This protocol describes a 96-well disruption of individual bacterially colonized Caenorhabditis elegans following cold paralysis and surface bleaching to remove external bacteria. The resulting suspension is plated on agar plates to allow accurate, medium-throughput quantification of bacterial load in large numbers of individual worms.
The nematode Caenorhabditis elegans is a model system for host-microbe and host-microbiome interactions. Many studies to date use batch digests rather than individual worm samples to quantify bacterial load in this organism. Here it is argued that the large inter-individual variability seen in bacterial colonization of the C. elegans intestine is informative, and that batch digest methods discard information that is important for accurate comparison across conditions. As describing the variation inherent to these samples requires large numbers of individuals, a convenient 96-well plate protocol for disruption and colony plating of individual worms is established.
Heterogeneity in host-microbe associations is observed ubiquitously, and variation between individuals is increasingly recognized as a contributing factor in population-level processes from competition and coexistence1 to disease transmission2,3,4. In C. elegans, "hidden heterogeneity" within isogenic populations has been observed repeatedly, with sub-populations of individuals showing distinct phenotypes in heat shock response5,6, ageing, and lifespan7,8,9,10,11, and many other aspects of physiology and development12. Most analyses that attempt to identify sub-population structure provide evidence for two sub-populations in experimental populations of isogenic, synchronized worms5,7,8, though other data suggest the possibility of within-population distributions of traits rather than distinct groups7,12,13. Of relevance here, substantial heterogeneity in intestinal populations is observed even within isogenic populations of worms colonized from a shared source of microbes13,14,15,16, and this heterogeneity can be concealed by the batch digest measurements that are widely used17,18,19,20 for bacterial quantification in the worm.
This work presents data suggesting a need for greater reliance on single-worm measurements in host-microbe association, as well as protocols for increasing accuracy and throughput in single-worm disruption. These protocols are designed to facilitate mechanical disruption of large numbers of individual C. elegans for quantification of viable bacterial load, while providing better repeatability and lower effort per sample than pestle-based disruption of individual worms. A recommended gut-purging step, where worms are permitted to feed on heat-killed E. coli prior to the preparation for disruption, is included to minimize contributions from recently ingested and other transient (non-adhered) bacteria. These protocols include a cold-paralysis method for cleaning the cuticle with a low-concentration surface bleach treatment; surface bleaching can be used as a preparatory step in single-worm disruption or as a method for preparing live, externally germ-free worms. This surface-bleaching method is sufficient to remove a wide range of external microbes, and cold treatment provides an alternative to conventional levamisole-based paralysis; while levamisole will be preferred for cold-sensitive experiments, cold paralysis minimizes contributions to hazardous waste streams and allows rapid resumption of normal activity. While the full protocol describes a laboratory experiment where worms are colonized with known bacteria, the procedures for cleaning worms and single-worm disruption can readily be applied to worms isolated from wild samples or colonized in microcosm experiments. The protocols described here produces live bacteria extracted from the worm intestine, suitable for plating and quantification of colony forming units (CFUs) in individual worms; for sequencing-based intestinal community analysis, subsequent cell lysis and nucleic acid extraction steps should be added to these protocols.
Worms used in these experiments were obtained from the Caenorhabditis Genetic Center, which is funded by NIH Office of Research Infrastructure Programs (P40 OD010440). Bristol N2 is the wild-type. DAF-2/IGF mutants daf-16(mu86) I (CGC CF1038) and daf-2(e1370) III (CGC CB1370) are used to illustrate differences in intestinal bacterial load.
HT115(DE3) E. coli carrying the pos-1 RNAi vector is from the Ahringer library21. The MYb collection of C. elegans native gut bacteria22 was obtained from the Schulenburg lab. Salmonella enterica LT2 (ATCC 700720) attB:GFP-KmR is from this lab23. Pseudomonas mosselii was isolated in this lab. Staphylococcus aureus MSSA Newman pTRKH3-mGFP was obtained from the LaRock lab at Emory University.
All worm buffers and media are prepared according to previously published literature24 with minor modifications (see Supplementary File 1).
1. Preparation of synchronized sterile C. elegans
NOTE: In this section, step-by-step procedures are described for generating a synchronized population of reproductively sterile adult worms. Feeding on pos-1 RNAi plates is used here to prevent production of progeny because this interference is embryonic lethal; L1 larvae raised to adulthood on pos-1 RNAi develop into egg-laying hermaphrodites, but these eggs are inviable25. The RNAi feeding protocol is as in the "Reverse genetics" chapter of Wormbook26.
2. Feeding worms on live bacteria in liquid culture
NOTE: This protocol is used to colonize worms with laboratory-grown bacteria in well-mixed conditions in liquid culture (Supplementary Figure 1). Worms can be colonized with individual isolates from pure culture (e.g., pathogens such as Enterococcus faecium28,29) or mixtures of isolates (e.g., minimal microbiome communities14).
3. Mechanical disruption of individual worms in a 96-well format
NOTE: This section describes a 96-well plate format protocol for mechanical disruption of individual bacterially colonized C. elegans. The first steps in the protocol (3.1-3.8) describe a method for purging non-adhered bacteria from the worm intestine and cleaning the exterior of the worms using cold paralysis and surface-bleaching. These steps will produce clean, live adult worms that can be mechanically disrupted for quantification of bacterial contents (3.8-end) or used for further experiments (Supplementary Figure 1). This protocol can be adapted to quantify bacteria in worms colonized in liquid culture (Section 2), on agar plates, or from natural or microcosm soil.
4. Cleaning silicon carbide grit for re-use
NOTE: This procedure is used to clean and sterilize the grinding material, silicon carbide grit, for re-use after experiments. This protocol should be followed in its entirety before first use, as silicon carbide grit is an industrial product and does not come pre-sterilized. Si-carbide grit (3.2 g/cc) is a dense, rough-edged material that works efficiently to disrupt tough samples. However, the particles can wear down over repeated use and should be replaced when wear becomes apparent. Fortunately, the material is inexpensive, and the sizes typically sold (~1 lb) are sufficient for many experiments.
Bleach sterilization of live worms
Surface-bleached worms are effectively free of external bacteria until motility returns and excretion resumes. Under the conditions used here, rapid extinction of bacteria in buffer is observed (Figure 1A–C, Supplementary Figure 2, Video 1) without disturbing the gut-associated bacteria in cold-paralyzed worms (Figure 1D–F, Video 2). These data indicate that surface bleaching can be used effectively to sanitize worms externally without compromising the intestinal contents (comparisons of surface-bleached vs. no-bleach worm-associated CFU counts are non-significant, Wilcoxon rank-sum test p > 0.05).
Variations on multi-sample mechanical disruption
The 96-well technique for mechanical disruption of worms is robust to the specific materials used, and practical considerations dictate the choice of grinding material. Similar to a previous report33, manual disruption (Figure 2A) resulted in more heterogeneity than the standard 96-well protocol (silicon carbide grit, Figure 2B) (var(log10CFU) = 0.499) across all buffer conditions, as compared with 0.229 for Si-carbide, 0.243 for large glass beads (Figure 2C), and 0.227 for small glass beads (Figure 2D). Nonetheless, most differences in CFU/worm distributions were not significant (Kruskal-Wallace, p = 0.017 with df = 3; significant post-hoc Wilcoxon tests for large beads vs. small beads, p = 0.021, and large beads vs. silicon carbide grit, p = 0.02). The use of Triton X-100 as a surfactant was not associated with any significant difference in yield when considered as an individual factor (Kruskal-Wallace, p = 0.94, df = 3), although there is an apparent increase in yield in no-Triton vs. Triton-containing samples when large beads (2.7 mm) were used (Figure 2C), possibly attributable to the excessive "foaming" observed in these wells when Triton was present. These results indicate that large glass beads, while ideal for use in homogenization tubes33, are not suitable for the 96-well technique. While small glass beads produced reasonable results (Figure 2D), they consistently clogged 200 µL pipette tips during mixing and plating. The standard material in this assay, silicon carbide grit, is inexpensive, too large to clog standard tips, and like glass beads can be washed and reused after autoclaving. The grit does release a small amount of "dust" into the buffer, which does not interfere with plating but needs to be filtered off if the products of disruption are to be used for flow cytometry.
Heterogeneity in bacterial colonization in adult worms
Successful disruption of individual worms reveals heterogeneity in bacterial colonization. Individuals from isogenic synchronized populations of worms, colonized at the same time on the same pool of bacteria, consistently show 100-fold or greater range in intestinal bacterial load. This is observed for different bacterial colonists (Figure 3A) and during colonization on multi-species bacterial communities (Figure 3B). This heterogeneity is also evident in individual-worm measurements of fluorescence when worms are colonized with bacteria expressing a fluorescent protein (GFP) (Figure 3C–D). The properties of the host play a role in shaping this heterogeneity, as can be seen by comparing colonization of wild-type Bristol N2 worms to colonization by the same bacteria in DAF-2/IGF mutants; this daf-16 mutant supports larger populations of many bacteria as compared with N2, while daf-2 is resistant to colonization by a range of bacteria36 (Figure 3B,D). This heterogeneity is characteristic, showing variation across different combinations of host and colonist(s), while retaining a consistent structure over different runs of the same experiment (Figure 3E–F).
Importance of individual heterogeneity for accurate comparison of groups
The importance of individual heterogeneity can be easily seen by considering how batch digests could alter the distributions of data. Colonization by native microbiome bacteria MYb53 (Rhodococcus erythropolis) and MYb120 (Chryseobacteria spp.) (Figure 3A, 4A) in N2 adults are used as examples. The individual worm data are clearly similar in distribution (two-tailed t-test, p = 0.9, Wilcoxon rank sum, p = 0.59). When resampling these data to simulate the effects of batch digests, the batch extrapolated CFU/worm pulls toward the upper quantiles of the data due to the positive skew in these distributions (mean > median). As batching effectively averages over the individuals within a batch, batch-extrapolated CFU/worm will center around the arithmetic mean of the individual data, with decreasing distance to this mean as batches become large according to the central limit theorem (Figure 4B–D). Accordingly, signal from biological variation is quickly lost; batch-inferred CFU/worm measurements converge toward the average, which is not a representative metric of these log-scale-distributed data. Differences in inferred colonization by MYb53 vs. MYb120 quickly become significant in simulated batch digests (t-test batch 5, p = 0.049; batch 10, p = 2.27e-4; batch 20, p = 1.19e-15; Wilcoxon rank sum test batch 5, p = 2.27e-4; batch 10, p = 2.70e-06; batch 20, p = 1.80e-09) as the original signal is obscured.
Effects of individual heterogeneity on microbial transmission
As individual worms show substantial heterogeneity in bacterial colonization, it is reasonable to ask whether this heterogeneity has downstream effects. For example, it is reasonable to expect that transmission might be a function of intestinal bacterial load. By transferring individual surface-bleached worms to a clean environment, it is possible to observe inoculation of the environment with excreted live bacteria. In these experiments, surface-bleached pre-colonized adults, carrying generally substantial populations (103-105 CFU/worm, Figure 5) of commensal Ochrobactrum MYb14-GFP or pathogenic S. aureus-GFP, were allowed to roam on heat-killed OP50 lawns on NGM agar for 1.5 h. When these worms are re-harvested from excretion plates and disrupted for bacterial quantification, there is no significant relationship between bacterial load and excretion rate of live bacteria (Pearson correlations between log-transformed colonies/hr and CFU/worm: MYb14 rho = 0.19, p = 0.45; S. aureus rho = 0.02, p = 0.9) (Figure 5). Nor is there a significant relationship between the presence/absence of colonies on a plate and intestinal bacterial load (binomial logistic regression with log-transformed CFU/worm as factor: p = 0.15 with df = 53). A substantial fraction of plates remained free of new growth (9/18 plates for MYb14, 10/36 plates for S. aureus), indicating low overall excretion rates.
When worms are allowed to excrete onto agar plates, the actual number of live excreted bacteria per worm is confounded by "farming", where worms pass through colonies and create trails of new growth (Figure 6)37. A plate with n colonies represents at least one, and at most n, events where live colony-forming bacteria were excreted. From this observation, it is not possible to know how many excretion events in (1,n) actually occurred, nor is it possible to know how many bacteria were excreted in each event. It is therefore not possible to precisely estimate excretion rates of live bacteria from the gut using these data. However, it is possible to infer some bounds. Although the number of colonies per plate is not very informative, presence/absence data can be used for rough inference of excretion rates. For simplicity, if it is assumed that excretion rate of live bacteria is not a function of bacterial load and that excretion is a Poisson process, there is a ~50% chance of observing at least nine events in 18 trials when λ ≈ 0.33 worm-1 hr-1 in MYb14. For S. aureus, similar plausible rates of λ ≈ 0.2 worm-1 hr-1 are obtained. While these rough calculations suggest low rates of excretion of live bacteria, more precise quantification of this process over larger numbers of individual worms will be necessary to obtain reliable estimates.
Data Availability:
Data shown here are available on Dryad (https://doi.org/10.5061/dryad.7wm37pvw2).
Figure 1. Low-concentration surface-bleaching treatment rapidly kills bacteria in buffer but does not disturb intestinal communities in cold-paralyzed worms. (A–C) Bacterial CFU/mL in M9 worm buffer during surface bleaching at three different concentrations (1:1000, 1:2000, 1:5000 v/v; unbleached control for comparison), targeting (A) S. aureus Newman, (B) S. enterica LT2, or (C) E. coli OP50. Samples were taken at indicated time points up to 20 min post-exposure and washed twice with sterile buffer to prevent bleach from killing colonies on plates. Data for the 1:1000 condition are offset slightly so that these data are visible on the plot. (D–F) Intestinal bacteria in individual N2 worms (n = 24 worms per experiment, two or three independent runs on separate days). All comparisons of surface-bleached and no-bleach worm-associated CFU counts are non-significant (Wilcoxon rank-sum test p > 0.05). Grey horizontal lines represent threshold of detection, defined as the density (40 CFU/worm) at which probability of observing at least one colony is ~60% when plating 10 µL aliquots from 200 µL volumes. Please click here to view a larger version of this figure.
Figure 2. The 96-well disruption protocol produces consistent results and is robust to the choice of materials. N2 adult worms colonized with a single bacterial species for 48 h (P. mosselii) were surface bleached and permeabilized according to standard protocols, then individual worms (n = 24 per condition) were mechanically disrupted for CFU plating using (A) manual disruption in individual 0.5 mL tubes, using a motorized pestle or (B–D) variations on the 96-well disruption protocol described in detail in the Protocol. Disruption was carried out in M9 worm buffer containing varying concentrations of Triton X-100 (x-axis, 0-0.1%, v/v) and one of (B) 36-grit silicon carbide, (C) small (425-600 µm) glass beads, or (D) large (2.7 mm) glass beads. For all plots, data shown are log10(CFU/worm), and each point is one individual worm. Please click here to view a larger version of this figure.
Figure 3. Heterogeneous bacterial colonization of the C. elegans intestine. (A) Single-species colonization of N2 adult hermaphrodites prepared as in Methods. Bacteria are four species from the MYb native worm microbiome collection (Dirksen et al. 2016) (n = 24 worms, one experiment each) and two pathogens, Staphylococcus aureus MSSA Newman (SA) and Salmonella enterica LT2 (SE) (n = 96-144 worms over two/three independent experiments). Colonization by native microbiome species was assessed after a 48 h incubation at 25°C in liquid S medium + 108 CFU/mL bacteria; colonization by pathogens was assessed after incubation on lawns on NGM worm agar for 24 (SA) or 48 (SE) h at 25°C. (At 48 h, worms on S. aureus have mostly died.) (B) Total CFU/worm in N2, daf-16(mu86), and daf-2(e1370) adults colonized for 4 days in liquid media on an eight-species minimal native microbiome (data from Taylor and Vega, 2021)14. (C–D) Green fluorescence in individual worms colonized with GFP-expressing bacteria, observed by large object flow cytometry. In (C), synchronized populations of N2 adults were colonized with OP50 (non-fluorescent, n = 1908 individual adult worms), S. aureus (GFP, n = 968), or S. enterica (GFP, n = 1153) as described in (A); the OP50 control indicates typical levels of green-channel autofluorescence in day-3 adult N2 worms. In (D), synchronized populations of N2 (n = 1165), daf-16(mu86) (n = 1180), and daf-2(e1370) (n = 2267) adults were colonized with commensal Ochrobactrum MYb14-GFP for 2 days on plates as described in (A). (E–F) Day-to-day variation in colonization by S. aureus (E) and S. enterica (F) (same data as in panel A and Figure 1, n = 48 worms per experiment). The x-axis indicates the day of sampling. Grey horizontal lines represent threshold of detection, defined as the density at which probability of observing at least one colony is ~60% (40 CFU/worm for single-species colonization and four CFU/worm for multi-species colonization, due to different plating volumes of 10 µL and 100 µL respectively out of 200 µL). Please click here to view a larger version of this figure.
Figure 4. Batching erases biological variation in skewed log-scale data. CFU/worm data from Figure 3 were resampled with replacement to create n = 25 replicate sets of simulated data for each batch size, where size is the number of individual worms per batch. CFU/worm is the total CFU in each simulated batch divided over the number of worms per batch. In the raw data (panel A), average CFU/worm for MYb53 is 4450.8 (103.6), and for MYb120, 1398.3 (103.1); the batch-inferred numbers converge to these values as batch size increases (B, five worms/batch; C, 10 worms/batch; D, 20 worms/batch), consistent with expectations from central limit theorem. Please click here to view a larger version of this figure.
Figure 5. Excretion of live bacteria is poorly correlated with CFU load in the intestine of individual worms. Here, N2 adults were colonized by feeding for 1 or 2 days respectively on lawns of S. aureus-GFP or MYb14-GFP. Worms with detectable GFP fluorescence (total GFP > 1.8 logs on large object flow cytometer) were sorted from the bulk population, surface bleached as described in Methods, and transferred individually to NGM + heat-killed OP50 plates. Pearson correlations between log-transformed colonies/h and CFU/worm are non-significant (MYb14 rho = 0.19, p = 0.45; S. aureus rho = 0.02, p = 0.9). Please click here to view a larger version of this figure.
Figure 6. Bacterial "farming" obscures the number of excretion events on agar plates. Here are two plates with MYb14-GFP colonies from worm excreta. The first plate (A) has clear evidence of "farming" along worm paths and appears to represent at least two separate excretion events based on differences in GFP expression (visible as yellowish pigmentation) across colonies. While the second plate (B) is more ambiguous, farming cannot be ruled out based on the positions of the colonies. In these experiments, N2 adult worms were pre-colonized for 48 h by feeding on agar plates containing lawns of MYb14-GFP. After colonization, worms were prepared and surface bleached according to Methods, then transferred in 5 µL aliquots of M9 worm buffer + 0.1% Triton X-100 to 6 cm NGM + heat-killed OP50 plates (prepared by allowing 50 µL spots of 5x concentrated heat-killed OP50 to dry on the surface). Worms were permitted to roam for 1.5 h at 25°C, then picked from plates and disrupted for CFU/worm plating (manual disruption in 20 µL buffer in individual 0.5 mL tubes, using a motorized pestle). Plates were incubated at 25°C for 2 days before counting. Please click here to view a larger version of this figure.
Video 1. Visualization of N2 worms colonized with GFP fluorescent S. aureus without surface bleaching. A small number of fluorescent cells on the cuticle move into and out of focus as the image passes through the body of the worm, and spatially heterogeneous colonization of the gut becomes visible as the field of view moves from the body surface into the intestine. Z-stack image was taken at 20x magnification on an inverted fluorescent microscope. Bright-field and GFP filtered fluorescent images were overlaid, and images across the Z-stack stitched together, using the vendor software. Image is from the same slide as in Supplementary Figure 2A. Please click here to download this Video.
Video 2. Visualization of a N2 worm colonized with GFP fluorescent S. aureus with surface bleaching (1:1000 v/v for 20 min). Spatially-heterogeneous colonization by fluorescent bacteria is visible in the intestine of this individual, and bacteria have infiltrated the body tissues, indicating advanced infection. No bacteria are visible on the cuticle. Z-stack image was taken at 20x magnification on an inverted fluorescent microscope. Bright-field and GFP filtered fluorescent images were overlaid, and images across the Z-stack stitched together, using the vendor software. Image is from the same slide as in Supplementary Figure 2B. Please click here to download this Video.
Supplementary Figure 1. Overview of the Protocol. Here, synchronized adult worms are mono-colonized with red bacteria, surface-bleached, and permeabilized before mechanical disruption of individual worms in a 96-well format. Bacteria released from the intestine are dilution plated in 10x series for CFU/worm quantification; plates shown are typical for observed heterogeneity. Please click here to download this File.
Supplementary Figure 2. Visualization of N2 worms colonized with GFP fluorescent S. aureus with and without surface bleaching (1:1000 v/v for 20 min). (A) In the unbleached sample, external bacteria are visible at low magnification as areas of green fluorescence not associated with worms or worm body fragments. (B) In the surface-bleached sample, GFP fluorescence is restricted to the interior of worm bodies (one worm body fragment is visible mid-image). All images were taken at 4x magnification on an inverted fluorescent microscope. Bright-field and GFP filtered fluorescent images were overlaid, and images from adjacent fields of view stitched together, using the vendor software. Please click here to download this File.
Supplementary File 1: Buffer and solution recipes. Please click here to download this File.
Here data are presented on the advantages of single-worm quantification of bacterial load in C. elegans, along with a 96-well disruption protocol to allow the rapid and consistent acquisition of large data sets of this type. As compared with existing methods33, these protocols allow higher-throughput measurement of intestinal microbial communities in the worm.
This approach has plating as a rate-limiting step and is not truly "high-throughput". Large-object flow cytometry (Figure 3C,D) is a useful high-throughput method for quantifying fluorescently labeled bacteria in individual worms16, although the number of simultaneous fluorophores is a limitation in multi-species communities. Linking multi-well plate disruption with community sequencing is another way to increase throughput; however, the 96-well disruption procedure described here was optimized specifically to leave bacterial cells intact. Sequencing-based analysis, where thorough lysis of cells is desirable, will require addition of a nucleic acid extraction step or modification of the beating protocol (Protocol 3.10-3.11) to extract cell contents instead of live bacteria. Protocols for single-worm disruption and extraction of nucleic acids have been published elsewhere38,39.
Bacterial total abundance in the worm intestine is heterogeneous, and the data shown here suggest that batch-based measurement can produce erroneous results in comparisons between groups. However, other measures of bacterial communities in the worm may be less sensitive to the effects of batching. Of note, relative abundances in worm-associated communities seem to vary very little if at all with total intestinal population size, regardless of whether interactions among microbes are neutral40 or not14. It is plausible that, compared to count data, relative abundance measures will be less susceptible to the false-positive rate issue described. Sequencing-based community analysis, which generates relative abundance data for community composition, may therefore not require measurement of single worms. Further investigation is needed on this point.
Here, we use cold treatment to paralyze worms for surface bleaching. Other work has found that worms resume normal activity rapidly (<15 min) if time on ice is kept under 30 min, allowing immediate use in further assays, in contrast with chemical paralysis agents which can require extended periods before full recovery34. If worms are to be disrupted immediately for bacterial quantification, this feature is dispensable, and the main advantage of chilling vs. chemical paralysis is avoiding the need for a controlled waste stream. Extended cold treatment should be used with caution when investigating stress responses, particularly if there is a known connection to temperature. The cold paralysis protocols described here entail shorter acute cold exposure than used in experiments for cold stress (20-30 min vs 2+ h at 2-4°C)41,42,43, and a 1 h cold shock produces no apparent phenotype in wild-type worms43. Short-term (90 min) incubation at 4 °C induces changes in cold-stress gene expression (measured by expression of a TMEM-135::GFP reporter), but expression returns to unstressed levels within minutes once worms are returned to room temperature34. However, the effects on stress-sensitive worm genotypes may be more severe than in wild-type. This procedure should be validated under the experimental conditions to be used.
The surface bleaching protocol described here can be used as a way to limit or eliminate passaging of external microbes in experiments. This method has additionally been used to clear fungal contaminants by surface bleaching and transferring only L1/L2 larvae to fresh plates (transfer of surface-bleached adults resulted in failure to clear the contaminant, presumably due to carriage in the intestines of the larger animals). It is critically important to ensure that bleach concentration does not exceed 1:1000 v/v, as damage to the worms and mortality will result. This procedure may be useful in experimental host-microbe evolution and host-pathogen interactions. For example, the low excretion rate of live bacteria observed here can help to explain the highly variable rates observed for bacterial transmission from hermaphrodites to offspring15. The lack of correlation between intestinal bacterial load and excretion rate observed here is interesting but requires further investigation; a larger number of data points across a range of conditions will be needed to determine where (or whether) this observation will hold.
It may not always be necessary to clean worms to the extent provided by surface bleaching. Multiple washes in sterile buffer are likely sufficient when worms are internally colonized with a single microbe if the minimum expected CFU/worm is much higher (10-100-fold) than the concentration of bacteria in buffer supernatant, as this carryover will minimally affect counts (see Figure 1). Additionally, if the microbe(s) of interest primarily colonize(s) the cuticle, surface bleaching should clearly be avoided. Thorough cleaning is more important to ensure accuracy when dealing with mixed microbial communities (to ensure that all colonies/reads in a sample are from worm-associated bacteria and not from the environment), when bacteria adhered to the cuticle interfere with reading the internalized population, when expected minimum CFU/worm is low, etc.
The authors have nothing to disclose.
The authors would like to acknowledge H. Schulenberg and C. LaRock for their generous sharing of bacterial strains used in these experiments. This work was supported by funding from Emory University and NSF (PHY2014173).
96-well flat-bottom polypropylene plates, 300 uL | Evergreen Labware | 290-8350-03F | |
96-well plate sealing mat, silicon, square wells (AxyMat) | Axygen | AM-2ML-SQ | |
96-well plates, 2 mL, square wells | Axygen | P-2ML-SQ-C-S | |
96-well polypropylene plate lids | Evergreen Labware | 290-8020-03L | |
Agar | Fisher Scientific | 443570050 | |
Bead mill adapter set for 96-well plates | QIAGEN | 119900 | Adapter plates for use with two 96-well plates on the TissueLyser II |
Bead mill tissue homogenizer (TissueLyser II) | QIAGEN | 85300 | Mechanical homogenizer for medium to high-throughput sample disruption |
BioSorter | Union Biometrica | By quotation | Large object sorter equipped with a 250 micron focus for C. elegans |
Bleach, commercial, 8.25% sodium hypochlorite | Clorox | ||
Breathe-Easy 96-well gas permeable sealing membrane | Diversified Biotech | BEM-1 | Multiwell plate gas permeable polyurethane membranes. Thin sealing film is permeable to O2, CO2, and water vapors and is UV transparent down to 300 nm. Sterile, 100/box. |
Calcium chloride dihydrate | Fisher Scientific | AC423525000 | |
Cholesterol | VWR | AAA11470-30 | |
Citric acid monohydrate | Fisher Scientific | AC124910010 | |
Copper (II) sulfate pentahydrate | Fisher Scientific | AC197722500 | |
Corning 6765 LSE Mini Microcentrifuge | Corning | COR-6765 | |
Disodium EDTA | Fisher Scientific | 409971000 | |
DL 1,4 Dithiothreitol, 99+%, for mol biology, DNAse, RNAse and Protease free, ACROS Organics | Fisher Scientific | 327190010 | |
Eppendorf 1.5 mL microcentrifuge tubes, natural | Eppendorf | ||
Eppendorf 5424R microcentrifuge | Eppendorf | 5406000640 | 24-place refrigerated benchtop microcentrifuge |
Eppendorf 5810R centrifuge with rotor S-4-104 | Eppendorf | 22627040 | 3L benchtop centrifuge with adaptors for 15-50 mL tubes and plates |
Eppendorf plate bucket (x2), for Rotor S-4-104 | Eppendorf | 22638930 | |
Ethanol 100% | Fisher Scientific | BP2818500 | |
Glass beads, 2.7 mm | Life Science Products | LS-79127 | |
Glass beads, acid-washed, 425-600 µm | Sigma | G877-500G | |
Glass plating beads | VWR | 76005-124 | |
Hydrochloric acid | VWR | BDH7204-1 | |
Iron (II) sulfate heptahydrate | Fisher Scientific | 423731000 | |
Kimble Kontes pellet pestle motor | DWK Life Sciences | 749540-0000 | |
Kimble Kontes polypropylene pellet pestles and microtubes, 0.5 mL | DWK Life Sciences | 749520-0590 | |
Leica DMi8 motorized inverted microscope with motorized stage | Leica | 11889113 | |
Leica LAS X Premium software | Leica | 11640687 | |
Magnesium sulfate heptahydrate | Fisher Scientific | AC124900010 | |
Manganese(II) chloride tetrahydrate | VWR | 470301-706 | |
PARAFILM M flexible laboratory sealing film | Amcor | PM996 | |
Peptone | Fisher Scientific | BP1420-500 | |
Petri dishes, round, 10 cm | VWR | 25384-094 | |
Petri dishes, round, 6 cm | VWR | 25384-092 | |
Petri dishes, square, 10 x 10 cm | VWR | 10799-140 | |
Phospho-buffered saline (1X PBS) | Gold Bio | P-271-200 | |
Polypropylene autoclave tray, shallow | Fisher Scientific | 13-361-10 | |
Potassium hydroxide | Fisher Scientific | AC134062500 | |
Potassium phosphate dibasic | Fisher Scientific | BP363-1 | |
Potassium phosphate monobasic | Fisher Scientific | BP362-1 | |
R 4.1.3/RStudio 2022.02.0 build 443 | R Foundation | n/a | |
Scoop-type laboratory spatula, metal | VWR | 470149-438 | |
Silicon carbide 36 grit | MJR Tumblers | n/a | Black extra coarse silicon carbide grit. Available in 0.5-5 lb sizes from this vendor. |
Sodium dodecyl sulfate | Fisher Scientific | BP166-100 | |
Sodium hydroxide | VWR | BDH7247-1 | |
Sodium phosphate dibasic anhydrous | Fisher Scientific | BP332-500 | |
Sodum chloride | Fisher Scientific | BP358-1 | |
Sucrose | Fisher Scientific | AC419760010 | |
Tri-potassium citrate monohydrate | Fisher Scientific | AC611755000 | |
Triton X-100 | Fisher Scientific | BP151-100 | |
Zinc sulfate heptahydrate | Fisher Scientific | AC205982500 |