To study the mutualism between Xenorhabdus bacteria and Steinernema nematodes, methods were developed to monitor bacterial presence and location within nematodes. The experimental approach, which can be applied to other systems, entails engineering bacteria to express the green fluorescent protein and visualizing, using fluorescence microscopy bacteria within the transparent nematode.
Symbioses, the living together of two or more organisms, are widespread throughout all kingdoms of life. As two of the most ubiquitous organisms on earth, nematodes and bacteria form a wide array of symbiotic associations that range from beneficial to pathogenic 1-3. One such association is the mutually beneficial relationship between Xenorhabdus bacteria and Steinernema nematodes, which has emerged as a model system of symbiosis 4. Steinernema nematodes are entomopathogenic, using their bacterial symbiont to kill insects 5. For transmission between insect hosts, the bacteria colonize the intestine of the nematode’s infective juvenile stage 6-8. Recently, several other nematode species have been shown to utilize bacteria to kill insects 9-13, and investigations have begun examining the interactions between the nematodes and bacteria in these systems 9.
We describe a method for visualization of a bacterial symbiont within or on a nematode host, taking advantage of the optical transparency of nematodes when viewed by microscopy. The bacteria are engineered to express a fluorescent protein, allowing their visualization by fluorescence microscopy. Many plasmids are available that carry genes encoding proteins that fluoresce at different wavelengths (i.e. green or red), and conjugation of plasmids from a donor Escherichia coli strain into a recipient bacterial symbiont is successful for a broad range of bacteria. The methods described were developed to investigate the association between Steinernema carpocapsae and Xenorhabdus nematophila 14. Similar methods have been used to investigate other nematode-bacterium associations 9,15-18and the approach therefore is generally applicable.
The method allows characterization of bacterial presence and localization within nematodes at different stages of development, providing insights into the nature of the association and the process of colonization 14,16,19. Microscopic analysis reveals both colonization frequency within a population and localization of bacteria to host tissues 14,16,19-21. This is an advantage over other methods of monitoring bacteria within nematode populations, such as sonication 22or grinding 23, which can provide average levels of colonization, but may not, for example, discriminate populations with a high frequency of low symbiont loads from populations with a low frequency of high symbiont loads. Discriminating the frequency and load of colonizing bacteria can be especially important when screening or characterizing bacterial mutants for colonization phenotypes 21,24. Indeed, fluorescence microscopy has been used in high throughput screening of bacterial mutants for defects in colonization 17,18, and is less laborious than other methods, including sonication 22,25-27and individual nematode dissection 28,29.
1. Construction of a Fluorescent Bacterial Strain via Conjugation
2. Production of Axenic Nematode Eggs
3. Co-cultivation Assay with Fluorescent Bacteria
4. Collection of Early Life Stages for Screening
5. Screening Nematodes for Bacterial Association by Microscopy
6. Representative Results
Example microscope images of Steinernema nematodes associated with Xenorhabdus bacteria are shown in Figure 3. To create the composite image seen in Figure 3A, a phase contrast image was overlaid with a fluorescent image. The arrow in Figure 3A indicates the bacteria present within the infective juvenile nematode (bar = 100 μm). Figure 3B was constructed in a similar manner and depicts a juvenile nematode with green fluorescent protein labeled bacteria (green rods) localized throughout the nematode intestinal lumen (bar = 20 μm). A population of the nematodes from two media were counted and scored for colonization by the bacterial symbiont (Table 1). For robust statistics, it is best to count at least 100 nematodes per sample with at least 30 falling in each category. As seen in Table 1, these nematodes are colonized at a level of approximately 14.6% when grown on lipid agar and 68.6% when grown on liver kidney agar. Other nematode and bacterial species have been shown to have different levels of colonization. For example X. nematophila colonizes 99% of S. carpocapsae infective juveniles (Martens 2003), and P. luminescens colonizes 26% of H. bacteriophora infective juveniles (Ciche 2003).
Figure 1. Schematic outline of the method. A. The bacterium is engineered to express a fluorescent protein. B. Nematode eggs are isolated from adult nematodes to produce sterile nematodes. C. The sterile nematodes are co-cultivated with the fluorescent bacteria. D. The resulting life stages are viewed under a microscope to evaluate bacterial presence within the nematode.
Figure 2. Depiction of adult females containing eggs. A. The schematic shows the general appearance of Steinernema females. Inset: DIC image of a S. feltiae gravid female. The black arrow indicates the vulva. White arrows show visible eggs. Image is at 20X magnification, and the scale bar represents 100 μm. B. DIC image of a developed but unhatched S. feltiae nematode egg. Image is 40X magnification, and the scale bar represents 50 μm. C. DIC image of eggs isolated from S. feltiae nematodes under 10X magnification. Scale bar is 100 μm.
Figure 3.Example microscope images of nematode-bacterial association. A. S. puntauvense nematodes were associated with their bacterial symbiont, X. bovienii, expressing GFP. The image is a composite image produced by overlaying a phase contrast image with a fluorescent image from the same field of view. The arrow indicates the fluorescent bacterial symbiont within the nematode host. Scale bar represents 100 μm. B. This image shows a S. carpocapsae juvenile nematode with GFP-expressing X. nematophila localized within the nematode intestine. This image was constructed through overlaying a fluorescent image over a differential interference contrast image. The scale bar represents 20 μm.
Strain | Number of Nematodes With acteria | Total Nematodes Counted | Percent of Nematodes olonized |
S. puntauvense Lipid Agar | 30 | 205 | 14.60% |
S. puntauvense Liver Kidney Agar | 72 | 105 | 68.60% |
Table 1. Example scoring of a nematode population for bacterial presence. In this experiment, axenic S. puntauvense nematodes were grown with their GFP-expressing symbiont on different growth media (lipid agar and liver kidney agar 32) to test for colonization defects. A total of at least one hundred nematodes per sample were counted and scored for the presence of bacteria. For statistical power, three experimental replicates should be counted with at least 30 nematodes falling in each category.
Table 2. Fluorescent protein containing plasmids. A list of potential plasmids for insertion of a fluorescent protein into the bacterial symbiont is given listed by name of plasmid. Other information included are the fluorescent protein encoded, antibiotic cassette used for plasmid maintenance, other instructions for use, the source of the plasmid. The concentration noted in parentheses is the concentration of the antibiotic used for X. nematophila. Each of these plasmids has been used successfully in either Xenorhabdus or Photorhabdus. Additional information can be obtained from the noted citations. Depending on the bacterium being tested, some plasmids may not work based upon the fluorescent protein, antibiotic selection, insertion site, or origin of replication. The plasmids listed above contain different features that may enable use in the bacterium of interest. For example, mini-Tn7-KSGFP inserts into the attTn7 site of the chromosome, while pECM20 inserts into the X. nematophila chromosome by homologous recombination. Alternatively, the pPROBE plasmids are maintained extrachromosomally, and each pPROBE plasmid has the same backbone and fluorophore but have different selectable markers or origins of replication to enable their use in a variety of taxonomic or mutant backgrounds.
The protocol described here provides a method for the optical detection of bacteria within a nematode host (Figure 1). This method takes advantage of the optical transparency of nematodes and the ability to fluorescently label bacteria, enabling in vivo analysis of bacteria within the nematode host (Figure 3). Specifically, this approach identifies bacterial localization within its host. By counting a nematode population and scoring for bacterial presence, the frequency of bacterial colonization across the nematode population can be determined (Table 1). This method is one of many potential techniques that can be utilized for studying the interactions between nematode hosts and bacterial symbionts. Related methods have been described previously to isolate the bacterial symbiont, grow nematodes axenically, and manipulate both partners 25-27.
The protocol described here was developed in the Xenorhabdus nematophila-Steinernema carpocapsae model system 14, and similar approaches have been used in other entomopathogenic nematode-bacterial associations 9,15,17,18 . Several conditions must be met in order to apply this method to other nematode-bacterial systems. First, the bacterial symbiont must be able to be isolated from the host and grown independently in culture. Second, the symbiont must be able to take up DNA through transformation or conjugation in order to introduce the fluorescent protein. Third, for the best results, the nematode must have a stage that can be made axenic and reintroduced to bacteria. However, even if the nematode cannot be isolated from the bacteria it might still be possible to visualize the association: instead of adding axenic eggs to the lawn of fluorescent bacteria, add a conventionally raised life stage and proceed with the protocol as described. Any results will suffer from the caveat that bacteria that do not express GFP will compete with GFP-expressing bacteria for localization to specific nematode tissues, and because of this, colonization should not be measured by microscopic counting. However, unless the bacteria that do not express GFP drastically outcompete GFP-expressing bacteria, GFP-expressing bacteria should localize to nematode tissues in at least some animals in the population and suggest important tissue sites for bacterial colonization. A long-standing caveat of using fluorescent-protein expressing strains is that they may colonize a host with different efficiency than a strain that does not express fluorescent proteins (e.g. 12).
Steps described in this protocol may require optimization depending on the nematode and bacterial species that are being used. For conjugation of the bacterium some conditions that can be altered are the length and temperature of growth, the ratio of donor to recipient, and plasmid used. Examples of relevant plasmids are listed in Table 2. All of these plasmids have been successfully used in either Xenorhabdus or Photorhabdus bacteria in association with their nematode host 14,17,20,24,33-35 . To ensure stable maintenance of the fluorescent protein during nematode colonization it is best to use a plasmid that will insert into the chromosome of the target bacterium. Further, some bacteria may not take up these plasmids. For example, pECM20 has only been used in X. nemtaophila and very closely related bacterial strains 14 because this plasmid inserts into the chromosome using a homologous region between the plasmid and chromosome; the plasmid will not insert if the target bacterium lacks this region. To use this plasmid for other bacteria, the X. nematophila-specific region must be substituted with a genomic region from the target bacterium. Some plasmid-transformation schemes will also require certain alterations from Protocol 1. For example, mini-Tn7-KSGFP and pBK-miniTn7-ΩGm-DsRed require a helper plasmid containing transposition genes (tsnABCDE)33,35,36 to insert into the bacterial chromosome. Conjugation is most efficient with these Tn7-based plasmids, when 2-hour cultures (~OD6000.3-0.4) are mixed in equal ratios.
After successful conjugation it is important to utilize correct selective plates corresponding to the plasmid used. The selective plates should contain the antibiotic encoded on the plasmid to select for the presence of the plasmid in the recipient and a counter-selection against the donor and helper strains. For example, when conjugating pECM20 in X. nematophila the counter-selection used is ampicillin because X. nematophila is ampicillin resistant and the donor strain is ampicillin sensitive. If antibiotic counter-selection is unavailable in your system, a diaminopimelic acid (DAP) requiring donor may be used. To grow DAP-requiring strains, DAP is added to solid and liquid growth media during pre-conjugation and conjugation steps, and omitted during counter-selection stages. When DAP is absent the donor strain will not grow and is effectively counter-selected.
To ensure successful egg isolation it is essential to accurately check for egg production (Figure 2). At the time of isolation, female nematodes should be full of eggs and some eggs should be visible in the media (Figure 2A). If female nematodes are producing fertilized eggs at least some supernatant eggs will have visibly developed as unhatched nematodes (Figure 2B). For S. carpocapsae, the eggs will change from spherical to slightly oblong prior to developing nematodes and hatching. The timing or conditions of nematode growth may also need to be altered to maximize fertilized egg yield, including shortening or lengthening the period of time before egg harvesting, or using alternate media appropriate to the specific nematode species. At the end of the isolation, eggs should be easily visible within the buffer (Figure 2C). Parameters in the egg isolation protocol that may require optimization include bleach and KOH concentrations in the egg solution, incubation time in egg solution, or centrifugation speed. If the egg isolation produces eggs but no nematodes develop during the co-cultivation, it is possible that the isolated eggs are non-viable. To check for viability, leave the isolated eggs in buffer and wait for nematodes to hatch. The eggs should hatch within one or two days of isolation and then the juvenile nematodes can be used in the co-cultivation assay. If the eggs hatch but do not develop during the co-cultivation assay, it may be necessary to change the growth conditions or medium used for co-cultivation. We note that previous studies have isolated bacteria-free nematodes by soaking the nematodes in an antibiotic cocktail (e.g. 3,37). The approach we describe here is free of some of the caveats of antibiotic soaking, including complications with using bacteriostatic antibiotics, antibiotic-resistant contaminating bacteria, and antibiotic effects on the nematodes. Certain stages of nematodes may also be resistant to antibiotics and retain their bacterial symbionts3. Finally, we note that for microscopy, the concentration of levamisole necessary for nematode immobilization may vary with different nematode clades.
Once this method has been established in the system of interest, it is possible to alter the technique to derive more information. By looking at earlier time points in the nematode life cycle (Protocol 4), one may be able to identify how the bacterium and nematode initiate their association 14,16 . In addition, this approach can be used to conduct a high throughput mutant screens to identify bacterial factors involved in the symbiosis, using fluorescence to detect colonization 17,18 . Other methods such as signature tagged mutagenesis require the use of radioactively labeled probes and are more time consuming 22,28 . Therefore, the use of fluorescence microscopy for bacterial symbiont visualization in nematodes is an effective and efficient screening tool for investigating bacterial interactions with a nematode host.
The authors have nothing to disclose.
The authors wish to thank Eugenio Vivas, Kurt Heungens, Eric Martens, Charles Cowles, Darby Sugar, Eric Stabb, and Todd Ciche for their contributions to the development of this protocol and tools used. KEM and JMC were supported by National Institutes of Health (NIH) National Research Service Award T32 (AI55397 “Microbes in Health and Disease”). JMC was supported by a National Science Foundation (NSF) Graduate Research Fellowship. This work was supported by grants from the National Science Foundation (IOS-0920631 and IOS- 0950873).
Name of the reagent | Company | Catalogue number | Comments (optional) |
Lipid Agar (sterile) |
8 grams nutrient broth, 15 grams agar, 5 grams yeast extract, 890 ml water, 10 ml 0.2 g/ml MgCl2. 6H20, 96 ml corn syrup solution*, 4 ml corn oil* Stir media while pouring plates *add sterile ingredient after autoclaving |
||
Corn Syrup Solution (sterile) |
7 ml corn syrup, 89 ml water mix and autoclave |
||
Egg Solution | 16.6 ml 12% sodium hypochlorite, 5 ml 5M KOH, 80 ml water | ||
Lysogeny Broth (sterile) |
5 grams yeast extract, 10 grams tryptone, 5 grams salt, 1 L water mix and autoclave |
||
Microfuge | Fisher | 13-100-675 | Any microfuge that holds microfuge tubes will work |
Centrifuge | Beckman | 366802 | Large table top centrifuge that holds 15 ml and 50 ml conical tubes |
Sterile 60 mm X 15 mm Petri Dish | Fisher | 0875713 | |
50 ml centrifuge tubes | Fisher | 05-539-6 | |
15 ml centrifuge tubes | Fisher | 05-531-6 | |
Sterile 100 mm X 20 mm Petri Dish | Fisher | 0875711Z | Deeper than standard Petri dishes |
24-well plate | Greiner Bio-One | 662000-06 | |
Microscope | The microscope needs florescent capabilities compatible with your fluorophore | ||
Paraformaldehyde | Electron Microscopy Sciences | 15710 | |
PBS (sterile) |
8 g NaCL 0.2 g KCL 1.44 g Na2HPO4 0.24 g KH2PO4 1 L water Adjust to a pH of 7.4 and water to 1 L and autoclave |
||
Microfuge tubes | Fisher | 05-408-138 | 2 ml or 1.5 ml tubes |
Shaker | Any shaker that causes the liquid to gently move will work | ||
Diaminopimelic acid | Sigma | D-1377 | If needed, supplement media to a concentration or 1 mM |