Skip to main content
  • Loading metrics

Thermal stress responses of Sodalis glossinidius, an indigenous bacterial symbiont of hematophagous tsetse flies

  • Jose Santinni Roma,

    Roles Investigation, Visualization, Writing – original draft

    Affiliation Department of Biology, University of Richmond, Richmond, Virginia, United States of America

  • Shaina D’Souza,

    Roles Investigation

    Affiliation Department of Biology, University of Richmond, Richmond, Virginia, United States of America

  • Patrick J. Somers,

    Roles Investigation

    Affiliation Department of Biology, University of Richmond, Richmond, Virginia, United States of America

  • Leah F. Cabo,

    Roles Investigation, Visualization

    Affiliation Department of Biology, University of Richmond, Richmond, Virginia, United States of America

  • Ruhan Farsin,

    Roles Investigation

    Affiliation Department of Biology, University of Richmond, Richmond, Virginia, United States of America

  • Serap Aksoy,

    Roles Resources, Writing – review & editing

    Affiliation Yale School of Public Health, Department of Epidemiology of Microbial Diseases, New Haven, Connecticut, United States of America

  • Laura J. Runyen-Janecky ,

    Roles Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing (LJR-J); (BLW)

    Affiliation Department of Biology, University of Richmond, Richmond, Virginia, United States of America

  • Brian L. Weiss

    Roles Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing (LJR-J); (BLW)

    Affiliation Yale School of Public Health, Department of Epidemiology of Microbial Diseases, New Haven, Connecticut, United States of America


Tsetse flies (Diptera: Glossinidae) house a taxonomically diverse microbiota that includes environmentally acquired bacteria, maternally transmitted symbiotic bacteria, and pathogenic African trypanosomes. Sodalis glossinidius, which is a facultative symbiont that resides intra and extracellularly within multiple tsetse tissues, has been implicated as a mediator of trypanosome infection establishment in the fly’s gut. Tsetse’s gut-associated population of Sodalis are subjected to marked temperature fluctuations each time their ectothermic fly host imbibes vertebrate blood. The molecular mechanisms that Sodalis employs to deal with this heat stress are unknown. In this study, we examined the thermal tolerance and heat shock response of Sodalis. When grown on BHI agar plates, the bacterium exhibited the most prolific growth at 25oC, and did not grow at temperatures above 30oC. Growth on BHI agar plates at 31°C was dependent on either the addition of blood to the agar or reduction in oxygen levels. Sodalis was viable in liquid cultures for 24 hours at 30oC, but began to die upon further exposure. The rate of death increased with increased temperature. Similarly, Sodalis was able to survive for 48 hours within tsetse flies housed at 30oC, while a higher temperature (37oC) was lethal. Sodalis’ genome contains homologues of the heat shock chaperone protein-encoding genes dnaK, dnaJ, and grpE, and their expression was up-regulated in thermally stressed Sodalis, both in vitro and in vivo within tsetse fly midguts. Arrested growth of E. coli dnaK, dnaJ, or grpE mutants under thermal stress was reversed when the cells were transformed with a low copy plasmid that encoded the Sodalis homologues of these genes. The information contained in this study provides insight into how arthropod vector enteric commensals, many of which mediate their host’s ability to transmit pathogens, mitigate heat shock associated with the ingestion of a blood meal.

Author summary

Microorganisms associated with insects must cope with fluctuating temperatures. Because symbiotic bacteria influence the biology of their host, how they respond to temperature changes will have an impact on the host and other microorganisms in the host. The tsetse fly and its symbionts represent an important model system for studying thermal tolerance because the fly feeds exclusively on vertebrate blood and is thus exposed to dramatic temperature shifts. Tsetse flies house a microbial community that can consist of symbiotic and environmentally acquired bacteria, viruses, and parasitic African trypanosomes. This work, which makes use of tsetse’s commensal endosymbiont, Sodalis glossinidius, is significance because it represents the only examination of thermal tolerance mechanisms in a bacterium that resides indigenously within an arthropod disease vector. A better understanding of the biology of thermal tolerance in Sodalis provides insight into thermal stress survival in other insect symbionts and may yield information to help control vector-borne disease.


Tsetse flies (Order: Diptera) house a microbial community that can consist of symbiotic and environmentally acquired bacteria, viruses, and parasitic trypanosomes. Among these are the primary endosymbiont Wigglesworthia glossinidia, a secondary symbiont Sodalis glossinidius, parasitic Wolbachia sp. (reviewed in [1, 2]) and Spiroplasma [3]. Sodalis (order: Enterobacteriaceae) resides intra- and extracellularly within the fly’s midgut, hemolymph, milk and salivary glands, muscle, and fat body tissues [47]. Both Sodalis and Wigglesworthia are passed vertically to tsetse progeny via maternal milk gland secretions [8, 9]. Although the population dynamics of Sodalis in laboratory reared and field-captured tsetse flies has been well-documented [1012], the functional relevance of this secondary symbiont to the fly’s physiology is currently unclear. Sodalis likely provides some benefit to tsetse, as flies exhibit a reduced lifespan when Sodalis is selectively eliminated via treatment with antibiotics [13]. Additionally, Sodalis may modulate tsetse’s susceptibility to infection with parasitic African trypanosomes (Trypanosoma brucei sp.) [14], which are the etiological agents of human and animal African trypanosomiases. Specifically, this bacterium’s chitinolytic activity results in the accumulation of N-acetyl-D-glucosamine, which is a sugar that inhibits the activity of trypanocidal lectins [15]. In support of this theory, several studies using field-captured tsetse have noted that the prevalence of trypanosome infections positively correlates with increased Sodalis density in the fly’s gut [1619].

Like many animal-bacterial symbiotic consortia, the ectothermic tsetse fly and its endosymbionts are sensitive to changes in temperature. In fact, the effect of temperature on the insect-symbiont relationship, and how symbionts contribute to insect host thermal tolerance, are common experimental considerations. Corbin et al [20] analyzed data from a large number of insect-symbiont systems and concluded that the nature of the insect hosts thermal environment influences symbiont density. Furthermore, the effect of symbionts on the thermal tolerance of the host can be complex. This relationship has been well studied in the pea aphid (Acyrthosiphon pisum) model system. This insect can house several symbiotic bacteria, including Buchnera aphidicola, Serratia symbiotica and Hamiltonella defensa, all of which increase their hosts tolerance to heat stress. Conversely, another symbiont, Regiella insecticola, decreases the aphid’s survival under high temperature conditions [21, 22]. These relationships are reflective of the fact that the thermal environment mediates a variety of interactions between insect hosts and their microbial symbionts.

The gut of hematophagous arthropods, including that of the tsetse fly, represents an environment that presents rapid and dramatic feeding status-based temperature shifts that likely alter the thermal stress physiology of the fly and its indigenous symbionts [23, 24]. In fact, temperature is one of the most important factors that controls bacterial growth and survival. Temperatures approaching and at the maximal temperature for a given bacterial species cause protein denaturation and membrane destabilization. Stabilization and refolding of denatured proteins via protein chaperones comprise mechanisms that bacteria use to combat thermal stress at elevated temperatures. One of the major chaperone systems is the ATP-dependent DnaK system, which is comprised of DnaK, the co-chaperone DnaJ and the nucleotide exchange factor GrpE (reviewed in [25, 26]). DnaK homologues are distributed across all three branches of life. With the help of DnaJ, DnaK binds to unfolded proteins. ATP hydrolysis facilitates a conformational change in DnaK, which then surrounds the substrate protein and enables refolding. GrpE then facilitates ATP regeneration at the complex, which causes a conformational change that releases the refolded protein. In E. coli the expression of these three genes increases upon exposure to elevated temperatures via the alternative σ32H) sigma factor, which directs RNA polymerase to the dnaK/dnaJ and grpE promoters [27, 28].

The functional role of DnaK as it relates to thermal stress has been studied using a select number of model bacterial species. As such, this topic is understudied in symbionts that reside within ectothermic animal hosts. With respect to the DnaK/DnaJ chaperone system, Brooks et al. [29] showed that the system is required for Vibrio fischeri colonization of its Euprymna scolopes squid host via regulation of proper biofilm formation. Manipulating genes of bacteria that are symbionts of insects, or that are vectored by insects, is technically challenging. As such, functional characterization of symbiont DnaK has heretofore been performed by ectopically expressing corresponding genes in heterologous bacteria. In these studies, the dnaK genes from Buchnera aphidicola, an aphid symbiont, and Borrelia bordoferii, which is vectored by ticks, showed partial to no complementation of the thermal sensitive phenotype in E. coli dnaK mutants [30, 31]. Additionally, a second chaperone, GroEL, is one of the most highly expressed proteins in many insect bacterial symbionts, including Sodalis [32, 33]. The molecular mechanisms that underlie Sodalis’ ability to reside successfully within the thermally fluctuating tsetse midgut environment are currently unknown. In this study we investigate Sodalis’ thermal tolerance profile and the functionality of the bacterium’s DnaK/DnaJ/GrpE chaperone system in response to thermal stress.


Bacterial strains, plasmids, and growth conditions

The bacterial strains and plasmids used in this study are listed in S1 Table. E. coli strains were grown in Luria-Bertani Broth (LB) or on Luria-Bertani Agar (L Agar) plates. Liquid cultures were incubated at 37°C with 200 rpm aeration. Sodalis glossinidius were grown at 25°C and 10% CO2 on Brain Heart Infusion (BHI) agar both with or without 10% horse blood (BBHI) (Quad Five, Ryegate, MT). The initial primary Sodalis culture used in these experiments (SODF) was established by washing two-week-old G. m. morsitans pupae consecutively in 40% EtOH, 30% EtOH and sterile BHI media for 30 minutes per solution. Sterilized pupae were then homogenized in 100 μl of fresh BHI and plated on BBHI plates without antibiotics. Liquid Sodalis cultures were started by inoculating colonies into liquid BHI in petri dishes and incubating without aeration in a 10% CO2 microaerophilic environment. New cultures of Sodalis were typically inoculated at optical densities at 600 nm (OD600) of approximately 0.08. Antibiotics were used for E. coli at the following concentrations: carbenicillin (carb) 125 μg/ml, ampicillin (amp) 50 μg/ml, chloramphenicol (cam) 30 μg/ml, and kanamycin (kan) 50 μg/ml.

Insect maintenance

Glossina morsitans were maintained at 25°C with 60-65% relative humidity in the insectary at the Yale School of Public Health. Unless indicated otherwise, all flies received defibrinated bovine blood (Quad Five) every 48 hours through an artificial membrane feeding system [34].

Plasmid construction

Plasmids were isolated using the QIAprep Spin Miniprep Kit (Qiagen, Valencia, CA). DNA fragments and enzyme reactions were purified with QIAquick gel extraction kit (Qiagen). All ligations were done with T4 DNA Ligase (Promega) and transformed into E. coli DH5α. The sequences of the PCR primers for cloning are listed in S2 Table, and all PCRs for cloning were done using PfuTurbo (Agilent, Santa Clara, CA). To construct pJR1 and pJS2, DNA fragments containing dnaK were PCR amplified from Sodalis or from E. coli (with primers UR423 and UR424 or UR518 and UR519, respectively), digested with XbaI and XhoI, and ligated to pWKS30 digested with the same enzymes. To construct pJR5, a modified procedure from the Quickchange II Site Directed Mutagenesis Kit (Agilent Technologies, Santa Clara, CA) was used. A DNA fragment containing dnaJ and ends that match the pJR1 plasmid was PCR amplified from Sodalis with primers UR530 and UR531. This PCR fragment was then annealed to pJR1 and used as the primers to synthesize a larger plasmid containing both Sodalis dnaK and dnaJ. The template pJR1 plasmid was then degraded with addition of DpnI, and the reaction mixture was transformed into E. coli NEB 5-alpha to recover pJR5. This plasmid was sequenced to verify that the DNA sequence of dnaKJ was the same as the chromosomal sequence. To construct pSD2, pJR5 was cut with MscI and a 1 kb internal segment of dnaK was deleted. To construct pRF2, a DNA fragment containing grpE was PCR amplified from Sodalis with primers UR455 and UR456 digested with XbaI and PstI, and ligated to pWKS30 digested with the same enzymes.

Thermal stress and survival assays for in vitro Sodalis

For growth assays, Sodalis were cultured in BHI at an OD600 of 0.08, grown for 2 days at 25oC, re-normalized to an OD of 0.02 (~107 bacteria per ml), and serially diluted in BHI broth. 10 μl (or 100 μl) of each serial dilution was spotted (or spread) on agar plates [BHI, BBHI or BHI supplemented with 0.6-10% horse serum (Hemostat Labs, Dixon, CA)]. Plates were incubated in a 10% CO2 incubator or in CampyPak microaerobic pouches (Becton, Dickinson and Company, Franklin Lakes, NJ) at the various temperatures indicated in the figure and table legends. Growth under each condition was assessed by scoring for the presence or absence of colonies after 7 days.

For thermal tolerance survival assays, Sodalis were inoculated in BHI at an OD600 of 0.08, grown for two days at 25oC, diluted to an OD600 of 0.1 in BHI broth, and aliquoted into samples that were incubated at 25oC, 30oC, 32oC, or 37oC for 3 days. Viable surviving bacteria were quantified each day by serially diluting the samples and plating on BHI agar plates and incubating the plates at 25°C. The number of colonies was counted seven days later, total number of viable cells at each temperature and time point were calculated by multiplying the number of colonies by the dilution factor and dividing by the amount of sample plated.

Thermal stress assays for in vivo Sodalis

Sodalis residing within tsetse were thermally stressed by holding flies at 30oC or 37oC in an incubator for 48 hours. Subsequently, midguts microscopically excised from these flies, and control flies maintained at 25oC, were homogenized in 0.85% NaCl, serially diluted and plated on BBHI agar [10]. Colony forming units per plate were counted manually.

Quantitation of Sodalis chaperone gene expression

For measuring in vitro gene expression, Sodalis cultures were grown in BHI from a starting OD600 of 0.08 for 1 day and then thermally stressed as follows: Cultures (1 ml) were transferred to microfuge tubes and placed in water baths at 25oC or 30oC for 15 minutes, and 37oC for 5, 10 and 15 minutes. 250 μl of RNA stabilizing reagent (95% acidic phenol/5% ethanol) was added to each sample to stabilize the RNA after incubation. The RNA was isolated using the RNeasy Mini Kit Procedure (Qiagen), and the isolated RNA was treated with DNase I (Qiagen) until the samples were free of contaminating DNA as confirmed by PCR. cDNA was generated from 200 ng of total RNA using Superscript III and random hexamers (Invitrogen, Carlsbad, CA).

Sodalis gene expression was measured ex vivo by dissecting midguts (n=5 biological replicates, each containing 5 midguts) from 8 day old wild-type tsetse that had previously been fed three times, the last feeding having occurred 72 hours prior to dissection. The intact Sodalis harboring tissues were placed on a 37°C heat block (housed in an incubator set at the same temperature) for two minutes. We chose to perform the experiment in this manner because the large quantity of blood present in tsetse’s midgut following consumption of a meal (the fly can consume 2-5 times its body weight each time it feeds [35]) interferes with the accurate measurement of symbiont gene expression in this tissue. We incubated the midguts at 37°C for two minutes because these parameters most accurately reflect the physiological conditions that Sodalis experiences when tsetse feeds. Specifically, the fly’s body reaches its maximum temperature of 37°C within 50 seconds of consuming a vertebrate blood meal [20]. We allowed the midguts to incubate at this temperature for an additional 70 seconds to account for the fact that our experimental radiant transfer of heat occurred from the outside in, while under natural conditions Sodalis present in tsetse’s midgut are immediately bathed directly in vertebrate blood when the fly consumes a meal. Control samples (n=5 biological replicates, each containing 5 midguts) were obtained by dissecting midguts from tsetse housed under normal insectary conditions (25°C). Midgut total RNA (including symbiont RNA) was then isolated using Trizol reagent (Invitrogen, Carlsbad, CA) and treated with DNase (Turbo DNA-free kit, Thermo Fisher) until the samples were free of contaminating DNA. cDNA was generated from 200 ng of total RNA using the iScript cDNA synthesis kit (Bio-Rad).

For both in vitro and in vivo analyses, quantitative PCR was performed using primers UR545 and UR546 for dnaK, UR556 and UR557 for dnaJ, and UR558 and UR559 for grpE., Constitutively expressed gene rplB (previously verified in [36, 37], amplified using primers QrplB1F and QrplB1R, was used to normalize transcript levels. All primer sequences are indicated in S2 Table.

Thermal stress assays for E. coli mutants expressing Sodalis chaperone genes

Overnight cultures of the E. coli strains carrying plasmids with Sodalis genes were grown in LB containing carb and the antibiotic marking the chromosomal mutations (cam for MC4100ΔdnaK and kan for JW0014-1) at 30°C. For assessment of growth on agar plates, the overnight cultures were serially diluted 1:10 six times in LB and 10 μL each dilution were spotted onto L agar plates containing carb. Plates were placed at 30°C (non-stress temperature) and 45±1°C (thermal stress temperature) overnight and growth was assessed the following day. For assessment of growth in liquid, the overnight cultures were subcultured 1:100 in LB containing carb and incubated at 30°C until midlog stage was reached. Then, each culture was diluted to an OD600 of 0.06 in LB containing carb and incubated at either 30°C or 45±1°C. Growth was measured via optical density at 600 nm.

Statistical analyses

All statistical analyses were carried out using GraphPad Prism (v.8) or Microsoft Excel. ‘Raw’ data underlying the results presented in the figures are contained in Supplemental data files 1 and 2. All statistical tests used, and statistical significance between treatments, and treatments and controls, are indicated on the figures or in their corresponding legends. All samples sizes are provided in corresponding figure legends or are indicated graphically as points on dot plots. Biological replication implies distinct aliquots of cultured Sodalis, and distinct groups of flies collected on different days, were used for all experiments.


Establishing the thermal range for Sodalis growth in vitro on un-supplemented BHI agar

To determine maximal temperature for Sodalis growth, Sodalis cultures were serially diluted and spotted on BHI agar plates, which were then incubated at various temperatures for 7 days. All of the spotted Sodalis dilutions (ranging from 10-1 to 10-4) formed colonies on BHI agar at temperatures up to 29°C. At 30°C on BHI, Sodalis was unable to form colonies at the highest two dilutions (10-3 and 10-4) and did not form any colonies at ≥ 31°C (Table 1, column 1).

Table 1. Thermal growth range for Sodalis on agar plates.

Presence of blood in the agar or growth in microaerobic environments extends Sodalis thermal tolerance

Tsetse’s gut is likely hypoxic, and after the first adult feeding, always contains some quantity of vertebrate blood. Thus, to more closely mimic gut conditions and determine how they influence Sodalis thermal stress survival, we tested the bacterium’s thermal tolerance when grown on BHI agar plates supplemented with blood and/or in a reduced oxygen environment (through the use of CampyPak microaerobic pouches, which reduces environmental oxygen levels to 5-15%). We found that either condition enabled Sodalis to form colonies at the highest dilution (10-4) at 30°C (Table 1). However, as the temperature increased, the ability of the bacteria to form colonies at all dilutions was diminished in these supplemented growth conditions. At 32°C, no colonies were present on the blood-supplemented plate incubated aerobically, and only a hazy film of growth formed at 10-1 upon CampyPak-incubation. The plates that had both blood-supplementation and CampyPak-incubation showed colonies at 10-2 dilutions at 32°C, but just a haze of growth at 33°C (Table 1).

Growth stimulation due to blood supplementation of agar could be due to either a component of the erythrocytes or the serum fraction. To determine which contained the stimulating factor(s), we added either purified erythrocytes or serum to the BHI agar plates. Only the agar supplemented with the erythrocytes supported growth (Table 2), suggesting that some component of this cell type facilitates Sodalis growth under stressed conditions.

Exposure to temperatures above 30°C is lethal to Sodalis in vitro and in vivo

The lack of Sodalis growth above 30°C on BHI agar could be because this temperature is either bacteriostatic or bactericidal (bacteria are alive but not growing, versus bacteria are dead). To distinguish between these two possibilities, we exposed Sodalis cultured in BHI broth to temperatures that were non-permissive for optimal growth (≥ 30°C) and then plated the bacteria on agar plates at a permissive temperature to quantify the number of cells that remained viable in the BHI broth culture. Within 24 hours of exposure to the non-permissive temperatures, less than 40% of the Sodalis survived temperatures above 30°C. By 72 hours, only 0.4% of the cells were recovered from samples exposed to 32°C, and no bacteria could be recovered from the samples at 37°C (Fig 1A). The samples incubated at 30°C were able to survive as well as the samples at 25°C for the first 24 hours. However, only 33% and 12% of the cells survived for 48 and 72 hours, respectively.

Fig 1. Sodalis survival at non-permissive growth temperatures.

(A) Two-day old Sodalis cultures were diluted to an OD600 of 0.1 and incubated at 25°, 30°C, 32°C and 37°C for two days. Viable surviving bacteria were quantified by plating on BHI agar plates at the indicated time points and incubating the plates at 25°C. The number of colonies was counted seven days later and used to determine the total number of viable cells at each temperature and time point. Each timepoint represents the average of three trials, ± standard deviation. (B) Tsetse flies were reared at 25°C, 30°C and 37°C for two days, after which midguts were excised, homogenized in 0.85% NaCl and plated on BBHI agar. Sodalis density per midgut was determined by manually counting colonies. Each data point on the graph represents one midgut (n=10 per treatment), and statistical significance was determined using a one-way ANOVA followed by Tukey’s HSD post-hoc analysis.

We next investigated the thermal tolerance of Sodalis that reside indigenously within tsetse by exposing wild-type flies to elevated temperatures for two days. Similar numbers of Sodalis were recovered from control (1.8x106±1.3x105 CFU) and treatment (2.2x106±1.8x105 CFU) tsetse housed at 25°C and 30°C, respectively. Conversely, tsetse housed at 37°C harbored 4.2x104±1.3x104 Sodalis CFU, which represents a 98% reduction in bacterial density compared to controls (Fig 1B). Thus, unlike cultured Sodalis, cells that reside within tsetse’s midgut are able to survive at 30°C or at least 2 days. This outcome suggests that the two Sodalis populations (in vitro and in vivo) present different thermal stress responses.

Sodalis contains DnaK, DnaJ, and GrpE homologues

The DnaK/DnaJ/GrpE chaperone system enables bacterial cells to respond to elevated temperature (reviewed in [26]). Sodalis’s genome encodes proteins annotated as DnaK, DnaJ, and GrpE [38]. The dnaK and dnaJ genes are tandemly organized on the chromosome, with a 117 bp intergenic region between them, while grpE is located 2 Mbp away. Well-conserved σ32 binding sites are located 124 and 47 bp 5’ of the dnaK and grpE start codons, respectively (S1A and S1B Fig). No σ32 or σ70 binding sites are present in the intergenic region between dnaK and dnaJ, but RNAs containing this sequence are capable of folding into an extended hairpin (S1C Fig).

Sodalis is less able to survive high temperatures than is closely-related, free-living E. coli. These differential phenotypes prompted us to compare the putative amino acid sequences of the Sodalis and E. coli chaperone proteins. Sodalis DnaK and DnaJ are 89% identical to their E. coli homologues, while Sodalis and E. coli GrpE exhibit 65% identity. Additionally, these Sodalis proteins have retained the conserved residues implicated in conferring thermal tolerance in E. coli (S2 Fig). We also examined the genomes of other insect bacterial symbionts for Sodalis DnaK homologues by performing a BLAST search with Sodalis DnaK. We found high levels of DnaK conservation with Sodalis DnaK in a wide variety of insect symbionts, including the primary symbiont of tsetse Wigglesworthia glossinidia (S3 Fig, S3 Table). Additionally, these proteins also retain conserved residues that confer thermal tolerance in E. coli. Our findings suggest that distinct bacterial taxa (e.g., E. coli and Sodalis) may exhibit very different thresholds of thermal stress despite the highly conserved nature of chaperone genes. Furthermore, elevated expression of these genes occurs at different temperatures, and other proteins may be involved in modulating thermal stress, in different bacteria.

dnaK, dnaJ, and grpE transcription increases at elevated temperatures in vitro and in vivo

In E. coli and many other bacteria, temperatures above the optimal growth temperature induce expression of the ATP-dependent dnaK system [39]. Thus, we hypothesized that expression of Sodalis homologues would similarly increase at temperatures above 25oC. To test this hypothesis initially, we isolated total RNA from Sodalis that had been exposed to 25oC and 30°C temperatures in vitro for 15 minutes and then used RT-qPCR to measure dnaK, dnaJ, and grpE expression levels. Sodalis dnaK expression increased an average of 5.4-fold when the bacterium was exposed to 30°C, as compared to 25°C (Fig 2A). Likewise, dnaJ and grpE expression were on average 8.4-fold and 6.7-fold induced, respectively, at elevated temperatures (Fig 2A).

Fig 2. Sodalis chaperone gene expression positively correlates with increased temperature.

(A) Cultured Sodalis were grown in BHI at 25°C for 1-2 days. Subsequently, the culture was split into two samples that were incubated at either 25°C or 30°C for 15 min. dnaK, dnaJ, grpE and constitutively expressed rplB transcript abundance was measured by RT-qPCR using gene specific primers. ns, not significant. (B) Cultured Sodalis were grown in BHI at 25°C for 1-2 days and subsequently split into two samples that were then incubated at either 25°C or 37°C for five, 10 or 15 minutes. Sodalis dnaK and rplB expression levels were then measured by RT-qPCR. (C) Sodalis harboring tsetse midguts were excised and incubated at either 25°C or 37°C for two minutes. Ex vivo expression of dnaK, dnaJ and grpE transcript abundance was measured by RT-qPCR using gene specific primers. In A, B and C, one dot on each graph represents one biological replicate, each of which contains Sodalis from three independent cultures (A and B) and midguts from five tsetse flies (C). Target gene (dnaK, dnaJ and grpE) expression in all experiments was normalized to the transcript abundance of constitutively expressed rplB. Horizontal bars in (A), (B) and (C) represent median values. Statistical analyses for (A) and (C) were unpaired t-tests, and for (B) one-way ANOVA followed by Tukey’s HSD post-hoc analysis.

We additionally analyzed the expression of Sodalis dnaK after incubation for 5, 10 and 15 minutes at 37oC, which mimics the temperature that the bacterium would be exposed to when its tsetse host consumes a vertebrate blood meal [20]. dnaK transcript abundance increased at the earliest time point and continued to significantly increase thereafter (Fig 2B).

Finally, we monitored how Sodalis that reside indigenously within tsetse’s gut respond to thermal stress. To do so we dissected Sodalis-harboring midguts from flies and then incubated the intact organ at 37°C for two minutes (the rationale behind why we used this ex vivo experimental protocol is described in detail in the Materials and Methods, subsection ‘Quantitation of Sodalis chaperone gene expression’). Control midguts came from flies housed at 25°C. Expression of Sodalis dnaK, dnaJ, and grpE increased on average 2.9-fold, 2.1-fold and 11.5-fold, respectively, when tsetse midguts were exposed to 37°C for two minutes (Fig 2C).

Taken together, these data indicate that Sodalis’ thermal stress tolerance system is induced at the temperature (30°C) that inhibits bacterial growth in vitro. Additionally, the system is also transcriptionally active both in vitro and in the bacterium’s native niche of tsetse’s midgut when exposed to ecological conditions reflective of those present in recently fed flies.

Sodalis DnaK, DnaJ, and GrpE mediate thermal tolerance in a heterologous E. coli host

To test whether the Sodalis DnaK, DnaJ, and GrpE proteins are functional chaperones, we expressed these Sodalis genes on plasmids in several different E. coli strains that lack the respective homologues. E. coli MC4100ΔdnaK, JW0014-1, and DA16 lack functional DnaK, DnaJ, and GrpE, respectively, and these strains cannot grow at temperatures at or above the elevated temperature of 45°C. To quantify the ability of the Sodalis chaperone proteins to functionally replace their E. coli homologues, we spotted serially diluted cultures of the mutant E. coli strains that express Sodalis genes on plates that were then incubated at permissive and elevated temperatures. All strains grew at the permissive temperature of 30°C (S4 Table). At 45°C, the E. coli mutant strains containing pWKS30, the vector control (empty plasmid), failed to grow. However, E. coli MC4100ΔdnaK expressing Sodalis dnaK survived as well as the parent strain at this elevated temperature (Fig 3). Likewise, E. coli JW0014-1 (ΔdnaJ) and DA16 (ΔgrpE) expressing Sodalis dnaJ or Sodalis grpE, respectively, survived elevated temperature as well as the parent strains (Fig 3).

Fig 3. Sodalis chaperone genes facilitate E. coli survival at elevated temperatures.

Overnight cultures of the indicated E. coli strains grown at 30°C were serial diluted (1:10) six times, and 10 μl of each dilution was spotted on L agar plates. The plates were incubated at 45±1oC for 18-24 hours. In all panels, experimental designations are indicated as the E. coli strain/introduced plasmid (containing the cloned Sodalis gene). All plasmids used are described in S1 Table. Panel A, wt (wild-type E. coli strain MC4100), mut [mutant E. coli strain MC4100ΔdnaK (ΔdnaK)]. Panel B, wt (wild-type E. coli strain BW25113, mut [mutant E. coli strain JW0014-1 (ΔdnaJ)]. Panel C, wt wild-type E. coli strain DA15, mut [mutant E. coli strain DA16 (grpE280)]. The data shown are representative of at least three independent experiments.

We also examined the growth kinetics of E. coli strains expressing Sodalis dnaK or dnakJ by performing growth curves at 30°C and 45±1°C. All strains grew at the permissive temperatures of 30°C (Fig 4), although the dnaK mutant containing the pWKS30 control plasmid grew slightly slower than the wildtype strain containing pWKS30. At 45±1°C, the E. coli dnaK mutant strain containing pWKS30 did not grow. E. coli MC4100ΔdnaK expressing Sodalis dnaK or Sodalis dnaKJ survived as well as the parent strain and control strains with E. coli dnaK (Fig 4). Taken together, these data suggest that the Sodalis DnaKJ/GrpE chaperone system is sufficient for mediating heat shock survival in an E. coli heterologous host strain deficient in these functions.

Fig 4. Growth kinetics of E. coli dnaK mutants containing Sodalis dnaK or dnaK and dnaJ.

Mid-log phase cultures (37°C) of E. coli wt/pWKS30 or mutant E. coli [MC4100ΔdnaK (ΔdnaK)] containing either pWKS30, pJR1, pJR5, or pJS2 (all plasmids used are described in S1 Table) were diluted to an OD600 of 0.06 in L broth containing carbenicillin and incubated at 30°C (top graph) or 45±1°C (bottom graph). Growth was measured via optical density at 600 nm. Each time point represents the mean of three individual experiments, ± the standard deviation.


Ectothermic insects, and the bacterial symbionts that reside within them, are sensitive to temperature shifts in their environment. Bacteria that reside within the gut of hematophagous insects must also deal with a rapid increase in the temperature of their niche when their host consumes vertebrate blood. This study utilized the tsetse fly and its secondary symbiont, Sodalis glossinidius, to examine molecular mechanisms that may mediate thermal stress tolerance in an obligate blood feeding insect/bacterial symbiont model system. Our initial work examined the upper thermal limit for growth and survival of Sodalis maintained in culture and residing endogenously within tsetse. We found that Sodalis do not survive for extended periods of time when exposed to temperatures above 30°C in BHI media or within tsetse at 37°C. This was surprising because, outside of a lab setting, tsetse flies and their Sodalis symbionts reside in sub-Saharan Africa where temperatures are often well above 30°C. Several scenarios may explain this unexpected finding. First, tsetse are crepuscular and thus feed during cooler dawn and dusk periods of the day [40]. Furthermore, although the environmental temperature routinely exceeds 30oC, tsetse seek cooler shade under these conditions [41]. Thus, even in the wild, Sodalis within tsetse might not be exposed to temperatures above 30°C for long periods of time. Although the fly’s body temperature rises during and immediately after feeding to 36°C [24], it likely returns to ambient temperature well within Sodalis’ window of survival. In fact, our results showed that in BHI broth, Sodalis can survive for limited periods of time at 30°C for 24 hours, 32°C for 8 hours, and 37°C for at least 2 hours. Second, the Sodalis used in this study were isolated from tsetse flies reared in a laboratory colony at 25°C for many years. Sodalis from these flies may have lost the ability to survive at higher temperatures, while recent environmental isolates of Sodalis may be more thermo-tolerant. Finally, environmental factors associated with the tsetse fly host (including other symbionts) may increase the upper thermal limit for Sodalis survival, and these factors may be absent from the tsetse colony and BHI growth media.

We discovered that modulating the composition and/or environment of agar plates on which Sodalis are cultured can restore growth at temperatures above 30oC. Specifically, either blood-supplementation of the agar, or growth within hypoxic CampyPaks (5-15% O2), increased the high temperature threshold at which Sodalis could grow. The stimulation of growth by addition of whole blood is likely due to a component of the erythrocytes, and not the serum, as addition of serum alone did not restore growth of Sodalis at 30oC. In addition to hemoglobin, erythrocytes contain high levels of catalase [42, 43], which detoxifies the oxidative stress molecule H2O2 that is generated during cellular metabolism. Thus, catalase-mediated mitigation of oxidative stress in the presence of blood/erythrocytes may allow Sodalis, which does not itself produce catalase [44], to divert more resources to surviving thermal stress situations. Consistent with this hypothesis are the observations that (a) incubation of Sodalis in CampyPaks, which reduce oxidative stress by generating a hypoxic environment, also increased the high temperature threshold at which Sodalis could grow, but addition of hemoglobin did not, and (b) thermal tolerance and oxidative stress are physiologically linked, as increased temperatures can result in an oxidative stress burden on bacteria cells [4547]. A similar phenomenon might be occurring in Sodalis in the fly. Specifically, tsetse’s gut is likely hypoxic, and this environment may increase the thermal limit for Sodalis within the fly. Consistent with this idea, we found that although Sodalis struggled to grow at 30°C on BHI agar, the bacteria grew fine in the tsetse fly gut at this temperature.

DnaKJ/GrpE chaperone systems that enable refolding of proteins during thermal stress have been found in all domains of life, suggesting that selection for maintenance of these systems is strong. An analysis of over 1200 genomes in 2012 showed that dnaK was found in all bacteria, with the exception of two thermophiles isolated from deep sea hydrothermal vents [48]]. More so, the function of DnaK chaperone systems is well characterized in several bacteria, the vast majority of which are pathogens. Comparatively, genomic conservation and physiological function systems is less well studied in symbionts that reside within ectothermic animal hosts. One exception is the aphid symbionts Buchnera aphidicola, Serratia symbiotica and Hamiltonella defensa, the latter two of which facilitate their insect host’s survival in high temperature environments (51). Additionally, dnaJ and grpE are lost or truncated in a few highly reduced genomes from vertically transmitted endosymbionts that share ancient associations with cicadas, mealybugs, and psyllids [48]. Our work herein shows that the dnaKJ locus is conserved between Sodalis and E. coli, including promoter regions and a potential hairpin in the intergenic region between dnaK and dnaJ. Additionally, at the protein level, Sodalis DnaK is 89% identical to its E. coli ortholog, and residues implicated in mediating thermal tolerance are conserved. Thus, the fact that Sodalis DnaK, DnaJ, and GrpE proteins can functionally replace the homologous E. coli proteins and promote growth at elevated temperature is not surprising. However, this finding of complementation is in contrast with DnaK from Buchnera, Borrelia burgdorferi and Vibrio harveyi, which exhibit partial (B. burgdorferi, at 37oC but not 43oC) or no complementation phenotypes when expressed ectopically in E. coli dnaK mutants [31, 49]. Our ability to complement was similar to complementation found with DnaK from Pseudomonas syringae, Agrobacterium tumefaciens, and Brucella ovis [5052]. No apparent correlation exists between evolutionary relatedness of the above-mentioned bacteria to E. coli and the ability of their DnaK proteins to complement an E. coli dnaK mutant.

For bacteria that experience rapid fluctuations in their environment, the DnaK chaperone system may be especially critical. Induction of Sodalis dnaK expression in response to elevated temperature suggests that the bacterium uses DnaK to tolerate heat shock. Retention of these systems in tsetse’s indigenous, enteric symbionts is indicative of their fundamental importance during times of thermal stress, such as that experienced when the fly host consumes vertebrate blood. In support of this theory, the highly reduced (~ 700 kB) genome of tsetse’s obligate mutualist, Wigglesworthia, also encodes a conserved and putatively functional DnaK gene [53] This bacterium resides intracellularly within tsetse’s bacteriome organ [54], which is attached to the fly’s anterior midgut and thus exposed to rapid changes in temperature following consumption of a blood meal. DnaK has also been implicated in protection against other environmental stresses, including acidic conditions, antibiotic resistance and oxidative stress [5557]. Our results suggest that Sodalis requires DnaK to survive growth in vitro, as our attempts to generate mutations in the gene caused cell death. We tried insertions at four different locations in dnaK using Targetron mutagenesis, as well as deletion of dnaK by allelic change. Notably, these mutagenesis techniques have been successfully used to make mutations in numerous other Sodalis genes [5860]. As Sodalis transitions from a free-living lifestyle to a mutualistic one [38], growth outside the tsetse host (on agar plates or in liquid BHI) may generate low-grade oxidative stress that requires DnaK for an appropriate cellular response that is critical for bacterial survival.

Maintenance of thermal tolerance homeostasis is an integral process that underlies successful insect host-bacterial symbiont interactions. In fact, symbiotic relationships are disrupted at elevated temperatures, and in some cases, heat-shock can result in the complete loss of these bacteria (reviewed in [61]). Symbionts that live within obligate hematophagous arthropods experience rapid changes in temperature as their host feeds, and these changes must be quickly mitigated to avoid disruption of epidemiologically relevant physiologies. For example, symbiotic bacteria that reside within arthropod vectors indirectly and directly mediate their host’s vector competency. Tsetse’s obligate symbiont Wigglesworthia [62], and mosquito [63] and tick [64] commensals, are responsible for mediating production of their host’s peritrophic matrix (PM). This structure is a chitinous and proteinaceous ‘sleeve’ that lines the arthropod midgut [65], and in each of the abovementioned vectors, the PM serves as a protective barrier that their respective vertebrate pathogens must circumvent in order to establish midgut infections that are required for transmission to a subsequent vertebrate host. Intriguingly, Sodalis produces chitinase [15] that may exert a two-fold impact trypanosome infection establishment. First, Sodalis chitinolytic activity would likely degrade the structural integrity of tsetse’s PM, thus making it easier for trypanosomes to cross the barrier and establish an infection in the fly’s ectoperitrophic space [66, 67]. This process would result in the accumulation of N-acetyl-D-glucosamine, which would further facilitate trypanosome infection establishment by inhibiting that activity of anti-parasitic tsetse lectins [15]. While these theories have never been experimentally proven, they are correlatively validated by the fact that Sodalis density positively correlates with trypanosome infection prevalence [16, 19, 68]. Finally, symbiotic bacteria from the genera Kosakonia and Chromobacterium, which are found naturally in the midgut of Anopheles gambiae and Aedes aegypti mosquitoes, produce and secrete reactive oxygen intermediates [69], histone deacetylases [70] and aminopeptidases [71] that exert direct anti-Plasmodium and anti-dengue activity.

In conclusion, information about Sodalis’ heat shock response provides insight into bacterial adaptations that allow symbionts residing within the gut of hematophagous arthropods to survive acute environmental stressors, including heat shock that ensues immediately after their host consumes a meal of vertebrate blood. This information increases our understanding of the physiological mechanisms that facilitate maintenance of bacterial symbioses, which are crucial mediators of host fitness and vector competency.

Supporting information

S1 Fig. Non-coding regulatory elements for Sodalis dnaK, dnaK, and grpE genes.

(A) Putative Sodalis promoters for the polycistronic dnaK and dnaJ mRNA and for monocistronic grpE mRNA are shown, based on homology to their E. coli promoters. Start codons are bolded, the Shine-Delgarno sequence is bolded and italicized, and the σ32 binding sites are bolded and underlined. (B) The consensus sequence for the σ32 binding site for E. coli [72]. (C) A potential secondary structure of the RNA corresponding to the dnaKdnaJ intergenic region, generated using RNAfold from the ViennaRNA package [73, 74].


S2 Fig. Comparison of Sodalis and E. coli heat shock chaperone proteins.

Alignment of Sodalis DnaK, DnaJ, and GrpE with homologues from Escherichia coli MG1665 using Clustal Omega ( An asterisk (*) indicates positions that have a single, fully conserved residue. A colon (:) indicates conservation between groups that exhibit strongly similar properties, roughly equivalent to scoring > 0.5 in the Gonnet PAM 250 matrix. A period (.) indicates conservation between groups that exhibit weakly similar properties, roughly equivalent to scoring ≤ 0.5 and > 0 in the Gonnet PAM 250 matrix. For DnaK, the boxed residues indicate a glycine (G) that interacts with GrpE, a glutamine (Q) that binds the unfolded protein substrate and an alanine (A) that is involved in synergistic activation of ATPase by DnaJ [7577]. The overlined residues indicate DnaK amino acids predicted to interact with Mg-ADP [76, 78, 79]. The dashed underline indicates a motif found in DnaK from all gram-negative bacteria that is thought to be essential for ATP-dependent cooperative function with DnaJ and GrpE [80]. The threonine (T) with the dot is required for ATPase activity [81]. For DnaJ, the bracketed residues are conserved residues in the J-domain that interact with DnaK [82, 83]. The underlined residues are zinc-binding motifs that are predicted to bind the unfolded protein substrate [8486]. The G/F region, which may modulate unfolded substrate binding to DnaK, is boxed, and the DIF motifs within this G/F region, which are involved in regulation of chaperone cycling by modulating a step after ATP hydrolysis [87, 88], are overlined.


S3 Fig. Comparison of DnaK proteins from E. coli, Sodalis glossinidius, and other insect symbionts.

Alignment of Sodalis glossinidius DnaK with homologues from Escherichia coli MG1665 and the insect symbionts using Clustal Omaga ( The species corresponding to the protein accession numbers are as follows: WP_074011646.1, Candidatus Sodalis sp. SoCistrobi; KYP97672.1, Sodalis-like endosymbiont of Proechinophthirus fluctus; WP_025244843.1, Candidatus Sodalis pierantonius; WP_067565807.1, Candidatus Doolittlea endobia; WP_067567978.1, Candidatus Hoaglandella endobia; WP_014888228.1, secondary endosymbiont of Ctenarytaina eucalypti; WP_067497883.1, Candidatus Gullanella endobia; WP_067568929.1, Candidatus Mikella endobia; AIN47473.1, Candidatus Baumannia cicadellinicola; WP_014888738.1; secondary endosymbiont of Heteropsylla cubana; WP_083172452.1, secondary endosymbiont of Trabutina mannipara; WP_013975497.1, Candidatus Moranella endobia. An asterisk (*) indicates positions which have a single, fully conserved residue. A colon (:) indicates conservation between groups of strongly similar properties, roughly equivalent to scoring > 0.5 in the Gonnet PAM 250 matrix. A period (.) indicates conservation between groups of weakly similar properties, roughly equivalent to scoring = < 0.5 and > 0 in the Gonnet PAM 250 matrix. The boxed residues indicate a glycine (G) that interacts with GrpE, a glutamine (Q) that binds the unfolded protein substrate and an alanine (A) that has been shown to be involved in synergistic activation of ATPase by DnaJ. The overlined residues indicate DnaK amino acids predicted to interact with Mg-ADP. The dashed underline indicates a motif found in DnaK from all gram-negative bacteria which is thought to be essential for ATP-dependent cooperative function with DnaJ and GrpE. The threonine (T) with the dot is required for ATPase activity.


S3 Table. Sodalis DnaK homology with other insect symbionts.


S4 Table. Sodalis chaperone genes facilitate elevated temperature survival in E. coli.



We gratefully thank Dr. Bernd Bukau at University Heidelberg for supplying us with strain MC4100ΔdnaK. We also thank the students in the Introduction to Biological Thinking class at the University of Richmond in the Fall of 2010 for help with preliminary data acquisition.


  1. 1. Snyder AK, Rio RV. Interwoven biology of the tsetse holobiont. J Bacteriol. 2013;195(19):4322–30. pmid:23836873; PubMed Central PMCID: PMC3807475.
  2. 2. Wang J, Weiss BL, Aksoy S. Tsetse fly microbiota: form and function. Frontiers in cellular and infection microbiology. 2013;3:69. pmid:24195062; PubMed Central PMCID: PMC3810596.
  3. 3. Doudoumis V, Blow F, Saridaki A, Augustinos A, Dyer NA, Goodhead I, et al. Challenging the Wigglesworthia, Sodalis, Wolbachia symbiosis dogma in tsetse flies: Spiroplasma is present in both laboratory and natural populations. Sci Rep. 2017;7(1):4699. pmid:28680117; PubMed Central PMCID: PMC5498494.
  4. 4. Cheng Q, Aksoy S. Tissue trophism, transmission and expression of foreign genes in vivo in midgut symbionts of tsetse flies. Insect Mol Biol. 1999;8(1):125–32. pmid:9927181
  5. 5. Pinnock DE, Hess RT. The occurrence of intracellular rickettsia-like organisms in the tsetse flies, Glossina morsitans, G. fuscipes, G. brevipalpis and G. pallidipes. Acta Trop. 1974;31(1):70–9. pmid:4151039.
  6. 6. Shaw MK, Moloo SK. Comparative study on Rickettsia-like organisms in the midgut epithelial cells of different Glossina species. Parasitology. 1991;102:193–9. pmid:1852486.
  7. 7. Weyda F, Soldan T, Matha V. Rickettsia-like organisms in the tsetse fly Glossina palpalis palpalis. Cytobios. 1995;81:223–8.
  8. 8. Attardo GM, Lohs C, Heddi A, Alam UH, Yildirim S, Aksoy S. Analysis of milk gland structure and function in Glossina morsitans: milk protein production, symbiont populations and fecundity. J Insect Physiol. 2008;54(8):1236–42. pmid:18647605
  9. 9. Denlinger D, Ma W. Dynamics of the pregnancy cycle in the tsetse Glossina morsitans. J Insect Physiol. 1974;20(6):1015–26. pmid:4839338
  10. 10. Maltz MA, Weiss BL, O'Neill M, Wu Y, Aksoy S. OmpA-Mediated Biofilm Formation Is Essential for the Commensal Bacterium Sodalis glossinidius To Colonize the Tsetse Fly Gut. Appl Environ Microbiol. 2012;78(21):7760–8. Epub 2012/09/04. [pii]. pmid:22941073.
  11. 11. Mbewe NJ, Mweempwa C, Guya S, Wamwiri FN. Microbiome frequency and their association with trypanosome infection in male Glossina morsitans centralis of Western Zambia. Vet Parasitol. 2015;211(1-2):93–8. pmid:25983231.
  12. 12. Wamwiri FN, Alam U, Thande PC, Aksoy E, Ngure RM, Aksoy S, et al. Wolbachia, Sodalis and trypanosome co-infections in natural populations of Glossina austeni and Glossina pallidipes. Parasit Vectors. 2013;6(1):232. pmid:23924682; PubMed Central PMCID: PMC3751944.
  13. 13. Dale C, Welburn SC. The endosymbionts of tsetse flies: manipulating host-parasite interactions. Int J Parasitol. 2001;31(5-6):628–31. pmid:11334953.
  14. 14. Moloo SK, Kabata JM, Waweru F, Gooding RH. Selection of susceptible and refractory lines of Glossina morsitans centralis for Trypanosoma congolense infection and their susceptibility to different pathogenic Trypanosoma species. Med Vet Entomol. 1998;12(4):391–8. pmid:9824823.
  15. 15. Welburn SC, Arnold K, Maudlin I, Gooday GW. Rickettsia-like organisms and chitinase production in relation to transmission of trypanosomes by tsetse flies. Parasitology. 1993;107 (Pt 2):141–5. pmid:8414668.
  16. 16. Aksoy E, Telleria EL, Echodu R, Wu Y, Okedi LM, Weiss BL, et al. Analysis of multiple tsetse fly populations in Uganda reveals limited diversity and species-specific gut microbiota. Appl Environ Microbiol. 2014;80(14):4301–12. pmid:24814785; PubMed Central PMCID: PMC4068677.
  17. 17. Farikou O, Thevenon S, Njiokou F, Allal F, Cuny G, Geiger A. Genetic diversity and population structure of the secondary symbiont of tsetse flies, Sodalis glossinidius, in sleeping sickness foci in Cameroon. PLoS neglected tropical diseases. 2011;5(8):e1281. pmid:21886849; PubMed Central PMCID: PMC3160304.
  18. 18. Griffith BC, Weiss BL, Aksoy E, Mireji PO, Auma JE, Wamwiri FN, et al. Analysis of the gut-specific microbiome from field-captured tsetse flies, and its potential relevance to host trypanosome vector competence. BMC microbiology. 2018;18(Suppl 1):146. pmid:30470178; PubMed Central PMCID: PMC6251097.
  19. 19. Soumana IH, Simo G, Njiokou F, Tchicaya B, Abd-Alla AM, Cuny G, et al. The bacterial flora of tsetse fly midgut and its effect on trypanosome transmission. J Invertebr Pathol. 2013;112 Suppl:S89–93. pmid:22841948.
  20. 20. Corbin C, Heyworth ER, Ferrari J, Hurst GD. Heritable symbionts in a world of varying temperature. Heredity (Edinb). 2017;118(1):10–20. Epub 2016/10/06. pmid:27703153; PubMed Central PMCID: PMC5176117.
  21. 21. Mourino S, Osorio CR, Lemos ML. Characterization of heme uptake cluster genes in the fish pathogen Vibrio anguillarum. J Bacteriol. 2004;186(18):6159–67. Epub 2004/09/03. pmid:15342586; PubMed Central PMCID: PMC515166.
  22. 22. Russell JA, Moran NA. Costs and benefits of symbiont infection in aphids: variation among symbionts and across temperatures. Proc Biol Sci. 2006;273(1586):603–10. pmid:16537132; PubMed Central PMCID: PMC1560055.
  23. 23. Benoit JB, Denlinger DL. Bugs battle stress from hot blood. Elife. 2017;6. Epub 2017/11/22. pmid:29157360; PubMed Central PMCID: PMC5697928.
  24. 24. Lahondere C, Lazzari CR. Thermal effect of blood feeding in the telmophagous fly Glossina morsitans morsitans. J Therm Biol. 2015;48:45–50. pmid:25660629.
  25. 25. Bhandari V, Houry WA. Substrate Interaction Networks of the Escherichia coli Chaperones: Trigger Factor, DnaK and GroEL. Adv Exp Med Biol. 2015;883:271–94. pmid:26621473.
  26. 26. Maleki F, Khosravi A, Nasser A, Taghinejad H, Azizian M. Bacterial Heat Shock Protein Activity. J Clin Diagn Res. 2016;10(3):BE01–3. pmid:27134861; PubMed Central PMCID: PMC4843247.
  27. 27. Cowing DW, Bardwell JC, Craig EA, Woolford C, Hendrix RW, Gross CA. Consensus sequence for Escherichia coli heat shock gene promoters. Proc Natl Acad Sci U S A. 1985;82(9):2679–83. pmid:3887408; PubMed Central PMCID: PMC397628.
  28. 28. Grossman AD, Straus DB, Walter WA, Gross CA. Sigma 32 synthesis can regulate the synthesis of heat shock proteins in Escherichia coli. Genes Dev. 1987;1(2):179–84. pmid:3315848.
  29. 29. Brooks JF 2nd, Gyllborg MC, Cronin DC, Quillin SJ, Mallama CA, Foxall R, et al. Global discovery of colonization determinants in the squid symbiont Vibrio fischeri. Proc Natl Acad Sci U S A. 2014;111(48):17284–9. pmid:25404340; PubMed Central PMCID: PMC4260577.
  30. 30. Sato S, Ishikawa H. Structure and expression of the dnaKJ operon of Buchnera, an intracellular symbiotic bacteria of aphid. J Biochem. 1997;122(1):41–8. pmid:9276669.
  31. 31. Tilly K, Hauser R, Campbell J, Ostheimer GJ. Isolation of dnaJ, dnaK, and grpE homologues from Borrelia burgdorferi and complementation of Escherichia coli mutants. Mol Microbiol. 1993;7(3):359–69. pmid:8459764.
  32. 32. Aksoy S. Molecular analysis of the endosymbionts of tsetse flies: 16S rDNA locus and over-expression of a chaperonin. Insect Mol Biol. 1995;4(1):23–9. pmid:7538012.
  33. 33. Kupper M, Gupta SK, Feldhaar H, Gross R. Versatile roles of the chaperonin GroEL in microorganism-insect interactions. FEMS Microbiol Lett. 2014;353(1):1–10. pmid:24460534.
  34. 34. Moloo SK. An artificial feeding technique for Glossina. Parasitology. 1977;63:507–12.
  35. 35. Bursell E, Taylor P. An energy budget for Glossina (Diptera: Glossinidae). B Entomol Res. 1980;70(2):187–96.
  36. 36. Dale C, Jones T, Pontes M. Degenerative evolution and functional diversification of type-III secretion systems in the insect endosymbiont Sodalis glossinidius. Mol Biol Evol. 2005;22(3):758–66. pmid:15574807.
  37. 37. Pontes MH, Smith KL, De Vooght L, Van Den Abbeele J, Dale C. Attenuation of the sensing capabilities of PhoQ in transition to obligate insect-bacterial association. PLoS genetics. 2011;7(11):e1002349. pmid:22072980; PubMed Central PMCID: PMC3207850.
  38. 38. Toh H, Weiss BL, Perkin SA, Yamashita A, Oshima K, Hattori M, et al. Massive genome erosion and functional adaptations provide insights into the symbiotic lifestyle of Sodalis glossinidius in the tsetse host. Genome Res. 2006;16(2):149–56. pmid:16365377.
  39. 39. Guisbert E, Yura T, Rhodius VA, Gross CA. Convergence of molecular, modeling, and systems approaches for an understanding of the Escherichia coli heat shock response. Microbiol Mol Biol Rev. 2008;72(3):545–54. pmid:18772288; PubMed Central PMCID: PMC2546862.
  40. 40. Makumi JN, Green C, Baylis M. Activity patterns in Glossina longipennis: a field study using different sampling methods. Medical and Veterinary Entomology. 1998;12(4):399–406. WOS:000076933100008. pmid:9824824
  41. 41. Pollock JN. Training Manual for Tsetse Control Personnel. In: Pollock JN, editor. Rome: Food and Agricultural Organization of the United Nations 1982.
  42. 42. Agar NS, Sadrzadeh SM, Hallaway PE, Eaton JW. Erythrocyte catalase. A somatic oxidant defense? J Clin Invest. 1986;77(1):319–21. pmid:3944256; PubMed Central PMCID: PMC423343.
  43. 43. Bolton FJ, Coates D, Hutchinson DN. The ability of campylobacter media supplements to neutralize photochemically induced toxicity and hydrogen peroxide. J Appl Bacteriol. 1984;56(1):151–7. pmid:6706882.
  44. 44. Dale C, Maudlin I. Sodalis gen. nov. and Sodalis glossinidius sp. nov., a microaerophilic secondary endosymbiont of the tsetse fly Glossina morsitans morsitans. Int J Syst Bacteriol. 1999;49 Pt 1:267–75. pmid:10028272.
  45. 45. Benov L, Fridovich I. Superoxide dismutase protects against aerobic heat shock in Escherichia coli. J Bacteriol. 1995;177(11):3344–6. pmid:7768839; PubMed Central PMCID: PMC177032.
  46. 46. Garenaux A, Jugiau F, Rama F, de Jonge R, Denis M, Federighi M, et al. Survival of Campylobacter jejuni strains from different origins under oxidative stress conditions: effect of temperature. Curr Microbiol. 2008;56(4):293–7. pmid:18180992.
  47. 47. Privalle CT, Fridovich I. Induction of superoxide dismutase in Escherichia coli by heat shock. Proc Natl Acad Sci U S A. 1987;84(9):2723–6. pmid:3554246; PubMed Central PMCID: PMC304730.
  48. 48. Warnecke T. Loss of the DnaK-DnaJ-GrpE chaperone system among the Aquificales. Mol Biol Evol. 2012;29(11):3485–95. pmid:22683810.
  49. 49. Zmijewski MA, Kwiatkowska JM, Lipinska B. Complementation studies of the DnaK-DnaJ-GrpE chaperone machineries from Vibrio harveyi and Escherichia coli, both in vivo and in vitro. Archives of microbiology. 2004;182(6):436–49. pmid:15448982.
  50. 50. Boshoff A, Hennessy F, Blatch GL. The in vivo and in vitro characterization of DnaK from Agrobacterium tumefaciens RUOR. Protein Expr Purif. 2004;38(2):161–9. pmid:15555931.
  51. 51. Cellier MF, Teyssier J, Nicolas M, Liautard JP, Marti J, Sri Widada J. Cloning and characterization of the Brucella ovis heat shock protein DnaK functionally expressed in Escherichia coli. J Bacteriol. 1992;174(24):8036–42. pmid:1459952; PubMed Central PMCID: PMC207542.
  52. 52. Keith LM, Partridge JE, Bender CL. dnaK and the heat stress response of Pseudomonas syringae pv. glycinea. Mol Plant Microbe Interact. 1999;12(7):563–74. pmid:10478477.
  53. 53. Akman L, Yamashita A, Watanabe H, Oshima K, Shiba T, Hattori M, et al. Genome sequence of the endocellular obligate symbiont of tsetse flies, Wigglesworthia glossinidia. Nat Genet. 2002;32(3):402–7. Epub 2002/09/10. [pii]. pmid:12219091.
  54. 54. Balmand S, Lohs C, Aksoy S, Heddi A. Tissue distribution and transmission routes for the tsetse fly endosymbionts. J Invertebr Pathol. 2013;112 Suppl:S116–22. pmid:22537833; PubMed Central PMCID: PMC3772537.
  55. 55. Hanawa T, Fukuda M, Kawakami H, Hirano H, Kamiya S, Yamamoto T. The Listeria monocytogenes DnaK chaperone is required for stress tolerance and efficient phagocytosis with macrophages. Cell Stress Chaperones. 1999;4(2):118–28. pmid:10547061; PubMed Central PMCID: PMC312926.
  56. 56. Singh VK, Utaida S, Jackson LS, Jayaswal RK, Wilkinson BJ, Chamberlain NR. Role for dnaK locus in tolerance of multiple stresses in Staphylococcus aureus. Microbiology. 2007;153(Pt 9):3162-73. pmid:17768259.
  57. 57. Yamaguchi Y, Tomoyasu T, Takaya A, Morioka M, Yamamoto T. Effects of disruption of heat shock genes on susceptibility of Escherichia coli to fluoroquinolones. BMC microbiology. 2003;3:16. pmid:12911840; PubMed Central PMCID: PMC184496.
  58. 58. Hrusa G, Farmer W, Weiss BL, Applebaum T, Roma JS, Szeto L, et al. TonB-dependent heme iron acquisition in the tsetse fly symbiont Sodalis glossinidius. Appl Environ Microbiol. 2015;81(8):2900–9. pmid:25681181; PubMed Central PMCID: PMC4375324.
  59. 59. Runyen-Janecky LJ, Brown AN, Ott B, Tujuba HG, Rio RV. Regulation of high-affinity iron acquisition homologues in the tsetse fly symbiont, Sodalis glossinidius. J Bacteriol. 2010. Epub 2010/05/25. pmid:20494987.
  60. 60. Smith CL, Weiss BL, Aksoy S, Runyen-Janecky LJ. Characterization of the achromobactin iron acquisition operon in Sodalis glossinidius. Appl Environ Microbiol. 2013;79(9):2872–81. pmid:23435882; PubMed Central PMCID: PMC3623160.
  61. 61. Wernegreen JJ. Mutualism meltdown in insects: bacteria constrain thermal adaptation. Curr Opin Microbiol. 2012;15(3):255–62. pmid:22381679; PubMed Central PMCID: PMC3590105.
  62. 62. Weiss BL, Wang J, Maltz MA, Wu Y, Aksoy S. Trypanosome infection establishment in the tsetse fly gut is influenced by microbiome-regulated host immune barriers. PLoS pathogens. 2013;9(4):e1003318. pmid:23637607; PubMed Central PMCID: PMC3630092.
  63. 63. Rodgers FH, Gendrin M, Wyer CAS, Christophides GK. Microbiota-induced peritrophic matrix regulates midgut homeostasis and prevents systemic infection of malaria vector mosquitoes. PLoS pathogens. 2017;13(5):e1006391. pmid:28545061; PubMed Central PMCID: PMC5448818.
  64. 64. Narasimhan S, Rajeevan N, Liu L, Zhao YO, Heisig J, Pan J, et al. Gut microbiota of the tick vector Ixodes scapularis modulate colonization of the Lyme disease spirochete. Cell Host Microbe. 2014;15(1):58–71. pmid:24439898; PubMed Central PMCID: PMC3905459.
  65. 65. Hegedus D, Erlandson M, Gillott C, Toprak U. New insights into peritrophic matrix synthesis, architecture, and function. Annu Rev Entomol. 2009;54:285–302. pmid:19067633.
  66. 66. Vigneron A, Aksoy E, Weiss BL, Bing X, Zhao X, Awuoche EO, et al. A fine-tuned vector-parasite dialogue in tsetse's cardia determines peritrophic matrix integrity and trypanosome transmission success. PLoS pathogens. 2018;14(4):e1006972. pmid:29614112; PubMed Central PMCID: PMC5898766.
  67. 67. Weiss BL, Savage AF, Griffith BC, Wu Y, Aksoy S. The peritrophic matrix mediates differential infection outcomes in the tsetse fly gut following challenge with commensal, pathogenic, and parasitic microbes. J Immunol. 2014;193(2):773–82. pmid:24913976; PubMed Central PMCID: PMC4107339.
  68. 68. Doudoumis V, Alam U, Aksoy E, Abd-Alla AM, Tsiamis G, Brelsfoard C, et al. Tsetse-Wolbachia symbiosis: comes of age and has great potential for pest and disease control. J Invertebr Pathol. 2013;112 Suppl:S94–103. pmid:22835476; PubMed Central PMCID: PMC3772542.
  69. 69. Cirimotich CM, Dong Y, Clayton AM, Sandiford SL, Souza-Neto JA, Mulenga M, et al. Natural microbe-mediated refractoriness to Plasmodium infection in Anopheles gambiae. Science. 2011;332(6031):855–8. pmid:21566196; PubMed Central PMCID: PMC4154605.
  70. 70. Saraiva RG, Huitt-Roehl CR, Tripathi A, Cheng YQ, Bosch J, Townsend CA, et al. Chromobacterium spp. mediate their anti-Plasmodium activity through secretion of the histone deacetylase inhibitor romidepsin. Sci Rep. 2018;8(1):6176. pmid:29670144; PubMed Central PMCID: PMC5906607.
  71. 71. Saraiva RG, Fang J, Kang S, Anglero-Rodriguez YI, Dong Y, Dimopoulos G. Aminopeptidase secreted by Chromobacterium sp. Panama inhibits dengue virus infection by degrading the E protein. PLoS neglected tropical diseases. 2018;12(4):e0006443. pmid:29694346; PubMed Central PMCID: PMC5937796.
  72. 72. Yura T, Kanemori M, Morita M. The Heat Shock Response: Regulation and Function. In: Storz G, Hengge-Aronis R, editors. Bacterial Stress Responses. Washington, D.C.: American Scoiety for Microbiology Press; 2000. p. 3–18.
  73. 73. Gruber AR, Lorenz R, Bernhart SH, Neubock R, Hofacker IL. The Vienna RNA websuite. Nucleic Acids Res. 2008;36(Web Server issue):W70–4. pmid:18424795; PubMed Central PMCID: PMC2447809.
  74. 74. Lorenz R, Bernhart SH, Honer Zu Siederdissen C, Tafer H, Flamm C, Stadler PF, et al. ViennaRNA Package 2.0. Algorithms Mol Biol. 2011;6:26. pmid:22115189; PubMed Central PMCID: PMC3319429.
  75. 75. Buchberger A, Schroder H, Buttner M, Valencia A, Bukau B. A conserved loop in the ATPase domain of the DnaK chaperone is essential for stable binding of GrpE. Nat Struct Biol. 1994;1(2):95–101. pmid:7656024.
  76. 76. Kamath-Loeb AS, Lu CZ, Suh WC, Lonetto MA, Gross CA. Analysis of three DnaK mutant proteins suggests that progression through the ATPase cycle requires conformational changes. J Biol Chem. 1995;270(50):30051–9. pmid:8530409.
  77. 77. Zhu X, Zhao X, Burkholder WF, Gragerov A, Ogata CM, Gottesman ME, et al. Structural analysis of substrate binding by the molecular chaperone DnaK. Science. 1996;272(5268):1606–14. pmid:8658133.
  78. 78. Buchberger A, Valencia A, McMacken R, Sander C, Bukau B. The chaperone function of DnaK requires the coupling of ATPase activity with substrate binding through residue E171. EMBO J. 1994;13(7):1687–95. pmid:7908876; PubMed Central PMCID: PMC395000.
  79. 79. Holmes KC, Sander C, Valencia A. A new ATP-binding fold in actin, hexokinase and Hsc70. Trends Cell Biol. 1993;3(2):53–9. pmid:14731729.
  80. 80. Sugimoto S, Higashi C, Saruwatari K, Nakayama J, Sonomoto K. A gram-negative characteristic segment in Escherichia coli DnaK is essential for the ATP-dependent cooperative function with the co-chaperones DnaJ and GrpE. FEBS Lett. 2007;581(16):2993–9. pmid:17544398.
  81. 81. McCarty JS, Walker GC. DnaK as a thermometer: threonine-199 is site of autophosphorylation and is critical for ATPase activity. Proc Natl Acad Sci U S A. 1991;88(21):9513–7. pmid:1835085; PubMed Central PMCID: PMC52748.
  82. 82. Qian YQ, Patel D, Hartl FU, McColl DJ. Nuclear magnetic resonance solution structure of the human Hsp40 (HDJ-1) J-domain. J Mol Biol. 1996;260(2):224–35. pmid:8764402.
  83. 83. Szyperski T, Pellecchia M, Wall D, Georgopoulos C, Wuthrich K. NMR structure determination of the Escherichia coli DnaJ molecular chaperone: secondary structure and backbone fold of the N-terminal region (residues 2-108) containing the highly conserved J domain. Proc Natl Acad Sci U S A. 1994;91(24):11343–7. pmid:7972061; PubMed Central PMCID: PMC45227.
  84. 84. Banecki B, Liberek K, Wall D, Wawrzynow A, Georgopoulos C, Bertoli E, et al. Structure-function analysis of the zinc finger region of the DnaJ molecular chaperone. J Biol Chem. 1996;271(25):14840–8. pmid:8662861.
  85. 85. Szabo A, Korszun R, Hartl FU, Flanagan J. A zinc finger-like domain of the molecular chaperone DnaJ is involved in binding to denatured protein substrates. EMBO J. 1996;15(2):408–17. pmid:8617216; PubMed Central PMCID: PMC449956.
  86. 86. Pellecchia M, Szyperski T, Wall D, Georgopoulos C, Wuthrich K. NMR structure of the J-domain and the Gly/Phe-rich region of the Escherichia coli DnaJ chaperone. J Mol Biol. 1996;260(2):236–50. pmid:8764403.
  87. 87. Cajo GC, Horne BE, Kelley WL, Schwager F, Georgopoulos C, Genevaux P. The role of the DIF motif of the DnaJ (Hsp40) co-chaperone in the regulation of the DnaK (Hsp70) chaperone cycle. J Biol Chem. 2006;281(18):12436–44. pmid:16533811.
  88. 88. Wall D, Zylicz M, Georgopoulos C. The conserved G/F motif of the DnaJ chaperone is necessary for the activation of the substrate binding properties of the DnaK chaperone. J Biol Chem. 1995;270(5):2139–44. pmid:7836443.