 1.

(a) Correlation of growth activity (log-transformed P:G ratios) of Porphymonas gingivalis and Filifactor alocis from paper point samples of periodontal pockets; n=51, linear regression, p=0.005. (b) Percent of periodontal pockets (sample sites) exhibiting increased or decreased prevalence (gDNA) or growth status (P:G ratios) of F. alocis or (c) P. ginvivalis following treatment (n=45, two-sample Z-test for proportions, *denotes statistical significance at p<0.05). (d) Prevalence in periodontal pockets (sample sites) of actively growing P. gingivalis and F. alocis before and after (6-18 weeks) non-surgical therapy (n=45, logistic regression, *denotes statistical significance, p<0.05).

If the user is interested in quantifying pre-rRNA levels in bacteria that can be cultured, and has access to the organism(s) and the necessary materials/equipment (anaerobic chamber or GasPaks, incubator, media, etc.), they should complete Support Protocol 1 before beginning Basic Protocol 1. The pre-rRNA/growth level characterization described in Support Protocol 1 will help the user anchor expectations for and contextualize results from Basic Protocol 1, as well as enable additional analyses. Some possible example analyses, with and without the prior completion of Support Protocol 1, are explored in the Understanding Results section. Before beginning Support Protocol 1, the user may benefit from reviewing methods and considerations for microbiological culture (Ahern, n.d.; Burdass et al., 2005; Lagier et al., 2015). To aid in planning for the microbiology steps (2-4) of Support Protocol 1, particularly if growth rate(s) of target organism(s) are not described in the literature, the user should perform a simple growth rate experiment (Hall et al., 2014). When obligate anaerobes are studied, it is critical to consistently limit exposure to oxygen; pre-reduce media and buffers (e.g. 0.01-0.1% L-cysteine) and monitor redox status (e.g. 1 mg/L resazurin).

Basic Protocol 1: Targeted Steady-state Pre-rRNA Analysis

This taxa-targeted approach is designed to characterize the growth status of a particular bacteria of interest within a sample. It requires knowledge of sequence data of relevant regions (e.g. 16S rRNA gene + leader), to allow design of specific oligonucleotide primers/probes. Nucleic acid is extracted from a sample, and species-specific pre-rRNA is quantified by RT-qPCR (preferably with hydrolysis probes) and normalized to genomic DNA (gDNA). When possible, characterizing the relationship between pre-rRNA levels and the growth response of targeted species, as described in Support Protocol 1, will enhance the utility of data generated from this protocol. In this example protocol, we describe how to analyze oral microbiome (periodontal pocket) paper point samples and quantify pre-rRNA levels of opportunistic oral pathogens (for simplicity, focusing on a single species, Filifactor alocis). Generalized notes are included for applying the protocol to other target species and sample types.

Materials:

• Microbiome samples expected to harbor species of interest (e.g. oral swabs, fecal matter, etc.)

• Solutions, reagents, and kits:

  • Nuclease-free TE buffer (see Reagents and Solutions)
  • Oligonucleotide primers targeting pre-rRNA (e.g. ETS/16S junction, See Step 1)
  • Hydrolysis probe(s) targeting pre-rRNA (e.g. ETS/16S junction, See Step 1)
  • RT and qPCR reagents (preferred, 1-step, e.g. Verso 1-step RT-qPCR #AB4101A; alternatively, 2-step, e.g. Thermo Superscript IV RT #11756050 and BioRad iTaq Universal Probes Supermix for qPCR #1725130).
  • Kit for DNA/RNA extraction (e.g. Qiagen AllPrep PowerFecal DNA/RNA [#80244] or MasterPure Complete DNA and RNA Purification Kit [Lucigen #MC85200])
  • DNase I (if not included in extraction kit)
  • RNAlater (Sigma #R0901) (if unable to quickly deep freeze samples during collection)
  • RNase-inactivating cleaners (e.g. RNaseZap [Sigma #R2020] and ELIMINase [Decon Labs #1101/2])
• Special equipment:

  • 0.1 mm acid-washed zirconia or glass beads for bacterial lysis (e.g. OPS Diagnostics #BAWZ 100-250-15)
  • Vortex adapter for bead-beating cell lysis (e.g. Qiagen 13000-V1-24)
  • Screw-cap o-ring tubes for safe lysis
  • UV spectrophotometer (e.g. Nanodrop)
  • Thermocycler
  • PCR plates and covers
  • Microcentrifuge tubes
  • Microcentrifuge
  • Ice and ice bucket
  • Dry ice or liquid nitrogen (if access to −80 °C freezer is not available during sample collection)

Protocol Steps

1. Design oligonucleotide primers and fluorescent hydrolysis probes to amplify and detect pre-rRNA from targeted taxa. The same set can be used if normalizing to gDNA.

   Primers targeting pre-rRNA typically straddle the ETS-16S junction (see Figure 1). Specificity is generally derived from F primer (ETS) and/or hydrolysis probe, due to sequence variability. ETS (F) primer should remain proximal to the 16S junction to ensure detection of transcribed sequence, at least within 80 bp (as shown in Figure 1). We recommend starting with a nucleotide alignment of target taxa along with close relatives that need to be excluded, using a tool such as NCBI’s Multiple Sequence Alignment Viewer. Note regions within the area of interest (e.g. ETS/16S) that contain exploitable sequence differences, and focus primer/probe design tools (e.g. Primer3 or IDT’s PrimerQuest) on those regions. Small (even single nucleotide) differences can still offer specificity if positioned on 3’ end of primers. Please see example primer sequences chosen for F. alocis in Support Protocol 1, step 7. We generally recommend amplicon sizes between 70-250bp; typically shorter amplicons are preferred over longer ones.

2. Validate primer/probe functionality with purified gDNA and/or RNA. Confirm specificity with off-target nucleic acid as needed. This can be done in silico (e.g. using NCBI’s BLAST tool), by endpoint PCR, and/or qPCR.

   Empirical validation of specificity and complete coverage of target taxa is recommended, but this is often not feasible with limited access to purified isolates. In silico validation in these cases is often sufficient, and increasingly so as more isolates are sequenced.

3. If possible, characterize pre-rRNA/growth relationship(s) of targeted taxa as described in Support Protocol 1.

   We encourage users to complete Support Protocol 1. Correlating a particular species’ normalized pre-rRNA levels to its various growth states can help provide context to the values obtained from natural samples in Basic Protocol 1. These characterization experiments can also anchor expectations based on the specific methods used (primers, reagents, etc.). However, where it is not possible to complete Support Protocol 1 (e.g. targeting non-culturable bacteria), relative comparisons can still be made between samples collected under different conditions.

4. Collect samples from experimental groups from which bacterial pre-rRNA levels will be assessed. To minimize variability and confounding effects, attempt to standardize time and method of collection for all samples. Include biological replicates to increase confidence in results.

   Freeze (−80 °C) samples as soon as possible (within 5 minutes) after collection to avoid pre-rRNA processing or degradation. Dry ice or liquid nitrogen should be used if ready access to a −80 °C freezer is not experimentally feasible. When immediate freezing after sample collection is not possible, we recommend the use of a preservative such as RNAlater.

   Sample collection method will vary with type. Fecal samples and oral swabs can be frozen directly, whereas samples with low numbers of bacteria (e.g. urine) should be centrifuged and supernatants discarded before freezing cell pellets. For our example periodontal paper points, we recommend direct freezing on dry ice immediately after collection. To avoid artifacts, strive to treat all samples within an experiment equally.

5. From equal volume/mass samples, extract RNA and DNA using a kit such as those suggested in the Materials list. Consider expected cell type and number, as well as state of sample (e.g. cell pellet, fecal homogenate, oral swab) and adjust procedure according to manufacturer’s recommendations.

   For periodontal paper points, use MasterPure Complete DNA and RNA Purification Kit. Briefly, suspend paper points in the provided lysis buffer (300 μL and proteinase k solution [1 μL]) prior to following the furnished protocol for purifying total nucleic acid from cell samples (ImProm-II Reverse Transcription System, n.d.). Vortex the suspension for 1 min and heat at 65 °C for 15 min to lyse the cells. Elute Total nucleic acid (TNA) in 30 μL TE, and from that, remove 10 μL for DNA measurement. From the remaining 20 μL, purify RNA by DNase I treatment and re-precipitate as directed by the included protocol, and resuspend in 20 μL of RNAse-free water.

   For microbiota or other samples where gram-positive species are present, a mechanical lysis step (e.g. bead-beating) is recommended. Three minute bead-beating (3 x 1 min with 1 min on wet ice between sessions) with vortex adapter at max speed is a good starting point.

   Longer or more energetic bead-beating may introduce heat that can degrade RNA. For tissues, an upfront homogenization step may be needed, (e.g. using larger dense beads, LN2 grinding, sonication, etc.).

6. Assess DNA/RNA concentration and purity by UV spectrophotometry; re-extract if concentration or purity is too low for effective RT or qPCR.

   DNA and RNA concentration are estimated by absorbance at 260 nm relative to the same (nucleic acid-free) solution (e.g. water, TE buffer) used to suspend/elute the nucleic acid in step 5. Note sample concentrations for subsequent steps. Contaminants often absorb at 230 nm and 280 nm (e.g. protein, guanidine, etc.). Ideally, 260/280 ratios should be ~1.7-2.0 for pure DNA and ~2.0 for pure RNA, whereas 260/230 ratios will be between ~1.8-2.2 for pure DNA and RNA. However, lower ratio values (if above ~1.4) do not necessarily indicate a fatal problem with the samples, particularly when they are consistent among the sample set. Note the ratio values for later troubleshooting, if needed. Please see Troubleshooting for notes on nucleic acid concentration.

7. Conduct RT-qPCR on purified RNA and qPCR on DNA. Follow RT and qPCR kit/reagent recommendations. One-step RT-qPCR kits utilizing gene-specific primers (GSPs) usually lead to the most consistent success and are what we recommend. However, 2-step (separate RT and qPCR) procedures using GSPs or random primers may also work. Sub-steps described below are for the 2-step RT-qPCR (with GSP) used to generate sample data for F. alocis (Figure 3):

   1. Convert purified F. alocis RNA to cDNA by reverse transcription (RT) with the ImProm II system according to the provided instructions, using 4 μL of template and 3 μM of 16S-specific primers (GSP: 5’-TACTGATCGTTGCCTTGGTG-3’) in 15 μL reactions. Choose a consistent RNA volume between samples that avoids any samples exceeding the maximum advised RNA input (e.g. 1 ug for Improm-II). Include “no-RT” controls to ensure DNA has been adequately removed from RNA preparations.
   2. Measure cDNA generated in step 7a and genomic DNA from step 5 by qPCR (using the same primer/probe set) and Power SYBR Green. Each 20 μL reaction should contain 2 μL of template (cDNA or gDNA) and 375 nM each of forward and reverse primers (Fwd: AACCGGAGCAAAACTGAGAA and Rev: CCGTCCGCCACTAACTTCTA). Include a ~7-log standard curve for absolute quantification of samples. Also include no-RT controls from step 7a and no-template controls (e.g. water) for evaluation of background signal. Run on an Applied Biosystems StepOnePlus under the following conditions: 95 °C for 10 minutes followed by 40 cycles of 95 °C for 15 seconds, 57 °C for 30 seconds, and 72 °C for 30 seconds. Use automatically set thresholds by the StepOnePlus software or manually adjust in the exponential range when necessary. Export Cq values and quantities to spreadsheet for analysis.

8. For a given targeted taxon, compare levels of pre-rRNA (P) and the normalizing gDNA (G). Calculate pre-rRNA:gDNA (P:G) ratios using quantity values generated relative to standard curves, or if no standard curve is available, by 2^(gDNA Cq – pre-rRNA Cq).

   Cq values between ~15-35 are generally best for accurate quantification. Dilution of samples may be needed to achieve such values. Adjust copy number from diluted samples by multiplication with dilution factor, or Cq values by subtracting log₂(dilution factor). Confirm PCR performance with standard curves if possible. For P:G ratio calculations, it is preferred to use target quantities derived from standard curves (if a sample has 100 pre-rRNA copies and 10 gDNA copies, its P:G ratio for that species is 10), rather than the approximating formula: 2^(gDNA Cq – pre-rRNA Cq), as this formula assumes 100% PCR efficiency, which is rarely the case; approximated ratios will be increasingly imprecise as efficiencies deviate from 100%.

9. Compare P:G values between experimental groups (e.g. different populations, clinical interventions, diets, disease status, etc.) to observe relationships with target species’ pre-rRNA levels. Compare P:G values to those from in vitro characterizations (see Support Protocol 1) to gauge growth status of target(s) within samples.

   Appropriate and desired analyses will vary depending on experimental design. Please see Understanding Results section for example analyses.

Support Protocol 1: Characterization of pre-rRNA/growth relationship

In vitro experiments characterizing the relationship between taxa-specific pre-rRNA levels and growth status may help provide meaning to values ascertained from natural samples (Basic Protocol 1). These characterization experiments can also help anchor expectations based on the specific methods used (primers, reagents, etc.). Here, pre-rRNA pools are “drained” from bacterial cultures by transiently withholding nutrients. Following the provision of nutrients, growth is monitored and samples are collected at time points to measure pre-rRNA levels. Pre-rRNA levels are measured and normalized using the primers/probes designed and validated in Basic Protocol 1 (steps 1-2). In this example protocol, we describe how to characterize the pre-rRNA/growth relationship of the opportunistic oral pathogen Filifactor alocis, and include some generalized notes for applying the method to other species.

Materials:

• Bacterial species/strain of interest (e.g. F. alocis ATCC 35896)

• Solutions, reagents, and kits:

  • Phosphate Buffered Saline (PBS) (see Reagents and Solutions) or similar nonnutritive buffer
  • Nutritive bacterial growth media (depends on targeted organisms). For example, for F. alocis: brain heart infusion (BHI) broth, supplemented with yeast extract (0.5 mg/mL), L-cysteine (50 μg/mL), and 20% arginine
  • L-cysteine (reducing agent for anaerobic bacteria; Sigma #168149)
  • Resazurin (redox indicator for anaerobic bacteria; Sigma #R7017)
  • Nuclease-free TE buffer (see Reagents and Solutions)
  • Oligonucleotide primers targeting pre-rRNA (e.g. ETS/16S junction, Basic Protocol 1)
  • Hydrolysis probe(s) targeting pre-rRNA (e.g. ETS/16S junction, Basic Protocol 1)
  • RT and qPCR reagents (preferred, 1-step, e.g. Verso 1-step RT-qPCR #AB4101A; alternatively, 2-step, e.g. Thermo Superscript IV RT #11756050 and BioRad iTaq Universal Probes Supermix for qPCR #1725130).
  • Kit for DNA/RNA extraction (e.g. Qiagen AllPrep PowerFecal DNA/RNA [#80244] or MasterPure Complete DNA and RNA Purification Kit [Lucigen #MC85200]))
  • DNase I (if not included in extraction kit)
  • Ice and ice bucket
  • RNase-inactivating cleaners (e.g. RNaseZap [Sigma #R2020]and ELIMINase [Decon Labs #1101/2])
• Special equipment:

  • 0.1 mm acid-washed zirconia or glass beads for bacterial lysis (e.g. OPS Diagnostics #BAWZ 100-250-15)
  • Vortex adapter for bead-beating cell lysis (e.g. Qiagen 13000-V1-24)
  • Vortexer
  • Screw-cap o-ring tubes (for safe lysis)
  • Thermocycler
  • PCR plates and covers
  • Microcentrifuge tubes
  • UV/Vis spectrophotometer (e.g. Nanodrop)
  • Anaerobic chamber (if needed) or anaerobic pouches (e.g. BD GasPak, #BD 260683)
  • Incubator with shaker or drum
  • Microcentrifuge

Protocol Steps:

1. Prepare at least three bacterial cultures according to established procedures. Ideally, streak for isolation on solid media and select single colonies to inoculate replicate broth cultures. We suggest broth cultures ≥ 1 mL.

   Please see Strategic Planning for references on microbiological methods.

2. Incubate broth cultures and allow bacteria to reach mid- to late-logarithmic growth (as assessed by optical density [OD₆₀₀]).

   The time needed to reach mid- to late-logarithmic growth and the final OD₆₀₀ achieved depends upon the species/strain, density and status of inoculum, and growth conditions (including media, aeration, temperature, etc.). For F. alocis grown in supplemented BHI (described above), it will typically take 12-24 hours, to reach an OD₆₀₀ of ~0.1-0.3. In contrast, E. coli grown in LB broth will typically take 1-2 hours for an OD₆₀₀ of 0.1-1.0.

3. Centrifuge cells at RT for 2 min at >13000 g, and discard the supernatant (removing cell-free nucleic acids, residual nutrients, and inhibitors). Resuspend in an equal volume of pre-warmed (and pre-reduced [e.g. 1 g/L L-cysteine] if using anaerobic bacteria) non-nutritive buffer (e.g. phosphate-buffered saline). Incubate cell suspension under growth conditions for 3-10 doubling times to thoroughly drain pre-rRNA pools.

   Bacterial growth is expected to stop when nutrients are scarce, and pre-rRNA levels will decline under these conditions. Cultures can alternatively be depleted of pre-rRNA pools by continued incubation in media in step 2 (rather than transferring cells to non-nutritive buffer) until well after growth ceases (3-10 doubling times). Surfactant (e.g. 0.05% Tween-80) in media and buffer is recommended to mitigate adhesion and cell loss. Please see Strategic Planning for microbiology notes.

4. Dilute stationary-phase cell suspensions 1:10 in pre-warmed growth media and incubate under preferred conditions. Immediately record OD and collect a “time 0” time point sample for SSP. At several time points, record OD and take samples for SSP. To fully capture growth curve dynamics, time points should range before the first doubling time until after the culture reaches late stationary phase.

   Samples volumes for OD and SSP time points should represent a negligible volume relative to the total volume of the culture (< 5%) to avoid unintentionally altering conditions. To collect samples, take a volume from the culture, centrifuge at >13000g for 1-2 min at RT, discard supernatant, and immediately store cell pellets at −80 °C until nucleic acid extraction. In addition to a time 0 time point, we generally recommend, at a minimum: one time point sample at ~half the doubling time, one at ~the doubling time, and three additional ones, at e.g. 2, 4, and 8 doubling times.

5. Extract RNA and DNA using kit such as those suggested in the materials list. Consider expected cell type and number, as well as state of sample (e.g. cell pellet, fecal homogenate, oral swab) and adjust the procedure according to the manufacturer’s recommendations.

   For the F. alocis example shown in Figure 2, MasterPure Complete DNA and RNA Purification Kit was used. Here, briefly, resuspend cell pellets in the provided lysis buffer (300 μL and proteinase k solution [1 μL]) prior to following the furnished protocol for purifying total nucleic acid from cell samples (MasterPure Complete DNA and RNA Purification Kit, n.d.). Vortex the suspension for 1 min and heat at 65 °C for 15 min to lyse the cells. Elute Total nucleic acid (TNA) in 30 μL TE, and from that, remove 10 μL for DNA measurement. From the remaining 20 μL, purify RNA by DNase I treatment and re-precipitate as directed by the included protocol, and resuspend in 20 μL of RNAse-free water. For microbiota or other samples where gram-positive species are present, a mechanical lysis step (e.g. bead-beating) is recommended. Three minute bead-beating (3 x 1 min with 1 min on wet ice between sessions) with vortex adapter at max speed is a good starting point. Longer or more energetic bead-beating may introduce heat that can degrade RNA. For tissues, an upfront homogenization step may be needed, (e.g. using larger dense beads, LN2 grinding, sonication, etc.).

6. Assess DNA/RNA concentration and purity by UV spectrophotometry. Re-extract if concentration or purity is too low for effective RT or qPCR.

   DNA and RNA concentration are estimated by absorbance at 260 nm relative to the same (nucleic acid-free) solution (e.g. water, TE buffer) used to suspend/elute the nucleic acid in step 5. Note sample concentrations for subsequent steps. Contaminants often absorb at 230 nm and 280 nm (e.g. protein, guanidine, etc.). Ideally, 260/280 ratios should be ~1.7-2.0 for pure DNA and ~2.0 for pure RNA, whereas 260/230 ratios will be between ~1.8-2.2 for pure DNA and RNA. However, lower ratio values (if above ~1.4) do not necessarily indicate a fatal problem with the samples, particularly when they are consistent among the sample set. Note the ratio values for later troubleshooting, if needed. Please see Troubleshooting for notes on nucleic acid concentration.

7. Conduct RT-qPCR on purified RNA, and qPCR on DNA. Follow RT and qPCR kit/reagent recommendations. One-step RT-qPCR kits utilizing gene-specific primers (GSPs) usually lead to the most consistent success and are what we recommend. However, 2-step (separate RT and qPCR) procedures using GSPs and/or random primers may also work. Sub-steps are described below for the 2-step RT-qPCR (with GSP) used to generate the sample data for F. alocis shown in Figure 2:

   1. Convert purified F. alocis RNA to cDNA by reverse transcription (RT) with the ImProm II system according to the provided instructions, using 4 μL template and 3 μM 16S-specific primers (GSP: 5’-TACTGATCGTTGCCTTGGTG-3’) in 15 μL reactions. Choose a consistent RNA volume between samples that avoids any samples exceeding the maximum advised RNA input (e.g. 1 ug for Improm-II). Include “no-RT” controls to ensure DNA has been adequately removed from RNA preparations.
   2. Measure cDNA generated in step 7a and genomic DNA from step 5 by qPCR (using the same primer/probe set) and Power SYBR Green. Each 20 μL reaction should contain 2 μL template (cDNA or gDNA) and 375 nM each of forward and reverse primers (Fwd: AACCGGAGCAAAACTGAGAA and Rev: CCGTCCGCCACTAACTTCTA). Include a ~7-log standard curve for absolute quantification of samples. Also include no-RT controls from step 7a and no-template controls (e.g. water) for evaluation of background signal. Run on an Applied Biosystems StepOnePlus under the following conditions: 95 °C for 10 minutes followed by 40 cycles of 95 °C for 15 seconds, 57 °C for 30 seconds, and 72 °C for 30 seconds. Use automatically set thresholds by the StepOnePlus software or manually adjust in the exponential range when necessary. Export Cq values and quantities to spreadsheet for analysis.

8. For a given targeted taxon, compare Cq values from pre-rRNA (P) and the normalizing gDNA (G). Calculate pre-rRNA:gDNA (P:G) ratios using quantity values generated relative to standard curves, or if no standard curve is available, by 2^(gDNA Cq – pre-rRNA Cq).

   Cq values between ~15-35 are generally best for accurate quantification. Dilution of samples may be needed to achieve such values. Adjust copy number from diluted samples by multiplication with dilution factor, or Cq values by subtracting log₂(dilution factor). Confirm PCR performance with standard curves if possible. For P:G ratio calculations, it is preferred to use target quantities derived from standard curves (if a sample has 100 pre-rRNA copies and 10 gDNA copies, its P:G ratio for that species is 10), rather the approximating formula: 2^(gDNA Cq – pre-rRNA Cq), as this formula assumes 100% PCR efficiency, which is rarely the case; approximated ratios will be increasingly imprecise as efficiencies deviate from 100%.

9. Calculate means and standard deviations of biological replicates of P:G from each time point. Correlate P:G values to OD measurements, as shown in Figure 2.

   Please see Understanding Results for guidance on calculating means and standard deviations. Generally, higher values indicate active or imminent growth whereas lower values indicate stasis or decline. Expected results should resemble those shown in Figure 2. Typically, normalized values (P:G ratios) range 10-fold or more.

REAGENTS AND SOLUTIONS

Nuclease-free TE buffer:

(50 mL Tris-Cl [10 mM], 10mL EDTA [1 mM]); store up to 2 years unopened at ambient temperature.

Phosphate Buffered Saline (PBS):

(8 g of NaCl [137 mM], 0.2g KCl [2.7 mM], 1.44 g of Na₂HPO₄ [10 mM], 0.245 g of KH₂PO₄ [2 mM]), pH=8. Store up to 2 years unopened at ambient temperature.

COMMENTARY

BACKGROUND INFORMATION

The biological diversity and complexity of gut microbiotas present formidable challenges to our investigation and understanding of these microbial inhabitants. To comprehend their physiological and pathological functions, it is crucial to characterize the basic microbiology of these microbes, such as how and why different species grow and divide in vivo (Schmidt et al., 2018). Despite intense and expanding research on gut microbiota, many basic questions remain unanswered (Almeida et al., 2019). The inability to culture many gut bacteria species in vitro limits researchers’ investigations in the laboratory. Significant progress has been made in this area (Lagier et al., 2018), but even as more bacteria become culturable, the translation of findings from in vitro studies to myriad in vivo contexts still remains difficult (Schmidt et al., 2018). Therefore, methods that can directly assess the physiology of microbiota members in situ are highly desirable (Biteen et al., 2016). Measuring pre-rRNA levels of bacteria within complex samples can reveal more information than traditional qPCR and sequencing methods (Cangelosi et al., 2010; Oerther et al., 2000; Spooner et al., 2016). The ability to obtain real-time quantifiable data on the physiology of individual bacterial species could open the door to new inquiries, and the discovery of microbial interactions in various diseases or in the wake of toxicant exposures.

The universally rapid and dynamic regulation of bacterial pre-rRNA levels was first exploited in a technology called molecular viability testing (MVT) (Cangelosi et al., 2010; Do et al., 2014; Weigel et al., 2013, 2017). In MVT, reverse transcription quantitative polymerase chain reaction (RT-qPCR) methods are used to sensitively and specifically detect viable bacterial cells within a clinical or environmental sample. To conduct MVT, a sample is split and each portion is briefly exposed to either favorable (e.g. nutritional media) or non-favorable (buffer) conditions. After less than one doubling time, nucleic acid is extracted and species-specific pre-rRNA levels in each fraction are compared. The synthesis of pre-rRNA in response to nutritional stimulation (stimulated > unstimulated) reliably reveals the presence of viable cells without the need for microbiological culture. MVT is highly sensitive (Do et al., 2014; Weigel et al., 2017), species-specific, and works for numerous species and diverse complex sample types (Cangelosi et al., 2010; Do et al., 2014; Weigel et al., 2013, 2017).

Here, we describe protocols based on a related technology that uses bacterial ribosomal RNA precursor (pre-rRNA) levels as a proxy for physiological status. Steady-state pre-rRNA analysis (SSP) uses RT-qPCR methods to measure species-specific pre-rRNA copy numbers in complex samples, without nutritional stimulation and without culturing bacteria from samples. In SSP, pre-rRNA is typically normalized to genomic DNA (gDNA) of the same species (P:G ratio), providing information on ongoing bacterial replication within natural samples (high P:G ratios correlate with ongoing growth activity) (Cangelosi & Brabant, 1997; Cangelosi & Meschke, 2014; Oerther et al., 2000). Recently, others have shown that Mycobacterium tuberculosis pre-rRNA levels (normalized to mature rRNA) correlated with bacillary replication, reflected the sterilizing potency of antibiotics (in vitro and in vivo), and were predictive of microbiologic relapse in a mouse model (Walter et al., 2021).

There are significant advantages and drawbacks to normalizing to either DNA or RNA. Normalizing to DNA requires extracting both RNA and DNA, complicating procedures and limiting available extraction methods. However, DNA is an inherently more stable target (~1 genome per cell) than RNA, and in SSP, DNA can be measured with the same primers/probes used for pre-rRNA. Extracting only RNA may simplify procedures and introduce less experimental variability because its isolation can be achieved through the use of a single method. Importantly, this approach (whether normalizing to gDNA or rRNA) provides a “snapshot” of the growth physiology of specific organisms within a sample, without requiring isolation and culture of the organisms from the sample.

The targeted RT-qPCR-based SSP approach described in Basic Protocol 1 is well suited to studies in which specific taxa are thought to be key players. The upfront investment is significant, involving primer/probe design and validation, and ideally, in vitro characterization(s) of the pre-rRNA/growth relationship (Support Protocol 1). However, this approach offers the distinct advantage of providing information on the in situ growth physiology of specific bacteria within complex samples.

We have developed pre-rRNA methods (SSP and molecular viability testing), and we and others have shown that they are highly sensitive biomarkers for diverse viable bacteria and human diseases (Agossou, n.d.; Cangelosi et al., 2010; Do et al., 2014; Lee & Bae, 2018; Spooner et al., 2016; Walter et al., 2021; Weigel et al., 2013, 2017). However, species-targeted PCR-based methods lack the throughput for convenient multi-species testing. 16S sequencing is high-throughput, but blind to the physiological status of microorganisms and often lacks the resolution to identify individual species. Furthermore, 16S sequencing only reveals relative abundances, making an increase in one population sometimes difficult to distinguish from a decrease in another. Metagenomics and transcriptomics can both identify individual species, but often require a high number of reads (and, therefore, a high cost) to do so due to the untargeted nature of the methods. Transcriptomic sequencing (RNA-seq) can powerfully inform on functional aspects of the microbiome, but remains expensive ($200/sample) and lacks a common biomarker for growth status, making inter-species comparisons difficult. In its current form, SSP methods remain low-throughput, but offer unique advantages to investigate microbial physiology within complex microbial consortia.

Potential applications of SSP are numerous, allowing for the comparison of species’ growth status from microbiota among various experimental groups (e.g. different populations, clinical interventions, diets, disease status, etc.) to discover new mechanistic relationships, between microorganisms and their interplay with host health. Given the strong need for greater understanding in this area, this innovative approach will undoubtedly lead to new research directions, and will lay the foundation for further mechanistic investigations of the interactions between microbiota and various diseases. Because this culture-independent approach can be applied in a hypothesis-driven or exploratory manner, discovery of unforeseen microbiomic relationships is enabled.

CRITICAL PARAMETERS

To ensure comparability, samples within an experiment must be treated identically throughout the protocols, from sample collection to quantification by (RT-)qPCR. Importantly, this also applies to the methods used in Support Protocol 1 vs Basic Protocol 1. The user should carefully choose methods for Support Protocol 1 that are compatible with constraints in the experimental design of Basic Protocol 1. For example, if in Basic Protocol 1 the user is analyzing stool samples that were (or must be) transported on ice packs for <1hr before being frozen at −80 °C, a similar treatment should be applied to the in vitro samples in Support Protocol 1.

Regarding the conditions needed for microbiological culture in Support Protocol 1, a specific example is provided in the protocol (F. alocis) and more generalized guidance can be found in Strategic Planning. Given the widely varied needs of different taxa, it is beyond the scope of this protocol to describe this complicated subject in depth. In addition to the resources in Strategic Planning, we recommend consulting the American Type Culture Collection (ATCC) for general (ATCC - Bacterial Culture Guide, n.d.) and strain/species-specific (ATCC Bacteria Catalog, n.d.) recommendations. Further reading of the available scientific literature may be helpful or necessary, depending on which taxa are targeted. In cases where species of interest appear nonculturable, the user can still complete Basic Protocol 1, with somewhat limited analysis options as discussed in step 9 and in the Understanding Results section.

Primer and probe design considerations are explored within Basic Protocol 1 and further guidance is provided in Strategic Planning. The linked resources should help the user throughout this process, including the validation of developed (RT-)qPCR assays. Regardless of whether SYBR Green (intercalating dye) or probe-based qPCR methods are used, it is important to estimate the efficiency of new assays (Real-Time PCR, n.d.), and make modifications if needed until efficiency is 100% ±10%. If intercalating dye (e.g. SYBR Green) chemistries are used, the user should incorporate melt-curve analyses to help ensure that detected amplification represents the desired product (Pryor & Wittwer, 2006). All users are encouraged to consult the MIQE (minimum information for publication of quantitative real-time PCR experiments) guidelines to maximize the reliability of qPCR data interpretations (Bustin et al., 2009).

TROUBLESHOOTING

Table 1 summarizes common issues that may arise during nucleic acid extractions, reverse transcription, and PCR amplification for the described protocols, along with possible causes and solutions.

Table 1:

Common troubleshooting issues.

Problem	Possible Cause	Solutions
Poor nucleic acid yield	Nucleic acid degradation	Ensure reagents and disposables are nuclease-free. Replace as needed. Minimize processing times and heat exposure. Use buffered solutions and chelating agents (e.g. TE buffer). Test extraction outputs (e.g. spectrophotometry, Qubit or gel electrophoresis) to isolate cause.
	Inadequate cell lysis	Try shorter/longer bead-beating times. Short sessions with incubations on ice in-between may help.
	Exceeding method parameters	Ensure samples are within performance expectations of kit (i.e. not overloaded) as described in included documentation. Please see “Input concentration” below.
RT failure	Poor RT primer hybridization	Spontaneous formation of secondary rRNA structures may interfere with RT primer hybridization. Attempt pre-heating and crash-cooling template/primer mix prior to RT. Try alternative enzymes, mixes, or conditions. Use gene-specific primers rather than random primers. Redesign primer if needed.
	Poor RNA quality	Please see "Nucleic acid degradation" above
	Inhibition	Confirm quality of nucleic acid spectrophotometrically. Try alternative extraction method or clean output by ethanol precipitation, etc.
	Input concentration	Refer to manufacturer’s instructions to avoid adding too much nucleic acid. Practical minimum concentrations cannot be generalized (depends on sample nature).
Poor amplification	Poor primer/probe functionality	Validate with known-good template if possible. Try alternative reaction parameters and chemistry. Replace or redesign oligos if needed.
	PCR inhibition	Please see "Inhibition" and “Input concentration” above.
Amplification of DNA but not RNA	Forward primer too upstream of 16S	Redesign forward primer closer to ETS/16S junction.
	RT failure	Please see "RT failure" above.
Poor specificity	Stringency too low	Increase PCR stringency (e.g. higher annealing temp, lower [Mg2+]. Try alternative mastermixes. Redesign oligos if needed.
	Non-specific primer annealing	Use hydrolysis probes for added specificity (compared to SYBR Green). Redesign oligos with less degeneracy.

UNDERSTANDING RESULTS

The range of normalized pre-rRNA (P:G) values will depend on the specific organisms and the exact procedural methods used. For maximum utility, we recommend in vitro characterization of the pre-rRNA/growth relationship as described in Support Protocol 1, using methods as similar as possible to those used in Basic Protocol 1.

Because pre-rRNA and gDNA are not measured with equal efficiency (identical methods cannot be used to isolate and measure both), efficiencies/yield vary depending on specific methods, and because targeted cells within a sample may be heterogeneous with regard to growth state, P:G ratios are not expected to accurately reflect the number of pre-rRNA molecules per cell. For example, the data shown in Figure 2 should not be interpreted to mean that F. alocis ever had fewer than 1 pre-rRNA molecule per genome or cell. The actual number of pre-rRNA molecules in sampled cells was not determined. However, relative changes in P:G values can be expected over the time course (as cells adjust their pre-rRNA levels to their environment). We expect to see increasing levels immediately before and during growth (typically 10-fold or more), then a decline in later time points as nutrients are consumed and growth slows. The completion of Support Protocol 1 informs the user how absolute values of P:G ratios correlate to various states of growth, for the specific methods applied to a specific species/strain. Moreover, P:G ratios are usually dynamic enough to allow the establishment of a threshold value (manually estimated, just above the highest P:G ratios observed before nutrient provision or after log-phase growth ends) delineating growing and non-growing cells (Figure 2). When Support Protocol 1 cannot be completed (e.g. non-culturable target organism[s]), relative comparisons can still be made between experimental groups in Basic Protocol 1. For example, by applying the same methods to periodontal pocket samples taken before and after non-surgical periodontal therapy (supra- and sub-gingival debridement via scaling and root planing, and oral hygiene instructions), it was possible to generate comparative results (Figure 3). It may be additionally helpful to spike samples with a characterized species in a known state to control for methods that cannot be held constant or have potentially varied efficiencies (e.g. stationary-phase E. coli [~10⁴ cells]).

Regarding data generated from Basic Protocol 1, the analyses that are possible, appropriate, and desired will vary depending on experimental design. Example analyses that can be done without the completion of Support Protocol 1 (e.g. when targeting non-culturable organisms) are described first. To start, we suggest calculating means and standard deviations of P:G ratios (among at least three replicates) from each sample group. Generally, higher values indicate active or imminent growth whereas lower values indicate stasis or decline. Comparing P:G ratios between groups (or before and after some intervention) can provide an indication of the relative environmental favorability (for a given bacterial species) in one condition versus another. For example, increased Clostridium difficile P:G values in gut microbiome samples following antibiotic treatment would suggest that a more favorable environment (for C. difficile) was created by the treatment. Student’s t-test (T.TEST Function, n.d.) or ANOVA (One Way ANOVA, n.d.) can test for statistically significant differences between two or more experimental groups, respectively.

Growth status (P:G ratios) can be correlated with experimental parameters (or with another organism’s growth status, as seen with the log-transformed P:G ratios in Figure 3a) by regression (Linear Regression Analysis in Excel, n.d.). The analysis shown in Figure 3a can reveal the extent to which environmental favorability for one organism (F. alocis) is correlated to favorability of another (P. gingivalis), or point to interactions between species. Of course, this specific analysis is only possible if two or more species are tested; experimental questions and biological relevance will dictate whether this should be done. In time courses or following a clinical intervention, P:G ratios can be compared over time or pre- and post-intervention (see Figures 3b and 3c). Regression analysis can be used for time course data, while direct comparisons of mean P:G ratios can be tested for significance by t-test. In Figures 3b and 3c, we compared the percentage of sample sites that exhibited increased or decreased P:G ratios following treatment, and tested for statistical significance using a two-sample z-test for proportions(Z Score Calculator for 2 Poulation Proportions, n.d.). This allowed us to determine the typical effect the non-surgical periodontal therapy had on the growth of the two target species. The majority of sample sites exhibited decreased P:G ratios for both organisms, indicating that the therapy made the periodontal pockets significantly less supportive of their growth.

If Support Protocol 1 has been completed, other relationships can be discerned, by discriminating between non-growing and actively-growing cells. The establishment of a threshold P:G ratio (Figure 2), above which cells can be assumed to be actively growing, allows for binomial analyses. For example, to gauge the effectiveness of therapy, the percent of periodontal pocket samples harboring actively growing bacteria (P:G above the threshold determined in Figure 2) were compared before and after non-surgical periodontal treatment (Figure 3d). Statistical significance in Figure 3d was determined by logistic regression (Logistic Regression in R Tutorial, 2018). This analysis showed significantly reduced growth activity of P. gingivalis, but not F. alocis, after therapy. Interestingly, retrospective studies have found P. gingivalis to be more responsive than F. alocis to this type of therapy (Bizzarro et al., 2016; Jentsch et al., 2013).

The example analyses discussed in this section should not be taken to be complete. Additional analyses are possible, depending on study design, clinical parameters, and treatments.

TIME CONSIDERATIONS

Many of the steps described in these protocols have varied time requirements depending on the specific experimental attributes. Estimates for the time required to process 6-30 samples are provided in Table 2.

Table 2:

Estimated timeframes for completion of each protocol step.

Protocol	Stage	Objective	Active time	Total time
Basic Protocol 1	Preparations	Primer/probe design	2 hrs	2 hrs
		Primer/probe validation	30 min	2 hrs
		pre-rRNA/growth characterization	see Support Protocol 1
	Sample prep	Sample collection	1 hr	1 hr
		RNA/DNA extraction	0.5-2 hours	0.5-2 hours
		Nucleic acid quality assessment	30 min	30 min
	Analysis	RT-qPCR (1-step)	30 min	2 hrs
		or RT-qPCR (2-step)	60 min	3.5 hrs
		Data analysis	15-60 min	15-60 min
	subtotal (minus one-time prep)		4.25-6.5 hrs	9.75-12 hrs
Support Protocol 1	Microbiology	Initial bacterial culture	15 min	2-7 days
		Drain pre-rRNA pools	15 min	3 hrs-5 days
		Nutritional stimulation (regrowth)	15 min	30 min-4 hr
	Sample prep	RNA/DNA extraction	0.5-2 hours	0.5-2 hours
		Nucleic acid quality assessment	30 min	30 min
	Analysis	RT-qPCR (1-step)	30 min	2 hrs
		or RT-qPCR (2-step)	60 min	3.5 hrs
		Data analysis	15-60 min	15-60 min
	subtotal		3.5-5.75 hrs	2-7 days

DATA AVAILABILITY STATEMENT

The data, tools and material (or their source) that support the protocols are available from the corresponding author upon reasonable request.