Delayed accumulation of intestinal coliform bacteria enhances life span and stress resistance in Caenorhabditis elegans fed respiratory deficient E. coli
- 7.5k Downloads
Studies with the nematode model Caenorhabditis elegans have identified conserved biochemical pathways that act to modulate life span. Life span can also be influenced by the composition of the intestinal microbiome, and C. elegans life span can be dramatically influenced by its diet of Escherichia coli. Although C. elegans is typically fed the standard OP50 strain of E. coli, nematodes fed E. coli strains rendered respiratory deficient, either due to a lack coenzyme Q or the absence of ATP synthase, show significant life span extension. Here we explore the mechanisms accounting for the enhanced nematode life span in response to these diets.
The intestinal load of E. coli was monitored by determination of worm-associated colony forming units (cfu/worm or coliform counts) as a function of age. The presence of GFP-expressing E. coli in the worm intestine was also monitored by fluorescence microscopy. Worms fed the standard OP50 E. coli strain have high cfu and GFP-labeled bacteria in their guts at the L4 larval stage, and show saturated coliform counts by day five of adulthood. In contrast, nematodes fed diets of respiratory deficient E. coli lacking coenzyme Q lived significantly longer and failed to accumulate bacteria within the lumen at early ages. Animals fed bacteria deficient in complex V showed intermediate coliform numbers and were not quite as long-lived. The results indicate that respiratory deficient Q-less E. coli are effectively degraded in the early adult worm, either at the pharynx or within the intestine, and do not accumulate in the intestinal tract until day ten of adulthood.
The findings of this study suggest that the nematodes fed the respiratory deficient E. coli diet live longer because the delay in bacterial colonization of the gut subjects the worms to less stress compared to worms fed the OP50 E. coli diet. This work suggests that bacterial respiration can act as a virulence factor, influencing the ability of bacteria to colonize and subsequently harm the animal host. Respiratory deficient bacteria may pose a useful model for probing probiotic relationships within the gut microbiome in higher organisms.
KeywordsAging Bacterial colonization Coenzyme Q Gut microbiome Intestine Life span Pharynx Probiotic Respiration
Caenorhabditis Genetics Center
The digestive tracts of living systems, from nematodes to humans, contain a zoo of microorganisms. Many of these microbiota fill a required role for the host. The microbiota in human gastrointestinal systems produce folate and vitamin K, break down excess sugars and fibers, and help activate certain medications [1, 2]. However, digestive tracts also play host to various bacteria associated with pathophysiological states. Ulcerative colitis, diabetes mellitus, and irritable bowel syndrome are just a few of the diseases influenced by intestinal microbiota .
Microorganisms of the intestinal tract have been shown to influence the aging process. Metchnikoff suggested that the longevity of Bulgarians was attributed to their consumption of lactic acid generating bacteria in yogurts . Although the composition of the intestinal microbiome seems to be unique to each individual , there are common trends when the gut microbiome of babies is compared across diverse cultures . Some studies have shown certain age-related diseases can be prevented or ameliorated with the use of certain microorganisms .
Model organisms can be utilized as a first step in assessing the relationship between longevity and the gut microbiome. Altering gut microorganism composition can influence the aging process in model systems in a safe and effective manner [7, 8]. Mice fed diets supplemented with Lactobacillus as a probiotic not only showed no pathogenic response, but also lived longer than littermates on a standard diet .
C. elegans is routinely maintained on the standard lab E. coli strain OP50. Wild-type (N2) worms fed this diet live an average of two weeks , and recapitulate many of the aging-related changes observed in humans. Old worms show muscular disorganization, diminished movement, and accumulate the aging-related pigment lipofuscin [11, 12]. Worms fed OP50 show an accumulation of bacteria in the pharynx and gut as they age [13, 14, 15] and old nematodes appear constipated . C. elegans fed diets of either Lactobacillus or Bifidobacterium were long-lived and more resistant to the enteropathogen Salmonella enterica as compared to worms fed the standard OP50 E. coli lab diet .
Feeding worms a diet of GD1 E. coli deficient in coenzyme Q (ubiquinone or Q) leads to an increased life span without a cost to fertility [17, 18]. Q is an essential lipid component of the electron transport chain and is required for respiration-dependent energy production. The life span increase of nematodes fed a GD1 Q-less E. coli diet is not due to a lack of Q, because supplying a water-soluble formulation of Q that is effectively assimilated by the worm does not revert the life span of the GD1-fed animals to that observed in OP50-fed animals . In addition, worms fed E. coli mutant strains with defects in ATP synthase (1100bc or AN120) lived longer than worms fed OP50 . This implied that the respiratory status of the bacteria was a crucial factor in the life span of the worms fed these diets.
The relationship between respiration in the E. coli diet and the survival of the worms fed these diets identifies Q and ATP synthase as potential virulence factors. A virulence factor is any process, structure or metabolite required by a microorganism to be pathogenic to its host . In this study we show that loss of respiration in E. coli yields delayed gut colonization and improved worm survival. Indeed, in young animals, few respiratory deficient E. coli are detected on the posterior side of the pharynx. Worms fed a mixture of Q-replete and Q-deficient E. coli show intermediate life span extension, indicating that the degree of bacterial colonization of the gut may be dose dependent. We hypothesize that decreased or delayed gut colonization confers a survival advantage to animals fed the Q-deficient E. coli by diminishing or delaying stress due to high numbers of coliform bacteria. C. elegans fed respiratory-deficient E. coli diets serves as a model for characterizing the effects of anti-aging probiotic therapies.
The GD1-mediated life span extension is independent of dietary restriction or worm Q content
Worms fed a mixture of GD1 and rescued GD1 show an intermediate life span extension
Statistical analyses of life spans
Strain, food, treatment
mean ± s.d. (dy)
% change in mean life span from control
N2, OP50 a
15 ± 4
31 ± 5
N2, OP50 b (Adult)
18 ± 3
30 ± 5
16 ± 3
27 ± 6
N2, GD1::pAHG, – UV c
18 ± 4
16 ± 4
N2, GD1::pAHG, + UVc
20 ± 3
N2, GD1::pBSK, + UVc
21 ± 3
N2, GD1::pAHG only d
23 ± 3
N2, GD1::pBSK onlyd
30 ± 6
25 ± 4
N2, OP50 e
19 ± 5
26 ± 8
15 ± 6
30 ± 12
16 ± 5
33 ± 18
N2, OP50 f (Adult)
16 ± 4
28 ± 7
8 ± 3
34 ± 8
Worms fed GD1 are more thermotolerant and resistant to juglone treatment
Excreted components present in GD1 E. colispent media are not responsible for life span extension
To determine if the excreted D-lactic acid (or other fermentation products) present in GD1 spent media is responsible for the increased life span in worms fed this diet, we performed media swap experiments. Actively respiring rescued GD1 cells containing the ubiG gene on a plasmid (GD1:pAHG) were suspended in either their own spent media or the spent media from non-rescued GD1 cells (GD1:pBSK). Surprisingly, worms fed the GD1:pAHG cells suspended in the D-lactic acid rich spent media from GD1 cells, lived shorter lives than worms fed GD1:pAHG cells suspended in their own spent media (Figure 5B, Table 1). A separate cohort of each plate type was subjected to UV-treatment in order to prevent cells from metabolizing the D-lactic acid in the spent media. As shown in Figure 5B, worms do not display a difference in survival when fed UV-treated GD1:pAHG cells suspended in either type of spent medium. Both results indicate that the excreted components present in GD1 E. coli spent media are not responsible for life span extension.
Bacterial proliferation in the gut has been implicated as a major contributor of mortality in the worm [14, 32]. We speculated that the respiring cells suspended in spent media containing large amounts of D-lactic acid were converting this fermentative product into energy-rich metabolites, fueling proliferation and other cellular functions. To test whether the D-lactic acid in the spent media does supply fuel for growth, we suspended overnight cultures of GD1:pAHG cells in either their own spent media, LB media or the spent media from GD1 cells. We found that the cells provided the GD1 spent media grew nearly as well as cells in LB media, whereas cells suspended in their own spent media showed negligible growth (Figure 5C). These results suggested that respiring E. coli cells utilize D-lactic acid and other metabolites present in the spent media as fuel for proliferation. Under these conditions, the utilization of D-lactic acid has a negative impact on worm life span (Figure 5B).
Q deficient E. coli replicate more slowly than wild-type or ATP synthase mutant E. coli
E. colideficient in respiration show lower colonization of the worm gut during early- to mid-adulthood
At day 5 of adulthood, worms fed the ATP synthase deficient E. coli AN120 strain display an intermediate degree of colonization of the intestine as compared to either OP50-fed worms or the AN180 parental strain (Figure 7B). Interestingly, worms fed AN180 displayed a diminished gut infiltration pattern as compared to OP50 at day two of worm adulthood (Figure 7A and B), despite growing to a thicker density on plates (data not shown). In contrast, from day five of adulthood onward, worms fed AN180 have intestinal GFP patterns identical to OP50-fed worms, indicating that the lag of AN180 infiltration occurs only during the early stage of worm adulthood (Figure 7A and B).
Worms fed OP50 to the tenth day of adulthood have distended intestinal lumens filled with GFP-labeled bacteria. Severe anatomical alterations of the gut deviating from the organ’s previously linear shape are prevalent (Figure 7A). Nonetheless, the degree of bacterial infiltration of the gut increased only slightly compared to day five animals (Figure 7B). By day 10, GD1-fed worms show appreciable amounts of gut bacteria-GFP fluorescence, yet the intestine is still not noticeably distended (Figures 7A and B). In contrast, 10 day-old worms fed AN120 accumulate gut bacteria-GFP fluorescence and acquire the distended gut appearance of worms fed OP50 (Figure 7A and B, and Additional file 4). By day 14 of adulthood all worms have large portions of the gut distended due to bacterial accumulation, regardless of the diet (Figure 7A). Every animal assayed at day 14 demonstrates intestinal accumulation of E. coli (Figure 7B). These results suggest that early accumulation of bacteria in the nematode gut is linked to a shorter nematode life span.
Worms fed GD1 have decreased coliform counts
Interestingly, the cfu/worm in C. elegans fed AN120 were intermediate as compared to OP50, AN180, or GD1, particularly at days 2 and 5 of adulthood (Figure 8). However, day 10 and 14 adult animals fed AN120 have similar coliform counts as compared to either OP50 or AN180 fed worms (Figure 8).
The life span of C. elegans fed diets of respiratory deficient E. coli is significantly enhanced as compared to C. elegans fed the standard lab diet of OP50 E. coli (Figures 1 and 2, Table 1) and [17, 18]. These benefits are not confined to long-term survival, because animals fed the GD1 bacterial strain fare better than worms fed OP50 during short-term stress assays such as exposure to the oxidative agent juglone or to high-temperature (Figure 4). The E. coli respiratory deficiency, due to either the lack of Q or a deficiency in complex V, mediates worm life span extension and increased stress resistance independent of dietary restriction or the worm Q content. Worms fed the standard OP50 E. coli diet have distended guts packed with E. coli and show maximal coliform counts (cfu/worm) by day five of adulthood. However, worms fed the Q-less GD1 E. coli show delayed gut colonization and coliform counts fail to reach maximal levels even by day 14. The findings reported here suggest that the delayed replication of respiratory deficient E. coli in the gut lumen confers a survival benefit to the animal that correlates with the longer worm life span and enhanced stress resistance.
A recent study has suggested that the degree of bacterial colonization of the intestine at day two of C. elegans adulthood can be utilized as a predictor of subsequent worm survival 6 – 24 days thereafter . We have found that this predictive window can be extended to the fifth day of adulthood. It has been previously shown that worms fed OP50 or AN180 have similar life spans . Coliform counts (cfu/worm) in animals fed these diets are similar (Figure 8) when assayed at the L4 larval stage and throughout adulthood. In contrast, worms fed the ATP synthase defective E. coli strain AN120 yield coliform counts intermediate to OP50 and GD1 until day ten, when the values become similar to those of OP50-fed animals (Figure 8). Similarly, coliform counts from GD1-fed worms are significantly lower than worms fed any of the other diets at day two, five, or ten of adulthood (Figure 8). These findings suggest that the coliform counts at days two and five are predictive of the enhanced life span in worms fed these diets.
What accounts for the dramatically low coliform counts in the GD1-fed animals? It seems likely that the pharynx, which is responsible for grinding the food taken up by the worm, efficiently breaks down the Q-deficient E. coli. This degradation could exert an “abiotic” condition in the guts of animals fed this diet. Subsequently, GD1-fed worms begin accumulating bacteria in their guts by day ten of adulthood (Figures 7A, 7B, and 8). The transition from mid to late adulthood marks a shift in pharyngeal function [13, 14]. Animals become plagued by the effects of sarcopenia, or muscle wasting, as they age . The pharynx muscle declines in pumping activity and shows increasing tissue disorder [13, 14]. Indeed, the pharyngeal pumping rate of worms significantly diminishes between early and mid-adulthood . The intestinal colonization patterns of phm 2 mutant worms, which have a poorly functioning pharynx, indicate that large amounts of bacteria pass through the grinder intact. These animals have life spans considerably shorter than wild-type worms, and concomitantly higher numbers of E. coli in the gut lumen . Interestingly, bacteria are considered to play only a minor role in the decline of this organ, implying that degeneration of the pharynx was due predominantly to the effects of long-term pumping . Feeding C. elegans pathogenic bacteria such as Salmonella or Serratia marcescens degrades performance of the pharyngeal grinder and allows early passage of bacterial cells to the worm intestine [34, 35]. Our results suggest that the type of E. coli diet can profoundly alter the “functional aging” of the pharynx. We speculate that the Q-less E. coli membranes may be especially fragile when subjected to the worm pharyngeal grinder due to the absence of Q, which normally serves to maintain membrane stability by acting as a crucial membrane chain-terminating antioxidant . Taken together, these findings underscore the importance of efficient bacterial degradation, as the number of intact bacteria that make it past the pharyngeal grinder clearly impact worm survival.
Replicating bacteria in the gut have already been implicated as a main contributor of worm death . Worms fed either UV-irradiated or antibiotic-treated OP50 had increased survival [14, 18, 37]. Similarly, C. elegans exposed to UV-irradiated Enterica faecalis or Salmonella displayed greater survival than animals fed viable cells of these pathogenic strains [38, 39]. However, worms fed UV-irradiated GD1 E. coli exhibited shorter life span than worms fed untreated GD1 . We have observed enhanced susceptibility of GD1 E. coli to UV treatment. We speculate that the UV-treatment of GD1 as performed previously  actually represents a vast overdose of that required for cell killing, and may result in a toxic food that fails to support larval development (data not shown). Alternatively, it is possible that worms recognize metabolites produced by GD1 cells, similarly to those produced by OP50, and respond through up-regulation of antimicrobial genes. Thus, GD1 cells that are able to reside within the gut lumen may act to elicit different worm signaling pathways that control innate immunity and the expression of antimicrobial genes such as lys 8. In our study, the delay in E. coli accumulation of the gut in worms fed GD1 confers a survival advantage in the animal, and it will be important to determine whether the GD1 diet-mediated longevity effects can be attributed to enhanced intestinal immunity through known signaling pathways . The diminished proliferative capacity of the Q-deficient E. coli (Figure 6) serves to retard the level of infection in the gut once intact bacterial cells settle within the lumen of the aged worm. Therefore, we propose that both Q and ATP synthase function be considered virulence factors.
Both Q and ATP synthase serve essential functions in respiratory metabolism. A growing body of evidence suggests that bacterial pathogens within the gastrointestinal tract must sense oxygen availability (or lack thereof) and their metabolic adaptation to the host environment plays a key role in the expression of virulence factors and in modulating host responses . In E. coli ArcB senses oxygen availability via the quinone redox status (Q/QH2 and menaquinone/menaquinol) and tunes aerobic and anaerobic respiratory metabolism through its phosphorylation of ArcA . ArcA functions as a transcriptional regulator of operons involved in respiratory and fermentative metabolism; ArcA plays a role in virulence in a wide variety of pathogenic bacteria in animals and humans including the enteric pathogens Vibrio cholerae and Shigella flexneri. Mutations in genes encoding respiratory chain complexes also identify components in pathogens essential for virulence. Rat lung fibroblasts exposed to Shigella flexneri with mutations in the cytochrome bd oxidase had lower numbers of plaques than fibroblasts infected with the wild-type parental strain . Brucella abortus, a zoological pathogen that causes spontaneous abortions in cattle, showed attenuated virulence against murine macrophages after the cytochrome bd oxidase gene was disrupted . Two examples directly underscore the relationship between respiration, proliferation and pathogenicity. Burkholderia cenocepacia mutants lacking a functional phenylacetic acid catabolism pathway, which degrades aromatic compounds and shunts electrons into the TCA cycle, grow slowly and are less virulent to C. elegans than wild-type B. cenocepacia. Bae and colleagues fed C. elegans mutated Staphylococcus aureus generated in a random disruption screen and found that disruption mutants in various TCA cycle genes showed attenuated killing activity . Taken together, the findings presented here and in other model systems identify respiration and energy production as important virulence factors.
Our findings indicate that excreted components present in GD1 E. coli spent media are not responsible for worm life span extension. GD1 excreted large amounts of D-lactic acid into its media during growth (Figure 5A). The E. coli ubiA mutant, deficient in a different Q biosynthetic reaction, also accumulates large amounts of D-lactate under normoxic conditions . Intriguingly, consumption of lactic acid is beneficial in a variety of organisms. Ikeda and colleagues showed that worms lived longer and were more resistant to Salmonella enterica infection when fed the D-lactic-acid producing bacteria Bifidobacterium sp. or Lactobacillus sp., although whether this was due to the lactic acid itself was not shown . Feeding heat-killed species of Lactobacilli increased worm life span, implying that the lactate in these cultures was responsible for the improved survival . However, these results are difficult to interpret because the authors compared killed Lactobacilli to living OP50. It is crucial to consider the stereoisomer of lactic acid provided during these analyses. E. coli produces D-lactic acid under hypoxic conditions , whereas C. elegans lactic acid dehydrogenase is considered specific for the L-stereoisomer . Thus, the worm is incapable of converting the D-lactic acid produced by the bacteria into pyruvate. These considerations clarify the results of the spent media/mixing experiment, because while worms cannot utilize the D-lactic acid present in the spent medium of GD1 cultures, rescued GD1 E. coli are able to utilize the D-lactic acid (Figure 5B and 5C). For this reason, the D-lactic acid present in the spent media had no effect on C. elegans life span unless it was provided in combination with respiratory competent E. coli, in which case it led to more bacterial proliferation and a shorter worm life span.
It is becoming clear that certain pathological and aging-related disorders are related to the composition of the intestinal microflora . The use of beneficial bacteria to influence the health status of humans is quickly becoming a viable therapeutic option. Premature infants given Lactobacilli soon after birth show significantly decreased incidents of necrotizing enterocolitis . Probiotic therapies have an anti-cancer effect in human patients , while changes in intestinal microbiota composition were associated with the decreased onset of intestinal tumors in the cancer prone ApcMin mouse strain . Mice fed Bifidobacterium animalis subspecies lactis lived longer than littermates fed a control diet and showed diminished gut inflammation . Fruit flies require certain bacteria in their guts for healthy metabolism . Probiotic interventions have yielded promising results in worms . A recent study showed that the folate status of the gut microbiome may slow C. elegans aging . In the presence of tetracycline, the worms assayed in our study responded well to a mixed diet composed of Q-replete and Q-deficient E. coli (Figure 2), indicating that the benefit of the GD1 diet takes effect even in the presence of respiratory-competent E. coli.
In summary, our study argues that E. coli respiration is a virulence factor of OP50 E. coli, the standard lab diet of C. elegans. The decreased coliform counts present in worms fed respiratory deficient E. coli may manifest in at least two ways: (1) the lack of Q increases the tendency of the pharyngeal grinder to break apart the E. coli GD1 cells; (2) the respiratory deficiency of both the Q-less and ATP synthase mutants may render them less able to colonize the gut once the intact bacteria have infiltrated. The delayed accumulation and diminished presence of bacteria in the gut of the animal attenuates the stress encountered by the aging animals, yielding longer-lived worms
We show that the respiratory deficient E. coli diet imparts not only longer life span, but also increased resistance to thermal stress and juglone treatment. The longevity observed is independent of the worm Q content and dietary restriction. We provide evidence that the decreased accumulation of respiratory deficient bacteria in the worm intestine is responsible for the increased longevity observed in C. elegans. The lack of Q in particular makes the bacteria more susceptible to degradation at the worm’s pharynx. In summary, we put forward the idea that respiration is a virulence factor that has a profound effect on the ability of E. coli to colonize and harm its host.
C. elegansstrain and growth conditions
C. elegans and E. coli strains used in this study
skn-1(zu169) IV/nT1 [unc?(n754) let?] (IV;V)
argE3, thi-1, str R , uncA401
argE3, thi-1, str R
Growth of E. coli
Nematode diets consisted of E. coli strains listed in Table 2. E. coli were cultured in LB medium with the designated antibiotic and incubated overnight at 37°C with shaking at 250 rpm. E. coli cells were then harvested and seeded onto NGM plates containing the stated antibiotic. OP50-1 E. coli carrying an integrated streptomycin resistance gene (CGC) were cultured in the presence of streptomycin (250 μg/mL final concentration). GD1 E. coli, a Q-less strain harboring an insertion in the ubiG gene (ubiG::Kan, zei::Tn10dTet) , were cultured in the presence of kanamycin (100 μg/mL final concentration).
GD1:pAHG harbors a wild-type copy of the E. coli ubiG gene on a multicopy plasmid (pAHG) . pBluescript (pBSK; Fermentas) was used as an empty vector control. Both GD1:pAHG and GD1:pBSK cells were grown overnight in LB media containing 100 μg/mL ampicillin.
The ATP synthase deficient E. coli strain AN120 and the parent strain AN180 were previously described . Cultures of AN120 and AN180 were grown overnight in LB medium.
OP50 containing the pFVP25.1 plasmid with the GFP marker was acquired from the Caenorhabditis Genetics Center. GD1, AN180 and AN120 E. coli were also transformed with this plasmid. All E. coli strains carrying the pFVP25.1 plasmid were cultured in LB containing 100 μg/mL ampicillin and seeded to NGM plates containing 100 μg/mL ampicillin as described above.
Determination of C. eleganstotal life span and adult life span
To determine C. elegans total life span (defined as the number of days from hatching until death), N2, CFC1005 and CFC315 gravid adults were hypochlorite lysed and eggs transferred to NGM plates containing the designated E. coli diet. Two days after hatching coq-3 homozygous mutant worms were selected and transferred to plates containing the designated diet. N2 worms were similarly treated. A total of five or six plates per condition were used (20 worms per plate). Worms were scored for survival and moved to new plates every day for the first six days, then every four days thereafter while scoring for survival every two days. Worms that responded to being gently prodded with a platinum wire by moving or pharyngeal pumping were counted as alive. Worms with internally hatched larvae, an extruded vulva, or that escaped were censored from the total count. One-way ANOVA analyses of life spans were performed with StatView 5.0.1 (SAS, CA) software at a significance level of 0.05. Similar results were attained when data were subjected to Kaplan-Meier Test at a 0.05 significance level. Maximum life span was calculated from the mean of the top 10% longest lived worms, for each condition.
To determine C. elegans adult life span, N2, CFC315 and EU35 heterozygous gravid adults were hypochlorite lysed and eggs placed on NGM plates containing fresh OP50. After reaching the L4 larval stage, N2, coq-3(ok506) –/ – and skn-1(zu169) –/ – L4 larvae were transferred to separate plates containing either OP50 or GD1 E. coli, and the life span determined as described above.
Media swap and UV-treatment of GD1:pAHG E. coli
GD1:pAHG and GD1:pBSK cells were grown as described above. The cells were pelleted, the spent media was removed and kept on ice, and the GD1:pBSK cells were discarded. An equal volume of GD1:pAHG cells were resuspended in either their own spent media or the spent media of the GD1:pBSK cells. These suspensions were then seeded onto regular NGM plates, allowed to dry at room temperature, and stored at 4°C until use. Half of the plates containing GD1:pAHG cells in GD1:pAHG spent media and half of the plates containing GD1:pAHG cells in GD1:pBSK spent media were UV-irradiated for 10 minutes at 365 nm on high setting with a Fluorchem2 imaging apparatus (Alpha Innotech, CA). N2 hatchlings were fed OP50 until the L4 larval stage, and then transferred to plates containing one of the designated diets: GD1:pAHG E. coli cells suspended in spent media obtained from cultures of either GD1:pAHG or GD1:pBSK; alternatively these two types of diets were first subjected to UV irradiation prior to the transfer of L4 larvae. Adult life span determinations were performed as described above.
Preparation of mixed E. colidiets
The GD1:pAHG and GD1:pBSK cells were grown as described above and the optical densities (A600nm) were adjusted to 6.0. One cohort of each cell type was seeded onto NGM plates containing 12 μg/mL tetracycline. Another cohort of GD1:pAHG and GD1:pBSK at an optical density of 6.0 (A600) cells were combined at equal volumes, mixed well and seeded onto NGM plates containing 12 μg/mL tetracycline. Wild-type worms were hypochlorite lysed, transferred to NGM plates and fed OP50 as hatchlings. The L4 larvae were transferred as described above onto plates bearing one of three diets: GD1:pAHG cells only, GD1:pBSK cells only or an equal mix of GD1:pAHG and GD1:pBSK cells. Adult life span determinations were performed as described above.
Measurement of D-lactic acid
OP50, GD1, GD1:pAHG and GD1:pBSK cells were grown overnight as described above. The cells were pelleted, the spent media was removed and saved on ice. Levels of D-lactic acid in the spent media were assayed using the Enzychrom D-lactate Assay Kit (BioAssay System Co., Hayward, CA), per the manufacturer’s instructions with an uQuant plate reader at 560 nm (Bio-Tec Instruments Inc., VT). The GD1 and GD1:pBSK spent media were diluted 1:10 with LB. One-way ANOVA analyses were performed with StatView 5.0.1 (SAS, CA) software at a significance level of 0.05, comparing all groups to D-lactic acid levels in OP50 spent media.
E. coligrowth determination
OP50:pFVP25.1, GD1:pFVP25.1, the ATP synthase deficient E. coli strain AN120:pFVP25.1 and its parent strain AN180:pFVP25.1 were grown overnight in LB media containing 100 μg/mL ampicillin. Optical densities were adjusted to 0.1 with LB media, and antibiotic was added for each strain. Bacteria were grown (37°C, 250 rpm) and the cell density was monitored over time by monitoring absorbance at 600 nm with a Shimadzu UV-160 spectrophotometer (Shimadzu, El Cajon, CA). One-way ANOVA analyses were performed with StatView 5.0.1 (SAS, CA) software at a significance level of 0.05, comparing optical density (A600 nm) of all groups versus OP50.
E. coligrowth determination in spent media
GD1:pAHG and GD1:pBSK cells were cultured overnight as described above. The cells were pelleted and the spent media saved on ice. The GD1:pAHG cells were diluted to an optical density of 0.1 in either LB media, spent media from GD1:pBSK cultures, or spent media from GD1:pAHG cultures. Absorbance (600 nm) was determined after 23 h of incubation. One-way ANOVA analyses were performed with StatView 5.0.1 (SAS, CA) software at a significance level of 0.05.
Determination of E. colicell size
OP50 and GD1 cells were grown as described above. Cells were placed onto glass slides and briefly heat fixed. The cells were DIC-imaged and photographed with a Deltavision Spectris Deconvolution Microscope system (Applied Precision). Linear measurements of cells were determined with the linear measurement tool. Fifteen cells per condition were measured. Student’s t-test at a significance level of 0.05 was used to analyze differences in size between the two strains.
Thermal tolerance assay
Gravid wild-type worms were hypochlorite lysed and transferred to NGM plates containing either OP50 or GD1. Ten L4 larvae per plate (three plates were used for each condition) were subjected to 35°C heat stress and monitored for survival until all the worms on OP50-seeded plates were exterminated. Survival was assayed by gently prodding with a platinum wire. Dead worms were removed. The assay was conducted four times. Student’s t-test at each time point was used to assess differences of survival at a significance level of p < 0.05.
Juglone survival assay
Gravid wild-type worms were hypochlorite lysed and eggs transferred to NGM plates containing either OP50 or GD1. Approximately 30 L4 worms were then placed in a 30 μL drop of S-media containing 250 μM juglone (Sigma) from a 12 mM stock solution in 100% ethanol. A drop of S-media containing an equal amount of alcohol was used as a vehicle control. The worms were maintained in the drop for 20 min and washed off the slide with 100 μL S-media onto NGM plates containing OP50. Worms were scored for survival 18 hours later.
For bacterial juglone survival assays, OP50 and GD1 were grown overnight in their respective media containing antibiotics. Cultures were diluted to 1.0 OD600 nm in water, and resuspended in either 125 μM juglone or equal amounts of ethanol as vehicle control. The cells were incubated at 37°C with aeration (250 rpm) and at the indicated time points 3 μL aliquots were spotted onto LB plates containing the respective antibiotic in 1/10 dilutions. Plates were incubated at 37°C for 12 to 16 hours. The assay was conducted three times.
Determination of coliform counts
Gravid wild-type worms were hypochlorite lysed onto NGM plates containing OP50:pFVP25.1, GD1:pFVP25.1, AN120:pFVP25.1 or AN180:pFVP25.1. The hatchlings were fed the designated diets and collected at the following stages: L4, two-, five-, ten-, or fourteen-days of adulthood. Five worms from each condition were washed in 5 μL of S-media with 0.1% Triton X-100 on a foodless NGM plate for 30 s. The worms were washed four times in total and then placed in a 0.5 mL microcentrifuge tube containing 20 μL of the S-media with 0.1% Triton X-100. Worms were mechanically disrupted with a micro-pestle (Sigma) for 200 strokes. The micro-pestle was placed in a 1.5 mL Eppendorf tube containing 100 μL S-media for 30 s, and the contents of the two tubes were combined. The contents of the tube were mixed well and spread onto plates containing 100 μg/mL ampicillin. Serial dilutions (1:1,000, 1:10,000 and 1:100,000) were prepared from worm lysates derived from the OP50- and AN180-diet conditions at the day two, five, ten, and fourteen adult time points. Serial dilutions (1:100, 1:1,000, and 1:10,000) were prepared from worm lysates derived from the GD1- and AN120-diet conditions at the day five, ten, and 14 adult time points. Colony forming units were determined the following day. In order to assay whether the micro-pestle mediated lysis of the worms affected the viability of the bacteria, an equal number of either OP50 or GD1 cells were subjected to mechanical disruption and the cfu quanitified in an identical fashion except that worms were omitted. The process of mechanical disruption did not affect the viability of either the OP50 or GD1 cells (data not shown). One-way ANOVA analyses were performed with StatView 5.0.1 (SAS, CA) software at a significance level of 0.05, comparing all conditions to OP50 fed worms at each indicated time point.
Fluorescence microscopy and intestinal infiltration assay
To monitor bacterial proliferation within animals, synchronized N2 embryos were extracted from gravid adults following hypochlorite treatment and cultivated on OP50:pFVP25.1, GD1:pFVP25.1, AN120:pFVP25.1 or AN180:pFVP25.1 bacterial lawns on NGM plates containing 100 μg/mL ampicillin. Adult animals were moved to new plates every two days to prevent larval contamination. For imaging, L4 larvae and day two, five, ten, and fourteen adult nematodes were washed three times for 30 s in 30 μL M9, then placed onto slides prepared with fresh 2% agar pads. Worms were anesthetized with 100 mM levimasole (tetramisole hydrochloride, Sigma). GFP fluorescence in the pharyngeal or intestinal lumen was determined by visual inspection at 10X magnification on the Zeiss Imager M1 Axioscope. Fluorescent and Nomarski images were captured at 10X magnification using a Zeiss Axioimager A2 with an attached Zeiss AxioCam camera controlled by the software package Zeiss AxioVision.
The number of worms displaying bacterial fluorescence in the pharynx only, the gut only, or both the pharynx and gut were scored based on these images. These categories were chosen to assay the presence of the above-background fluorescence imparted by the bacteria carrying the GFP-expressing plasmid along the entire gastrointestinal tract; no distinction was made in the absolute levels of fluorescence in these categories. Representative mages were chosen to display the predominant category for each time point and diet. The results were pooled and subjected to Chi-squared analysis. The null hypothesis was ascertained as the values attained from OP50 fed animals.
Student’s T-tests were used to determine significance of single comparisons. One-way ANOVA analyses with Fisher's test were performed with StatView 5.0.1 software (SAS, CA) at a significance level of 0.05 for all multiple comparisons. Chi-square tests were utilized in Figure 7B and Additional file 4.
F.G. was supported by the Ruth L. Kirschstein National Service Award (GM007185), the NIH-NRSA Ruth L. Kirchstein Pre-doctoral Fellowship (F31GM082094-04), a Philip Whitcome Pre-doctoral Fellowship, and an UCLA Dissertation Year Fellowship Award. G.C.M. was supported by the Ford Foundation and a National Science Foundation Graduate Research Fellowship. This work was supported by UCLA School of Medicine (A.R.F.) and by the National Institutes of Health Grant AG19777 and National Science Foundation Grant 0919609 (C.F.C.). We thank Steve Clarke, Sean Curran, Lars Dreier, Nancy Freitag, David Gems, Shauna Hill, Theresa Nguyen, Alex van der Bliek and David Weinkove for helpful discussions, advice, and comments on the manuscript. We thank James Gober and Courtney White for help with bacterial microscopy and Mannon Guillermin, Michelle Castelleto, and Elissa Hallem for advice and help with GFP-worm microscopy. Some strains were provided by the CGC, which is funded by NIH Office of Research Infrastructure Programs (P40 OD010440).
- 1.Prakash S, Rodes L, Coussa-Charley M, Tomaro-Duchesneau C: Gut microbiota: next frontier in understanding human health and development of biotherapeutics. Biologics: targets & therapy. 2011, 5: 71-86.Google Scholar
- 7.Ottaviani E, Ventura N, Mandrioli M, Candela M, Franchini A, Franceschi C: Gut microbiota as a candidate for lifespan extension: an ecological/evolutionary perspective targeted on living organisms as metaorganisms. Biogerontology. 2011, 12 (6): 599-609. 10.1007/s10522-011-9352-5.PubMedCrossRefGoogle Scholar
- 9.Matsumoto M, Kurihara S, Kibe R, Ashida H, Benno Y: Longevity in mice is promoted by probiotic-induced suppression of colonic senescence dependent on upregulation of gut bacterial polyamine production. PLoS One. 2011, 6 (8): e23652-10.1371/journal.pone.0023652.PubMedPubMedCentralCrossRefGoogle Scholar
- 16.Ikeda T, Yasui C, Hoshino K, Arikawa K, Nishikawa Y: Influence of lactic acid bacteria on longevity of Caenorhabditis elegans and host defense against Salmonella enterica serovar enteritidis. Appl Environ Microbiol. 2007, 73 (20): 6404-6409. 10.1128/AEM.00704-07.PubMedPubMedCentralCrossRefGoogle Scholar
- 18.Saiki R, Lunceford AL, Bixler T, Dang P, Lee W, Furukawa S, Larsen PL, Clarke CF: Altered bacterial metabolism, not coenzyme Q content, is responsible for the lifespan extension in Caenorhabditis elegans fed an Escherichia coli diet lacking coenzyme Q. Aging Cell. 2008, 7 (3): 291-304. 10.1111/j.1474-9726.2008.00378.x.PubMedPubMedCentralCrossRefGoogle Scholar
- 20.Gomez F, Saiki R, Chin R, Srinivasan C, Clarke CF: Restoring de novo coenzyme Q biosynthesis in Caenorhabditis elegans coq-3 mutants yields profound rescue compared to exogenous coenzyme Q supplementation. Gene. 2012, 506: 106-116. 10.1016/j.gene.2012.06.023.PubMedPubMedCentralCrossRefGoogle Scholar
- 35.Kurz CL, Chauvet S, Andres E, Aurouze M, Vallet I, Michel GP, Uh M, Celli J, Filloux A, De Bentzmann S, et al: Virulence factors of the human opportunistic pathogen Serratia marcescens identified by in vivo screening. EMBO J. 2003, 22 (7): 1451-1460. 10.1093/emboj/cdg159.PubMedPubMedCentralCrossRefGoogle Scholar
- 42.Bekker M, Alexeeva S, Laan W, Sawers G, de Teixeira Mattos J, Hellingwerf K: The ArcBA two-component system of Escherichia coli is regulated by the redox state of both the ubiquinone and the menaquinone pool. J Bacteriol. 2010, 192 (3): 746-754. 10.1128/JB.01156-09.PubMedPubMedCentralCrossRefGoogle Scholar
- 47.Law RJ, Hamlin JN, Sivro A, McCorrister SJ, Cardama GA, Cardona ST: A functional phenylacetic acid catabolic pathway is required for full pathogenicity of Burkholderia cenocepacia in the Caenorhabditis elegans host model. J Bacteriol. 2008, 190 (21): 7209-7218. 10.1128/JB.00481-08.PubMedPubMedCentralCrossRefGoogle Scholar
- 51.Tsoi SC, Li SS: The nucleotide and deduced amino-acid sequences of a cDNA encoding lactate dehydrogenase from Caenorhabditis elegans: the evolutionary relationships of lactate dehydrogenases from mammals, birds, amphibian, fish, nematode, plants, bacteria, mycoplasma, and plasmodium. Biochem Biophys Res Commun. 1994, 205 (1): 558-564. 10.1006/bbrc.1994.2701.PubMedCrossRefGoogle Scholar
- 57.Hsu AY, Poon WW, Shepherd JA, Myles DC, Clarke CF: Complementation of coq3 mutant yeast by mitochondrial targeting of the Escherichia coli UbiG polypeptide: evidence that UbiG catalyzes both O-methylation steps in ubiquinone biosynthesis. Biochemistry. 1996, 35 (30): 9797-9806. 10.1021/bi9602932.PubMedCrossRefGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.