Here, we present a protocol to spatially and temporally assess the presence of viable microbiota in tick guts using a modified whole-mount in situ hybridization approach.
Infectious diseases transmitted by arthropod vectors continue to pose a significant threat to human health worldwide. The pathogens causing these diseases, do not exist in isolation when they colonize the vector; rather, they likely engage in interactions with resident microorganisms in the gut lumen. The vector microbiota has been demonstrated to play an important role in pathogen transmission for several vector-borne diseases. Whether resident bacteria in the gut of the Ixodes scapularis tick, the vector of several human pathogens including Borrelia burgdorferi, influence tick transmission of pathogens is not determined. We require methods for characterizing the composition of the bacteria associated with the tick gut to facilitate a better understanding of potential interspecies interactions in the tick gut. Using whole-mount in situ hybridization to visualize RNA transcripts associated with particular bacterial species allows for the collection of qualitative data regarding the abundance and distribution of the microbiota in intact tissue. This technique can be used to examine changes in the gut microbiota milieu over the course of tick feeding and can also be applied to analyze expression of tick genes. Staining of whole tick guts yield information about the gross spatial distribution of target RNA in the tissue without the need for three-dimensional reconstruction and is less affected by environmental contamination, which often confounds the sequencing-based methods frequently used to study complex microbial communities. Overall, this technique is a valuable tool that can be used to better understand vector-pathogen-microbiota interactions and their role in disease transmission.
Human and livestock pathogens transmitted by arthropod vectors are found worldwide and account for about 20% of infectious diseases globally1, but effective and safe vaccines against most of these pathogens are not available. Our understanding of the important role of commensal, symbiotic and pathogenic microorganisms, collectively known as the microbiome2, in modulating and shaping the health of almost all metazoans3 is expanding. It is now evident that arthropod vectors of pathogens also harbor gut microbiota and these vector-associated microbiota have been shown to influence diverse vector-borne pathogens4,5. The arthropod microbiome is composed of eubacteria, archaea, viruses, and eukaryotic microbes such as protozoa, nematodes, and fungi6. However, the predominant research focus has been on eubacteria due, in part, to the availability of marker genes and reference databases to identify specific bacterial members.
With a focus on Ixodes scapularis, the tick vector of multiple human pathogens including Borrelia burgdorferi7, the causative agent of Lyme disease, the optimization of a microbial visualization technique was aimed at improving our understanding of tick gut microbiota in the context of vector-pathogen interactions. Several questions remain to be answered in the tick microbiome field. The gut is the site of the first extended encounter between the tick and the incoming pathogen in the context of horizontally transferred pathogens; therefore, understanding the role of vector gut microbiota in modulating vector-pathogen interactions will reveal meaningful insights. Ticks have a unique mode of blood meal digestion, where processing of blood meal components takes place intracellularly8. The gut lumen seemingly serves as a vessel to contain the blood meal as the tick feeds, and nutrient digestion and assimilation ensue throughout the several days of feeding and continue post-repletion. The pathogens acquired by the tick during feeding enter the gut lumen along with the bloodmeal and thus the lumen becomes a primary site of interactions among the tick, pathogen, and resident microbiota. As digestion proceeds through the repletion and Ixodid tick molting, the gut undergoes structural and functional changes9. The composition and the spatial organization of gut bacteria is also likely to vary in concert with the changing gut milieu. It is, therefore, important to understand the architecture of resident bacteria in the tick gut to fully understand the interplay of tick, pathogen, and gut microbiota.
Molecular techniques to describe host-associated microbiota routinely utilize high-throughput parallel sequencing strategies10 to amplify and sequence bacterial 16S ribosomal DNA (rDNA). These sequencing strategies circumvent the need to obtain axenic cultures of specific bacteria and provide an in-depth description of all bacterial members represented in the sample. Nevertheless, such strategies are confounded by the inability to distinguish environmental contaminations from bona fide residents. Further, when assessing samples, such as ticks, that are small in size and hence contain low microbiota-specific DNA yields, the likelihood of amplification of environmental contaminants is increased11 and results in the ambiguous interpretation of microbiome composition. Functional characterization in conjunction with the visualization of specific viable bacteria will, therefore, be critical to define and discern the microbiome of the tick temporally and spatially. Towards this goal, we took advantage of the whole-mount RNA in situ hybridization. This technique is routinely used to assess gene expression patterns in organs and embryos12,13,14 and allows semiquantitative analysis of expression over the entire sample of interest. This differs from traditional in situ hybridization techniques which utilize tissue sections and often require extensive analysis of sectioned material with a computational assembly to predict expression in whole organs15. While whole-mount generally refers to whole organisms12, here whole-mount refers to whole guts or organs. The advantages of using the whole-mount RNA in situ hybridization approach to assess the architecture of tick gut microbiota are multifold. The tick gut is composed of 7 pairs of diverticula, each pair varying in size16. The functional differences, if any, among these diverticula, are not understood in the context of tick biology, tick microbiota or tick-pathogen interactions. Manipulations of the gut that rupture the gut diverticula would displace microbiota present in the gut lumen or those associated loosely with the gut and result in misinterpretation of the spatial localization of microbiota. Fluorescence-labeled RNA in situ hybridization has been utilized earlier to examine tick gut transcripts17 by fixing and opening individual gut diverticula to ensure probe hybridization and to localize B. burgdorferi transcripts by sectioning paraffin-embedded whole ticks18. Both these approaches require manipulations of the tick tissues prior to hybridization that would affect the gut microbiota architecture.
In this report, we describe in detail the protocol to examine viable tick gut microbiota using whole-mount in situ hybridization (WMISH). The use of whole-mount RNA in situ hybridization enables a global understanding of the presence and abundance of specific gut bacteria in the different regions of the gut and may spur new insights into tick gut biology in the context of pathogen colonization and transmission. Further, the use of RNA probes directed against specific bacterial RNA allows detection of viable bacteria in the tick gut.
1. Preparation of DNA Templates
2. Construction of Digoxygenin-UTP RNA Probes
3. Visualization of RNA Probes by RNA Formaldehyde Gel Electrophoresis to Assess RNA Purity
CAUTION: Formaldehyde poses an inhalation hazard; therefore, generate the solutions required for RNA gel analysis in a fume hood. Ethidium bromide is a suspected carcinogen; handle with care.
4. Tick Gut Collection and Fixation
5. Construction of Mesh Sample Baskets and 24-well Basket Holder
6. In Situ Hybridization: Day 1 (Timing: 4 – 5 h)
7. In Situ Hybridization: Day 2 (Timing: 4.5 – 5.5 h)
8. In Situ Hybridization: Day 3 (Timing: 6 h to overnight)
9. In Situ Hybridization: Day 4 (Timing: 3 – 3.5 h)
The measurement and estimation of the quality of the RNA probes are critical prior to beginning the staining. In vitro transcription efficiency depends highly on the amount and quality of the DNA template. We routinely visualized the RNA probes on a formaldehyde gel to verify the purity and amount of probe generated by the transcription reactions. The probes should appear as bright, discrete bands (Figure 2). Spectrophotometric measurements of RNA concentration were highly indicative of probe quality and correlated well with the appearance of the bands on the gel. Thus, the gel visualization step may be skipped if desired once familiarity with the protocol has been attained. Either way, it is important to ensure that sufficient RNA of high quality is produced, as all downstream steps depend upon the robustness of the probes.
The strength of this technique is in facilitating broad observations about the tick microbiota, including presence or absence of bacterial species, changes in abundance of bacteria over time as the tick feeds, and localization of species within the whole tissue. Some variation in staining amount and distribution is normal among biological replicates as well as among the 7 pairs of diverticula of individual guts (Figure 3), therefore, it is best to stain at least five guts per condition to get an idea of average staining pattern. The overall success of the protocol is best gauged by comparing the staining of the sense and antisense samples for each condition. The sense samples, important negative controls for this type of analysis, should have minimal to no purple color (Figure 4). Conversely, antisense samples should display robust staining with minimal background, the intensity of which will depend on the abundance of the specific bacterial transcripts being targeted. The staining pattern within the tissue will also vary based on these parameters. Importantly, the sense and antisense samples must be developed for the same amount of time in order to be able to directly compare them, so these samples should be processed in parallel. Overdevelopment of the color reaction can result in a high amount of background staining, where the sense samples begin to look similar to the antisense (Figure 5). Particular care should be taken at this step in the protocol to avoid overstaining, as there is no way to reverse the reaction once it occurs.
It is critical to make sure all guts are fully submerged in each reagent during every step; an increase in the variation across samples may be due to uneven exposure to reagents. The amount of time that the ticks are fed before the guts are collected also has a significant effect on the results. Ticks that were unfed or only fed for 24 h were difficult to work with; the guts were also more easily damaged and did not stain well (Figure 6) and is a limitation of this approach at this time. Ticks fed for 48 – 72 h were easiest to work with, due to the larger size of these guts when compared to guts from ticks fed for 24 h. Guts from later time points also stained better than 24 h-fed ticks, possibly due to increases in bacterial burden during the later stages of feeding. The 72-h tissue often had a slight background of brown tint from blood, but the purple staining was clearly visible over the tint (Figure 4). The guts are best viewed at low magnification (5X or 10X) to be able to view the staining patterns across the entire tissue using a bright-field microscope. Viewing whole-mount tissue at higher magnification, such as with a 63X oil objective, runs the risk of damaging the tissue as the objective would press against the slide and compress the tissue due to the thickness of the samples. If higher resolution imaging is desired, the potential for tissue sectioning should be examined.
Figure 1. Schematic of template preparation for probe generation. Clone the 16S rRNA amplicon into a TA cloning vector containing T7 and Sp6 promoters. Linearize the construct using restriction enzymes to cut at sites a or b for in vitro transcription using T7 or Sp6 promoter to generate sense or antisense probes respectively. Please click here to view a larger version of this figure.
Figure 2. Visualization of the RNA probe by gel electrophoresis. RNA probes (~700 ng) were electrophoresed on a formaldehyde agarose gel, visualized under UV light, and imaged using a commercial gel imaging system. A representative good quality, lane 1; and poor-quality RNA probe, lane 2.
Figure 3. A representative overview of whole-mount in situ hybridization of 48 h-fed guts. Guts from ticks fed for 48 h stained with antisense RNA complementary to a conserved 16S ribosomal RNA sequence shows differential staining of gut diverticula. Scale bar represents 200 µm. Please click here to view a larger version of this figure.
Figure 4. Representative images of successful whole-mount in situ hybridization in Ixodes scapularis guts. Guts from ticks fed for the indicated amount of time were stained using either sense (negative control) or antisense RNA probes. Antisense probes were complementary to a conserved 16S ribosomal RNA sequence (Universal 16S) or a 16S RNA sequence specific to a particular bacterial genus (as indicated). Enterococcus did not yield robust staining at 48 h. Scale bars represent 140 µm. Please click here to view a larger version of this figure.
Figure 5. Prolonged chromogenic reaction time leads to non-specific staining. Guts from ticks fed for 48 h were probed with a sense RNA probe or an antisense probe complementary to a conserved 16S RNA sequence. The tissue was incubated with the alkaline phosphatase substrate BM purple overnight at 4 °C and then at room temperature for about 4 h before stopping the reaction. Scale bars represent 140 µm. Please click here to view a larger version of this figure.
Figure 6. Whole-mount in situ hybridization using unfed or 24h fed tick guts. Guts from unfed or 24 h-fed ticks were stained using sense RNA probes or antisense probes complementary to a conserved 16S RNA sequence or a sequence specific to the genus Rickettsia. Staining in the antisense samples is less robust than observed at later time points and the guts are also more easily damaged. Scale bars represent 60 µm. Please click here to view a larger version of this figure.
Name | Final Concentrations | Final pH | Storage notes |
MEMFA | 0.1 M MOPS | – | Room temperature |
2 mM EGTA | |||
1 mM magnesium sulfate | |||
3.7% formaldehyde | |||
PBS-Tween (PTw) | 1x PBS, 0.1% Tween-20 | 7.0 | Room temperature |
0.1% Tween-20 | |||
Triethanolamine buffer | 1x PBS | 7.0 – 8.0 | Room temperature |
0.1 M triethanolamine | |||
Hybridization buffer | 50% formamide | – | -20 °C |
5x SSC | |||
1 mg/mL Torula RNA | |||
100 µg/mL heparin | |||
1x Denhart's solution | |||
0.1% Tween-20 | |||
0.1% CHAPS | |||
10 mM EDTA | |||
Maleic acid buffer (MAB) | 100 mM maleic acid | 7.0 | Room temperature |
150 mM sodium chloride | |||
Alkaline phosphatase (AP) buffer | 100 mM Tris | 9.5 | -20 °C |
50 mM magnesium chloride | |||
100 mM sodium chloride | |||
0.1% Tween-20 | |||
5 mM levamisol | |||
Bleach solution | 1% hydrogen peroxide | – | Prepare fresh |
5% formamide | |||
0.5x SSC |
Table 1. Recipes of solutions used in the protocol.
This is the first use of a whole-mount in situ hybridization (WMISH) technique to study the microbiota of an arthropod vector of pathogens. Our protocol was adapted from one used to study development in Drosophila and in frog embryos25,26. Whole-mount RNA in situ hybridization has been routinely used to localize gene transcripts spatially and temporally27 and visualization of transcripts can be done by bright-field or by fluorescent microscopy. We have adopted the former towards detection of microbiota in tick guts. It is important to note that attempts to perform whole-mount in situ hybridization using whole ticks in which the head region was removed to allow penetration of fixative and hybridization solutions was not successful. The protocol described here utilizes dissected guts and has been implemented to study the effects of specific tick proteins on bacterial colonization in the midgut23. Information obtained from WMISH is comparable to immunohistochemistry but has the advantage of not requiring the generation of protein and antibodies to the gene or protein of interest. Genomic information is sufficient to generate the reagents for WMISH. Both WMISH and FISH (fluorescence in situ hybridization) can detect the gene expression. However, due to it being an enzymatic and chromogen-based assay, WMISH has the advantage that color development can be actively monitored, and the reaction is allowed to proceed until desired signal and sensitivity is achieved. It is fully amenable to automation and is easily scalable to high-throughput format. Unlike FISH and immunohistochemistry, no sectioning is required. The disadvantage of WMISH over FISH is that only 2 different mRNAs or genes can be addressed simultaneously and would require that probes be labeled with different nucleotide analogs such as digoxigenin-UTP and fluorescein-UTP28. With FISH, up to 6 different mRNAs can be examined in one assay29. While both assays require comparable time, the cost of fluorescence dyes are higher than that of chromogenic substrates. FISH assays would also require a fluorescence microscope for image visualization, while WMISH image visualization can be performed with any light microscope.
It is crucial that the protocol is followed closely; however, there are some steps that require particular care. Dissection of the midgut from the tick without damaging the tissue requires some dexterity. It may be prudent to collect extra ticks to practice the dissections and gain proficiency. Due to the heterogeneity of staining in the different diverticula of individual guts (Figure 3), it is important to examine all the diverticula of each gut to derive conclusive evidence of bacterial abundance. It is also important to process several guts (at least 5) for each experimental condition in order to obtain conclusive information on the presence, spatial distribution, and abundance of specific bacteria within different diverticula. Production of high-quality RNA probes is also critical to effective tissue staining. It is crucial to also confirm the sequence of the probe template in the cloning vector to ensure that antisense and sense probes are appropriately generated. A productive in vitro transcription reaction requires sufficient template DNA; a minimum of 100 ng may be used, but 0.5 – 1 µg is recommended for optimal probe generation.
The use of sample baskets and 24-well plates in this protocol avoids direct manipulation of the tissue at every wash step, which minimizes damage during the staining process. However, samples are still occasionally lost or damaged; guts from unfed or 24h-fed ticks, in particular, are difficult to see and easy to lose when transferring from the sample baskets to a long-term storage container. Additional care should be taken while handling these samples. Due to this limitation, we have mainly used 48 and 72 h-fed tick guts. The presence of large amounts of blood meal in the guts at these time-points (longer than 72 h) interferes with the staining and visualization due to non-specific background from the blood-meal and also presents a limitation of this technique. It is likely that the use of fluorescently-labeled probes might circumvent this issue.
The most important aspect of the staining procedure is to establish the balance between robust staining and low background signal. The timing of ideal color development can vary depending on the experimental conditions. Thus, it is best to perform the color reaction at room temperature and monitor color development approximately every 30 min. The reaction may also be set up to proceed overnight at 4 °C to slow color development. However, it is extremely critical not to let this reaction proceed for long. The staining reaction can be difficult to monitor by eye, so the samples should be examined under a dissection microscope. A purple tint should be visible on samples probed with antisense RNA, while the samples probed with sense RNA should retain minimal color. If the sense RNA-probed samples stain brightly, troubleshooting steps should begin with reducing the incubation time with BM purple and increasing the blocking time. Sometimes sense probes can cause anomalously high background compared to the antisense probes; in these cases, different regions of the gene might have to be considered for generating RNA probes. DNA sequences not represented in a given sample may also be considered as templates to generate negative control probes. For example, a DNA template encoding a region of the green fluorescent protein, not represented in the tick genome or its microbiome, can be utilized.
One limitation of this technique is that the analysis is largely qualitative; however, image analysis software (Table of Materials) could be used to measure the amount of staining signal in each sample. This would allow a semi-quantitative comparison of the abundance of various bacterial species under different experimental conditions. While this is a time-consuming protocol that takes 3 – 4 days to complete, it is not laborious; each day hands-on time is only about 3 – 5 h on average. However, the ability to process many samples in parallel with the use of the sample baskets and holders allows for high-throughput analysis. Further, this process can be automated using commercially available instruments (Table of Materials) if this protocol is expected to be an essential and routine component of the research laboratory by using commercially available automation-compatible platforms (Table of Materials) to further reduce hands-on time.
The described protocol makes use of digoxygenin-labeled probes that allow the production of a colorimetric signal that can be imaged using bright field microscopy. While we have utilized one chromogenic substrate to detect hybridized RNA probes specific to individual targets, it is also possible to detect multiple targets simultaneously by labeling probes with different nucleotide analogs, such as digoxigenin-UTP and fluorescein-UTP and different chromogenic substrates30. The tissue samples can then be incubated sequentially with alkaline-phosphatase-coupled antibodies against digoxigenin and fluorescein, with different chromogenic reactions for the two steps. This technique could also be adapted for fluorescently-labeled probes, which may facilitate higher-resolution imaging using confocal microscopy31 and provide the option of using multiple probes of different colors in parallel. This technique can only specifically assess a few microorganisms at one time in one sample. However, several samples can be assessed in high-throughput assays to address multiple microorganisms that may be represented in tick microbiota.
Finally, this technique is not limited to the investigation of the interactions between ticks and their associated microbiota. Probes complementary to any gene of interest within the tick genome may be generated and used to examine the abundance and localization of specific genes in different tissues. The salivary glands of the tick can also be processed and analyzed similarly to the gut. Overall, this technique has broad potential for adaptation to studies on the dynamics of microbial communities within other arthropod disease vectors of interest.
The authors have nothing to disclose.
We sincerely thank Dr. Mustafa Khokha, Yale University, for providing the use of his laboratory resources. We are grateful to Mr. Ming-Jie Wu for excellent technical assistance. EF is an HHMI investigator.This work was supported by a gift from the John Monsky and Jennifer Weis Monsky Lyme Disease Research Fund.
Sefar NITEX Nylon Mesh, 110 micron | Amazon | 03-110/47 | |
pGEM-T Easy Vector System | Promega | A1360 | |
Digoxygenin-11-UTP | Roche | 1209256910 | |
dNTP | New England Biolabs | N0447S | |
DNAse I(RNAse-free) | New England Biolabs | M0303S | |
HiScribe SP6 RNA synthesis kit | New England Biolabs | E2070S | |
HiScribe T7 High Yield RNA Synthesis Kit | New England Biolabs | E2040S | |
Water, RNase-free, DEPC-treated | American Bioanalytical | AB02128-00500 | |
EDTA, 0.5M, pH 8.0 | American Bioanalytical | AB00502-01000 | |
Formaldehyde, 37% | JT Baker | 2106-01 | |
Formamide | American Bioanalytical | AB00600-00500 | |
EGTA | Sigma Aldrich | E-4378 | |
DPBS, 10X | Gibco | 14300-075 | |
Tween-20 | Sigma Aldrich | P1379-25ML | |
Proteinase K | Sigma Aldrich | 3115879001 | |
Triethanolamine HCl | Sigma Aldrich | T1502-100G | |
Acetic anhydride | Sigma Aldrich | 320102-100ML | |
Paraformaldehyde | ThermoScientific/Pierce | 28906 | |
SSC, 20X | American Bioanalytical | AB13156-01000 | |
RNA from torula yeast | Sigma Aldrich | R3629-5G | |
Heparin, sodium salt | Sigma Aldrich | H3393-10KU | |
Denhardt's Solution, 50X | Sigma Aldrich | D2532-5ML | |
CHAPS hydrate | Sigma Aldrich | C3023-1G | |
RNase A | Sigma Aldrich | 10109142001 | |
RNase T1 | ThermoScientific | EN0541 | |
Maleic acid | Sigma Aldrich | M0375-100G | |
Blocking reagent | Sigma Aldrich | 11096176001 | |
Anti-Digoxigenin-AP, Fab fragments | Sigma Aldrich | 11093274910 | |
Levamisol hydrochloride | Sigma Aldrich | 31742-250MG | |
Chromogenic substrate for alkaline phosphatase | Sigma Aldrich | 11442074001 | |
Bouin's solution | Sigma Aldrich | HT10132-1L | |
Hydrogen peroxide | Mallinkrodt Baker, Inc | 2186-01 | |
Single stranded RNA ladder | Ambion -Millenium | AM7151 | |
#11 High-Carbon steel blades | C and A Scientific Premiere | #11-9411 | |
Thermocycler | BioRad, CA | 1851148 | |
Spectrophotometer | ThermoScientific | NanoDrop 2000C | |
Orbital shaker | VWR | DS-500E Digital Orbital shaker | |
Shaking water bath | BELLCO Glass, Inc | Hot Shaker-7746-12110 | |
Gel documentation system | BioRad | Gel Doc XR+ Gel documentation system | |
Bright-field Microscope | Nikon | NikonSM2745T | |
Bright-field Microscope | Zeiss | AXIO Scope.A1 | |
Dissection microscope | Zeiss | STEMI 2000-C | |
Light box | VWR | 102097-658 | |
PCR purification kit | Qiagen | 28104 | |
Image capture software | Zeiss | Zen lite | |
Image editing software | Adobe | Adobe Photoshop CS4 version 11.0 | |
Image analysis software | National Institutes of Health | ImageJ-NIH /imagej.nih.gov/ij/ | |
Automation compatible instrumentation | Intavis Bioanalytical Instruments, Tubingen, Germany). | Intavis, Biolane HT1.16v |