Microbiome studies have identified correlations between bacteria and host aging (Kundu et al., 2017; O'Toole and Jeffery, 2015). For example, 16S rRNA and metagenomic DNA sequencing are used to associate the presence or abundance of specific bacteria to human centenarians (Biagi et al., 2016; Claesson et al., 2012; Claesson et al., 2011). However, given the complexity and heterogeneity of the human gut environment, these approaches are unable to elucidate how a specific microbial species contributes to longevity.

The nematode Caenorhabditis elegans has a short and easily-measured lifespan, features that have revolutionized our understanding of the molecular genetics of aging and longevity (Kenyon, 2010). Studies using C. elegans also provide mechanistic insight into the association between bacterial species and host longevity (Gusarov et al., 2013; Kim, 2013). Importantly, recent studies have revealed that bacterial metabolism can produce specific products to directly influence the aging process in the host C. elegans or modulate the effects of environmental cues on C. elegans lifespan (Cabreiro et al., 2013; Pryor et al., 2019; Virk et al., 2016). These findings highlight the significance of bacterial metabolism in regulating host physiology during the aging process and have inspired interest in directly manipulating bacterial metabolism in situ in the host gastrointestinal (GI) tract.

In several recent studies, researchers have administered antibiotic- or carbohydrate-based small molecule inducers to modulate gene expression from gut bacteria (Kotula et al., 2014; Lim et al., 2017; Mimee et al., 2015). While this approach can be used to perform in situ analyses of gut bacterial-host interactions (Lim et al., 2017), chemical effectors may have unwanted side-effects on host or microbial physiology and are subject to slow and poorly-controlled transport and degradation processes that ultimately limit their precision.

Optogenetics combines light and genetically-engineered photoreceptors to enable unrivaled control of biological processes (Olson and Tabor, 2014). Previously, we and others have engineered photoreceptors that regulate bacterial gene expression in response to specific wavelengths of light (Levskaya et al., 2005; Li et al., 2020; Ohlendorf et al., 2012; Ong and Tabor, 2018; Ong et al., 2018; Ramakrishnan and Tabor, 2016; Ryu and Gomelsky, 2014; Schmidl et al., 2014). These photoreceptors have been used to achieve precise quantitative (Olson et al., 2017; Olson et al., 2014), temporal (Chait et al., 2017; Milias-Argeitis et al., 2016; Olson et al., 2017; Olson et al., 2014), and spatial (Chait et al., 2017; Levskaya et al., 2005; Ohlendorf et al., 2012; Tabor et al., 2011) control of bacterial gene expression in in vitro culture conditions. They have also been used to characterize and control transcriptional regulatory circuits (Chait et al., 2017; Olson et al., 2014; Tabor et al., 2009) and bacterial metabolic pathways (Fernandez-Rodriguez et al., 2017; Tandar et al., 2019) in vitro. Here, we hypothesized that optogenetic control of bacterial gene expression might provide a new way to fine-tune bacterial metabolism in the host GI tract with high temporal precision and no unwanted side-effects.

We address this possibility using the E. coli and C. elegans interaction model, an optically transparent system with known mechanistic links between bacterial metabolism and host longevity. In particular, CA is an exopolysaccharide synthesized and secreted by E. coli that can extend the lifespan of the host C. elegans by modulating mitochondrial dynamics (Han et al., 2017). First, we engineer E. coli strains that express fluorescent reporter proteins under control of our previously engineered green light-activated, red light de-activated two-component system CcaSR (Schmidl et al., 2014). We then expose worms carrying these engineered E. coli in their GI tracts to time-varying green and red light signals to demonstrate optical activation and de-activation of bacterial gene expression in worm gut. Next, we genetically engineer an E. coli strain that biosynthesizes CA under control of CcaSR, and utilize green light to induce CA production from this strain in the gut of the host C. elegans. Our experiments reveal that green light-induced CA from bacteria residing in the host gut is sufficient to modulate mitochondrial dynamics and lifespan. We use this approach to investigate the local effect of CA production on intestinal cells in a time-controlled manner and its systemic effect on organisms in a tunable way, neither of which is possible using the previous methods of administering purified CA to worms or feeding worms E. coli mutants that constitutively overproduce CA. This work paves the way for future studies of bacterial-host interactions with high temporal, spatial, and quantitative control without the shortcomings of chemical inducers.

To demonstrate optogenetic control over gut bacterial gene expression, we first engineered E. coli strain LH01, wherein CcaSR controls expression of superfolder green fluorescent protein (sfgfp), and mcherry is expressed constitutively to facilitate identification of the bacteria (Figure 1a, Figure 1—figure supplement 1, Supplementary file 1a, 1b). In LH01, green light-exposure switches CcaS to an active state in the presence of chromophore phycocyanobilin (PCB), wherein it phosphorylates the response regulator CcaR. Phosphorylated CcaR then activates transcription of sfgfp from the P_cpcG2-172 output promoter. Red light reverts active CcaS to the inactive form, de-activating sfgfp expression (Figure 1a).

Figure 1 with 4 supplements see all

Download asset Open asset

We then reared two groups of C. elegans from the larval to the adult stage on plates of LH01 under red or green light, respectively (Figure 1b). Next, we washed away external bacteria, applied the paralyzing agent levamisole to prevent expulsion of gut contents, and transferred the worms to agar pads. Finally, we switched the applied light color from red to green (step ON), or green to red (step OFF), and used epi-fluorescence microscopy to image the resulting changes in sfGFP and mCherry fluorescence in the gut lumen over time (Figure 1c). In the step ON experiment, sfGFP fluorescence in the worm gut lumen starts low, begins to increase within 2 hr, and reaches a saturated high level at 6 hr (Figure 1d). On the other hand, in the step OFF experiment, sfGFP fluorescence begins high, and decreases exponentially between hours 1–7 (Figure 1d). This light response is abolished when the PCB biosynthetic operon is removed (ΔPCB) (Figure 1d), demonstrating that changes in sfGFP fluorescence are due to CcaSR photoswitching in the worm gut.

Next, we used flow cytometry to analyze these bacterial light responses with single-cell resolution. Specifically, we reared worms on light sensing bacteria in red and green light as before, but then washed, paralyzed, and placed them into microtubes prior to light switching (Figure 1b). At several time points over the course of 8 hr, we homogenized the animals, harvested the gut contents, sorted bacterial cells, and measured fluorescence via flow cytometry. We observed that our engineered bacteria remain intact in the host gut (Figure 1—figure supplement 2) and respond to light in a unimodal fashion (Figure 1e–g). The temporal dynamics and PCB dependence of the gene expression responses recapitulate the microscopy results (Figure 1—figure supplement 3). We also confirmed that residual bacteria on the exterior of worms do not contribute to the flow cytometry measurements (Figure 1—figure supplement 4). Together, these microscopy and flow cytometry experiments demonstrate that we can use optogenetics to rapidly and reversibly induce gene expression of E. coli residing in the C. elegans gut.

Next, we sought to utilize our optogenetic method to modulate the production of specific metabolites in bacteria residing in the gut of live hosts. Bacterial genes involved in the same metabolic process are often clustered into operons and co-regulated at the transcriptional level. We took advantage of this coordinated mode of regulation and chose the cps operon and its transcription activator RcsA for testing optogenetic control of bacterial metabolism. The cps operon in E. coli consists of 19 genes that encode enzymes required for the biosynthesis and secretion of CA (Torres-Cabassa and Gottesman, 1987), and CA-overproducing bacterial mutants Δlon and Δhns promote longevity in the host C. elegans (Han et al., 2017). To place CA biosynthesis under optogenetic control, we engineered E. coli strain MVK29 which lacks genomic rcsA, and expresses a heterologous copy of rcsA under the control of CcaSR (Figure 2a, Supplementary file 1a, 1b). We first examined whether MVK29 could respond to green light and induce CA production and secretion. To this end, we grew the strain in batch culture under red or green light and quantified supernatant CA levels. In red light, MVK29 secretes CA to concentrations below the limit of detection of the assay, similar to the ΔrcsA mutant (Figure 2b). Green light, on the other hand, induces MVK29 to secrete high levels of CA, and removal of the PCB biosynthetic operon abolishes this response (Figure 2b). Moreover, mutation of the CcaS catalytic histidine to a non-functional alanine (H534A), or the CcaR phosphorylation site from an aspartic acid to a non-functional asparagine (D51N) abolishes detectable CA production (Figure 2b). Importantly, the level of secreted CA increases sigmoidally with the increasing intensity of green light (Figure 2c), which is consistent with the response of CcaSR (Schmidl et al., 2014). We conclude that we have placed CA production under the control of CcaSR, and that we can use light to tune the biosynthesis and secretion of bacterial metabolites.

Figure 2

Download asset Open asset

We then used this approach to study a gut bacterial metabolite-host interaction pathway in vivo. We first focused on cellular phenotypes in the worm gut that directly interact with bacteria and examined mitochondrial morphology that is known to be affected by CA (Han et al., 2017). Lon is an ATP-dependent protease that degrades RcsA, and its deletion mutant Δlon shows increased CA production (Han et al., 2017). When grown on Δlon mutant bacteria, worms exhibit elevated mitochondrial fragmentation in intestinal cells, similar to worms with CA supplementation (Han et al., 2017). To determine whether light-induced CA overproduction in MVK29 could affect the host as effectively as the CA overproduction caused by the Δlon mutant, we reared two different transgenic worm strains, expressing either mitochondrially-localized GFP (mito-GFP) or mitochondrially-localized RFP (mito-RFP) in the intestine (Supplementary file 1b), on the plate with MVK29 and exposed the plate to red light. For each worm strain, we continued to expose one group to red light but switched a second to green for an additional 6 hr (unparalyzed light-exposure, Figure 3a). After paralyzing the worms with levamisole, we immediately imaged intestinal cell mitochondrial morphology using confocal microscopy and categorized the morphology into the tubular, intermediate, or fragmented (Figure 3b). We found that green light increases mitochondrial fragmentation in MVK29-fed worms, but not in worms fed a MVK29 ΔPCB control strain that does not respond to light (Figure 3c, Figure 3—figure supplement 1). Thus, our light-inducible CA biosynthetic strain recapitulates the effect of the Δlon mutant or purified CA on the host.

Figure 3 with 1 supplement see all

Download asset Open asset

We next set out to induce CA production directly from bacteria residing within the host gut and examine its effect on intestinal mitochondria. To this end, we removed worms from plates, washed away external bacteria, paralyzed them using levamisole to stop bacteria being expelled from the gut, exposed them to green or red light for 6 hr, and then imaged intestinal mitochondrial morphology using confocal microscopy (paralyzed light-exposure, Figure 3d). In these conditions, CA production is inactivated on the plate and worms are only exposed to CA produced from bacteria residing with the host gut. Unexpectedly, we found that 6 hr of levamisole treatment does not kill worms but leads to mitochondrial hyper-fragmentation in the intestine (Figure 3c,e), possibly due to the inhibitory effect of levamisole on mitochondrial NADH-oxidizing enzymes (Köhler and Bachmann, 1978) and its associated mitochondrial stress. This hyper-fragmentation phenotype was not observed in worms treated with levamisole less than 1 hr. This stress-induced fragmentation resembles mitochondrial changes related to aging and age-related neurodegenerative diseases (Cho et al., 2009; Exner et al., 2007; Sebastián et al., 2017). Interestingly, we found that green light exposure suppresses stress-induced hyperfragmentation in paralyzed worms bearing MVK29 in the gut (Figure 3e, Figure 3—figure supplement 1). Importantly, we found no such effects in worms bearing the ΔPCB, CcaS(H534A), CcaR(D51N) or ΔrcsA mutant strains (Figure 3e, Figure 3—figure supplement 1), suggesting that this protective effect is a result of light-induced CA overproduction in the gut. As a control using transgenic worms expressing mito-GFP in the muscle, we also examined muscular mitochondrial morphology under paralyzed light-exposure conditions. We found that 6 hr levamisole treatment induces mitochondrial hyper-fragmentation in the muscle as it does in the intestine, but that light-induced CA production from the gut bacteria does not protect muscular mitochondria against hyper-fragmentation (Figure 3f, Figure 3—figure supplement 1). These results not only show that optogenetics can be utilized to induce CA secretion from gut-borne E. coli in vivo, but also reveal a local protective effect of CA on intestinal cells.

Finally, we examined how the extent of light-induced CA production relates to host lifespan extension. We first confirmed that green-light-exposed MVK29 can induce CA overproduction to a level comparable to the Δlon mutant (Figure 4a, Figure 4—figure supplement 1), and neither the Δlon nor the ΔrcsA mutant responds to light (Figure 4a). Next, beginning at the day-1 adult stage, we exposed worms bearing MVK29 to red, or low and high green light intensities resulting in intermediate or saturating levels of CA secretion in vitro (Figure 2c, Figure 4—figure supplement 1), and measured their lifespans. We found that green light exposure extends worm lifespan compared to red light exposure (Figure 4b). Furthermore, the extent of lifespan extension at high green light intensity is 1.7-fold higher than that at low intensity (Figure 4b, Supplementary file 1c), suggesting a dose-dependent effect. In parallel, we measured the lifespan of worms bearing the CA-overproducing Δlon mutant and showed that the lifespan extension caused by Δlon is independent of light exposure (Figure 4c). The extent of lifespan extension caused by Δlon is comparable to that caused by MVK29 (Figure 4b,c, Supplementary file 1c), which is consistent with the result that Δlon and green-light-exposed MVK29 induce CA overproduction to a similar level (Figure 4a). In addition, the lifespan of worms bearing the ΔrcsA mutant is also light-independent and is comparable to that of MVK29 worms under red light (Figure 4d, Supplementary file 1c). These results suggest that optogenetic control is sufficient to induce bacterial production of pro-longevity compounds and improve host health. Unlike administration of a bacterial mutant, optogenetic control of bacterial metabolism can modulate a host-level phenotype in a dose-dependent manner.

Figure 4 with 1 supplement see all

Download asset Open asset

Our method has broad applications for studying microbe-host interactions in situ. For example, we have identified about two dozen additional E. coli genes that are unrelated to CA biosynthesis and that enhance worm longevity when knocked out (Han et al., 2017), though the mechanisms by which they act remain largely unclear. By using light to induce their expression in the gut and measuring acute host responses such as changes in mitochondrial dynamics, the role of these genes in gut microbe-host interactions could be further explored. In another example, the quorum-sensing peptide CSF and nitric oxide, both of which are produced by Bacillus subtilis during biofilm formation, have been found to extend worm lifespan through downregulation of the insulin-like signaling pathway (Donato et al., 2017). We have recently ported CcaSR into B. subtilis and demonstrated that it enables rapid and precise control of gene expression dynamics (Castillo-Hair et al., 2019). The method we report here should enable optogenetic control of gene expression and metabolite production in B. subtilis and in situ studies of how this important Gram-positive model bacterium impacts longevity as well.

Multiple photoreceptors could also be combined to study more complex microbe-host interaction pathways. Specifically, we and others have co-expressed CcaSR with independently-controllable blue/dark and red/far-red reversible light sensors to achieve simultaneous and independent control of the expression of up to three genes in the same bacterial cell (Fernandez-Rodriguez et al., 2017; Olson et al., 2017; Tabor et al., 2011). Such optogenetic multiplexing could be performed in situ and used to study potential synergistic, antagonistic, or other higher-order effects of multiple bacterial genes or pathways. A large number of eukaryotic photoreceptors have also been developed, enabling optical control of many cell- and neurobiological processes (Deisseroth, 2015; Gautier et al., 2014; Goglia and Toettcher, 2019; Leopold et al., 2018). Bacterial and eukaryotic photoreceptors could be combined to enable simultaneous optical manipulation of bacterial and host pathways to interrogate whether or how they interact. Optogenetics could also be used to manipulate bacterial and/or host pathways at specific locations within the gut to examine location- or tissue-dependent phenomena.

Finally, optogenetic approaches could be extended to other bacteria or hosts. Although E. coli is an important model of Gram-negative bacteria, it is not a natural symbiont of C. elegans. Because E. coli does not stably colonize in the C. elegans gut, we have to paralyze worms to facilitate longer-term experiments. It would be interesting future work to port CcaSR or other bacterial photoreceptors into native C. elegans symbionts (Zhang et al., 2017) or pathogens (Couillault and Ewbank, 2002) and apply them for studying the physiological and pathological effects of bacterial symbionts in vivo in non-paralyzing hosts. Moreover, it is likely that light can also be used to control gut bacterial gene expression in other model hosts such as flies, zebrafish, or mammals. Red-shifted wavelengths and corresponding optogenetic tools (Ong et al., 2018; Ryu and Gomelsky, 2014) may prove superior for less optically transparent or larger animals. Overall, by enabling precision control of bacterial gene expression and metabolism in situ, we believe that optogenetics will greatly improve our understanding of a wide range of microbe-host interactions. In particular, through aiding the digestion of ingested food, intestinal bacteria alter their gene expression and metabolism and can consequently generate specific metabolites to influence other microbes and the host in close proximity. Understanding dynamic diet-microbe-host interplays demands a quantitative method to manipulate bacterial gene expression in a temporally controlled manner. For example, upon the supplementation of a specific food source, turning on or off a specific gene in bacteria and in turn examining molecular changes in the host would help reveal primary nodes in the complex diet-microbe-host network. Given their quantitative nature, rapid and reversible switching ability, and suitability for in vivo application, optogenetic approaches will provide a new toolbox for this area of research to decipher how mechanistically microbial genetic factors participate in those diet-microbe-host interplays and contribute to host health and diseases.

All data generated or analysed during this study are included in the manuscript and supporting files.

1. 1. Baba T
   2. Ara T
   3. Hasegawa M
   4. Takai Y
   5. Okumura Y
   6. Baba M
   7. Datsenko KA
   8. Tomita M
   9. Wanner BL
   10. Mori H

   (2006) Construction of Escherichia coli K-12 in-frame, single-gene knockout mutants: the keio collection

   Molecular Systems Biology 2:2006.0008.

   https://doi.org/10.1038/msb4100050

   • PubMed
   • Google Scholar
2. 1. Biagi E
   2. Franceschi C
   3. Rampelli S
   4. Severgnini M
   5. Ostan R
   6. Turroni S
   7. Consolandi C
   8. Quercia S
   9. Scurti M
   10. Monti D
   11. Capri M
   12. Brigidi P
   13. Candela M

   (2016) Gut Microbiota and extreme longevity

   Current Biology : CB 26:1480–1485.

   https://doi.org/10.1016/j.cub.2016.04.016

   • PubMed
   • Google Scholar
3. 1. Cabreiro F
   2. Au C
   3. Leung KY
   4. Vergara-Irigaray N
   5. Cochemé HM
   6. Noori T
   7. Weinkove D
   8. Schuster E
   9. Greene ND
   10. Gems D

   (2013) Metformin retards aging in C. elegans by altering microbial folate and methionine metabolism

   Cell 153:228–239.

   https://doi.org/10.1016/j.cell.2013.02.035

   • PubMed
   • Google Scholar
4. 1. Castillo-Hair SM
   2. Sexton JT
   3. Landry BP
   4. Olson EJ
   5. Igoshin OA
   6. Tabor JJ

   (2016) FlowCal: a User-Friendly, open source software tool for automatically converting flow cytometry data from arbitrary to calibrated units

   ACS Synthetic Biology 5:774–780.

   https://doi.org/10.1021/acssynbio.5b00284

   • PubMed
   • Google Scholar
5. 1. Castillo-Hair SM
   2. Baerman EA
   3. Fujita M
   4. Igoshin OA
   5. Tabor JJ

   (2019) Optogenetic control of Bacillus subtilis gene expression

   Nature Communications 10:3099.

   https://doi.org/10.1038/s41467-019-10906-6

   • PubMed
   • Google Scholar
6. 1. Chait R
   2. Ruess J
   3. Bergmiller T
   4. Tkačik G
   5. Guet CC

   (2017) Shaping bacterial population behavior through computer-interfaced control of individual cells

   Nature Communications 8:1535.

   https://doi.org/10.1038/s41467-017-01683-1

   • PubMed
   • Google Scholar
7. 1. Chen YJ
   2. Liu P
   3. Nielsen AA
   4. Brophy JA
   5. Clancy K
   6. Peterson T
   7. Voigt CA

   (2013) Characterization of 582 natural and synthetic terminators and quantification of their design constraints

   Nature Methods 10:659–664.

   https://doi.org/10.1038/nmeth.2515

   • PubMed
   • Google Scholar
8. 1. Cho DH
   2. Nakamura T
   3. Fang J
   4. Cieplak P
   5. Godzik A
   6. Gu Z
   7. Lipton SA

   (2009) S-nitrosylation of Drp1 mediates beta-amyloid-related mitochondrial fission and neuronal injury

   Science 324:102–105.

   https://doi.org/10.1126/science.1171091

   • PubMed
   • Google Scholar
9. 1. Claesson MJ
   2. Cusack S
   3. O'Sullivan O
   4. Greene-Diniz R
   5. de Weerd H
   6. Flannery E
   7. Marchesi JR
   8. Falush D
   9. Dinan T
   10. Fitzgerald G
   11. Stanton C
   12. van Sinderen D
   13. O'Connor M
   14. Harnedy N
   15. O'Connor K
   16. Henry C
   17. O'Mahony D
   18. Fitzgerald AP
   19. Shanahan F
   20. Twomey C
   21. Hill C
   22. Ross RP
   23. O'Toole PW

   (2011) Composition, variability, and temporal stability of the intestinal Microbiota of the elderly

   PNAS 108:4586–4591.

   https://doi.org/10.1073/pnas.1000097107

   • PubMed
   • Google Scholar
10. 1. Claesson MJ
    2. Jeffery IB
    3. Conde S
    4. Power SE
    5. O'Connor EM
    6. Cusack S
    7. Harris HM
    8. Coakley M
    9. Lakshminarayanan B
    10. O'Sullivan O
    11. Fitzgerald GF
    12. Deane J
    13. O'Connor M
    14. Harnedy N
    15. O'Connor K
    16. O'Mahony D
    17. van Sinderen D
    18. Wallace M
    19. Brennan L
    20. Stanton C
    21. Marchesi JR
    22. Fitzgerald AP
    23. Shanahan F
    24. Hill C
    25. Ross RP
    26. O'Toole PW

    (2012) Gut Microbiota composition correlates with diet and health in the elderly

    Nature 488:178–184.

    https://doi.org/10.1038/nature11319

    • PubMed
    • Google Scholar
11. 1. Couillault C
    2. Ewbank JJ

    (2002) Diverse Bacteria are pathogens of Caenorhabditis elegans

    Infection and Immunity 70:4705–4707.

    https://doi.org/10.1128/iai.70.8.4705-4707.2002

    • PubMed
    • Google Scholar
12. 1. Deisseroth K

    (2015) Optogenetics: 10 years of microbial opsins in neuroscience

    Nature Neuroscience 18:1213–1225.

    https://doi.org/10.1038/nn.4091

    • PubMed
    • Google Scholar
13. 1. Dische Z

    (1947)

    A new specific color reaction of hexuronic acids

    J Biological Chem 167:189–198.

    • Google Scholar
14. 1. Dische Z
    2. Shettles LB

    (1948)

    A specific color reaction of methylpentoses and a spectrophotometric micromethod for their determination

    J Biological Chem 175:595–603.

    • Google Scholar
15. 1. Donato V
    2. Ayala FR
    3. Cogliati S
    4. Bauman C
    5. Costa JG
    6. Leñini C
    7. Grau R

    (2017) Bacillus subtilis biofilm extends Caenorhabditis elegans longevity through downregulation of the insulin-like signalling pathway

    Nature Communications 8:14332.

    https://doi.org/10.1038/ncomms14332

    • PubMed
    • Google Scholar
16. 1. Engler C
    2. Gruetzner R
    3. Kandzia R
    4. Marillonnet S

    (2009) Golden gate shuffling: a one-pot DNA shuffling method based on type IIs restriction enzymes

    PLOS ONE 4:e5553.

    https://doi.org/10.1371/journal.pone.0005553

    • PubMed
    • Google Scholar
17. 1. Exner N
    2. Treske B
    3. Paquet D
    4. Holmström K
    5. Schiesling C
    6. Gispert S
    7. Carballo-Carbajal I
    8. Berg D
    9. Hoepken HH
    10. Gasser T
    11. Krüger R
    12. Winklhofer KF
    13. Vogel F
    14. Reichert AS
    15. Auburger G
    16. Kahle PJ
    17. Schmid B
    18. Haass C

    (2007) Loss-of-function of human PINK1 results in mitochondrial pathology and can be rescued by parkin

    Journal of Neuroscience 27:12413–12418.

    https://doi.org/10.1523/JNEUROSCI.0719-07.2007

    • PubMed
    • Google Scholar
18. 1. Fernandez-Rodriguez J
    2. Moser F
    3. Song M
    4. Voigt CA

    (2017) Engineering RGB color vision into Escherichia coli

    Nature Chemical Biology 13:706–708.

    https://doi.org/10.1038/nchembio.2390

    • PubMed
    • Google Scholar
19. 1. Gautier A
    2. Gauron C
    3. Volovitch M
    4. Bensimon D
    5. Jullien L
    6. Vriz S

    (2014) How to control proteins with light in living systems

    Nature Chemical Biology 10:533–541.

    https://doi.org/10.1038/nchembio.1534

    • PubMed
    • Google Scholar
20. 1. Gerhardt KP
    2. Olson EJ
    3. Castillo-Hair SM
    4. Hartsough LA
    5. Landry BP
    6. Ekness F
    7. Yokoo R
    8. Gomez EJ
    9. Ramakrishnan P
    10. Suh J
    11. Savage DF
    12. Tabor JJ

    (2016) An open-hardware platform for optogenetics and photobiology

    Scientific Reports 6:35363.

    https://doi.org/10.1038/srep35363

    • PubMed
    • Google Scholar
21. 1. Goglia AG
    2. Toettcher JE

    (2019) A bright future: optogenetics to dissect the spatiotemporal control of cell behavior

    Current Opinion in Chemical Biology 48:106–113.

    https://doi.org/10.1016/j.cbpa.2018.11.010

    • PubMed
    • Google Scholar
22. 1. Gusarov I
    2. Gautier L
    3. Smolentseva O
    4. Shamovsky I
    5. Eremina S
    6. Mironov A
    7. Nudler E

    (2013) Bacterial nitric oxide extends the lifespan of C. elegans

    Cell 152:818–830.

    https://doi.org/10.1016/j.cell.2012.12.043

    • PubMed
    • Google Scholar
23. 1. Han B
    2. Sivaramakrishnan P
    3. Lin CJ
    4. Neve IAA
    5. He J
    6. Tay LWR
    7. Sowa JN
    8. Sizovs A
    9. Du G
    10. Wang J
    11. Herman C
    12. Wang MC

    (2017) Microbial genetic composition tunes host longevity

    Cell 169:1249–1262.

    https://doi.org/10.1016/j.cell.2017.05.036

    • PubMed
    • Google Scholar
24. 1. Hermann GJ
    2. Schroeder LK
    3. Hieb CA
    4. Kershner AM
    5. Rabbitts BM
    6. Fonarev P
    7. Grant BD
    8. Priess JR

    (2005) Genetic analysis of lysosomal trafficking in Caenorhabditis elegans

    Molecular Biology of the Cell 16:.

    https://doi.org/10.1091/mbc.e05-01-0060

    • PubMed
    • Google Scholar
25. 1. Kenyon CJ

    (2010) The genetics of ageing

    Nature 464:504–512.

    https://doi.org/10.1038/nature08980

    • PubMed
    • Google Scholar
26. 1. Kim DH

    (2013) Bacteria and the aging and longevity of Caenorhabditis elegans

    Annual Review of Genetics 47:233–246.

    https://doi.org/10.1146/annurev-genet-111212-133352

    • PubMed
    • Google Scholar
27. 1. Köhler P
    2. Bachmann R

    (1978)

    The effects of the antiparasitic drugs levamisole, thiabendazole, praziquantel, and chloroquine on mitochondrial electron transport in muscle tissue from Ascaris suum

    Molecular Pharmacology 14:155–163.

    • PubMed
    • Google Scholar
28. 1. Kotula JW
    2. Kerns SJ
    3. Shaket LA
    4. Siraj L
    5. Collins JJ
    6. Way JC
    7. Silver PA

    (2014) Programmable Bacteria detect and record an environmental signal in the mammalian gut

    PNAS 111:4838–4843.

    https://doi.org/10.1073/pnas.1321321111

    • PubMed
    • Google Scholar
29. 1. Kundu P
    2. Blacher E
    3. Elinav E
    4. Pettersson S

    (2017) Our gut microbiome: the evolving inner self

    Cell 171:1481–1493.

    https://doi.org/10.1016/j.cell.2017.11.024

    • PubMed
    • Google Scholar
30. 1. Leopold AV
    2. Chernov KG
    3. Verkhusha VV

    (2018) Optogenetically controlled protein kinases for regulation of cellular signaling

    Chemical Society Reviews 47:2454–2484.

    https://doi.org/10.1039/C7CS00404D

    • PubMed
    • Google Scholar
31. 1. Levskaya A
    2. Chevalier AA
    3. Tabor JJ
    4. Simpson ZB
    5. Lavery LA
    6. Levy M
    7. Davidson EA
    8. Scouras A
    9. Ellington AD
    10. Marcotte EM
    11. Voigt CA

    (2005) Engineering Escherichia coli to see light

    Nature 438:441–442.

    https://doi.org/10.1038/nature04405

    • Google Scholar
32. 1. Li X
    2. Zhang C
    3. Xu X
    4. Miao J
    5. Yao J
    6. Liu R
    7. Zhao Y
    8. Chen X
    9. Yang Y

    (2020) A single-component light sensor system allows highly tunable and direct activation of gene expression in bacterial cells

    Nucleic Acids Research 48:e33.

    https://doi.org/10.1093/nar/gkaa044

    • PubMed
    • Google Scholar
33. 1. Lim B
    2. Zimmermann M
    3. Barry NA
    4. Goodman AL

    (2017) Engineered regulatory systems modulate gene expression of human commensals in the gut

    Cell 169:547–558.

    https://doi.org/10.1016/j.cell.2017.03.045

    • PubMed
    • Google Scholar
34. 1. Milias-Argeitis A
    2. Rullan M
    3. Aoki SK
    4. Buchmann P
    5. Khammash M

    (2016) Automated optogenetic feedback control for precise and robust regulation of gene expression and cell growth

    Nature Communications 7:12546.

    https://doi.org/10.1038/ncomms12546

    • PubMed
    • Google Scholar
35. 1. Mimee M
    2. Tucker AC
    3. Voigt CA
    4. Lu TK

    (2015) Programming a human commensal bacterium, Bacteroides thetaiotaomicron, to Sense and Respond to Stimuli in the Murine Gut Microbiota

    Cell Systems 1:62–71.

    https://doi.org/10.1016/j.cels.2015.06.001

    • PubMed
    • Google Scholar
36. 1. O'Toole PW
    2. Jeffery IB

    (2015) Gut microbiota and aging

    Science 350:1214–1215.

    https://doi.org/10.1126/science.aac8469

    • PubMed
    • Google Scholar
37. 1. Ohlendorf R
    2. Vidavski RR
    3. Eldar A
    4. Moffat K
    5. Möglich A

    (2012) From dusk till dawn: one-plasmid systems for light-regulated gene expression

    Journal of Molecular Biology 416:534–542.

    https://doi.org/10.1016/j.jmb.2012.01.001

    • PubMed
    • Google Scholar
38. 1. Olson EJ
    2. Hartsough LA
    3. Landry BP
    4. Shroff R
    5. Tabor JJ

    (2014) Characterizing bacterial gene circuit dynamics with optically programmed gene expression signals

    Nature Methods 11:449–455.

    https://doi.org/10.1038/nmeth.2884

    • PubMed
    • Google Scholar
39. 1. Olson EJ
    2. Tzouanas CN
    3. Tabor JJ

    (2017) A photoconversion model for full spectral programming and multiplexing of optogenetic systems

    Molecular Systems Biology 13:926.

    https://doi.org/10.15252/msb.20167456

    • PubMed
    • Google Scholar
40. 1. Olson EJ
    2. Tabor JJ

    (2014) Optogenetic characterization methods overcome key challenges in synthetic and systems biology

    Nature Chemical Biology 10:502–511.

    https://doi.org/10.1038/nchembio.1559

    • Google Scholar
41. 1. Ong NT
    2. Olson EJ
    3. Tabor JJ

    (2018) Engineering an E. coli Near-Infrared light sensor

    ACS Synthetic Biology 7:240–248.

    https://doi.org/10.1021/acssynbio.7b00289

    • PubMed
    • Google Scholar
42. 1. Ong NT
    2. Tabor JJ

    (2018) A miniaturized Escherichia coli Green Light Sensor with High Dynamic Range

    Chembiochem : A European Journal of Chemical Biology 19:1255–1258.

    https://doi.org/10.1002/cbic.201800007

    • PubMed
    • Google Scholar
43. 1. Portal-Celhay C
    2. Bradley ER
    3. Blaser MJ

    (2012) Control of intestinal bacterial proliferation in regulation of lifespan in Caenorhabditis elegans

    BMC Microbiology 12:49.

    https://doi.org/10.1186/1471-2180-12-49

    • PubMed
    • Google Scholar
44. 1. Pryor R
    2. Norvaisas P
    3. Marinos G
    4. Best L
    5. Thingholm LB
    6. Quintaneiro LM
    7. De Haes W
    8. Esser D
    9. Waschina S
    10. Lujan C
    11. Smith RL
    12. Scott TA
    13. Martinez-Martinez D
    14. Woodward O
    15. Bryson K
    16. Laudes M
    17. Lieb W
    18. Houtkooper RH
    19. Franke A
    20. Temmerman L
    21. Bjedov I
    22. Cochemé HM
    23. Kaleta C
    24. Cabreiro F

    (2019) Host-Microbe-Drug-Nutrient screen identifies bacterial effectors of metformin therapy

    Cell 178:1299–1312.

    https://doi.org/10.1016/j.cell.2019.08.003

    • PubMed
    • Google Scholar
45. 1. Ramakrishnan P
    2. Tabor JJ

    (2016) Repurposing Synechocystis PCC6803 UirS-UirR as a UV-Violet/Green photoreversible transcriptional regulatory tool in E. coli

    ACS Synthetic Biology 5:733–740.

    https://doi.org/10.1021/acssynbio.6b00068

    • PubMed
    • Google Scholar
46. 1. Ryu MH
    2. Gomelsky M

    (2014) Near-infrared light responsive synthetic c-di-GMP module for optogenetic applications

    ACS Synthetic Biology 3:802–810.

    https://doi.org/10.1021/sb400182x

    • PubMed
    • Google Scholar
47. 1. Schmidl SR
    2. Sheth RU
    3. Wu A
    4. Tabor JJ

    (2014) Refactoring and optimization of light-switchable Escherichia coli two-component systems

    ACS Synthetic Biology 3:820–831.

    https://doi.org/10.1021/sb500273n

    • PubMed
    • Google Scholar
48. 1. Sebastián D
    2. Palacín M
    3. Zorzano A

    (2017) Mitochondrial dynamics: coupling mitochondrial fitness with healthy aging

    Trends in Molecular Medicine 23:201–215.

    https://doi.org/10.1016/j.molmed.2017.01.003

    • PubMed
    • Google Scholar
49. 1. Tabor JJ
    2. Salis HM
    3. Simpson ZB
    4. Chevalier AA
    5. Levskaya A
    6. Marcotte EM
    7. Voigt CA
    8. Ellington AD

    (2009) A synthetic genetic edge detection program

    Cell 137:1272–1281.

    https://doi.org/10.1016/j.cell.2009.04.048

    • PubMed
    • Google Scholar
50. 1. Tabor JJ
    2. Levskaya A
    3. Voigt CA

    (2011) Multichromatic control of gene expression in Escherichia coli

    Journal of Molecular Biology 405:315–324.

    https://doi.org/10.1016/j.jmb.2010.10.038

    • PubMed
    • Google Scholar
51. 1. Tandar ST
    2. Senoo S
    3. Toya Y
    4. Shimizu H

    (2019) Optogenetic switch for controlling the central metabolic flux of Escherichia coli

    Metabolic Engineering 55:68–75.

    https://doi.org/10.1016/j.ymben.2019.06.002

    • PubMed
    • Google Scholar
52. 1. Torres-Cabassa AS
    2. Gottesman S

    (1987) Capsule synthesis in Escherichia coli K-12 is regulated by proteolysis

    Journal of Bacteriology 169:981–989.

    https://doi.org/10.1128/JB.169.3.981-989.1987

    • PubMed
    • Google Scholar
53. 1. Virk B
    2. Jia J
    3. Maynard CA
    4. Raimundo A
    5. Lefebvre J
    6. Richards SA
    7. Chetina N
    8. Liang Y
    9. Helliwell N
    10. Cipinska M
    11. Weinkove D

    (2016) Folate acts in E. coli to Accelerate C. elegans Aging Independently of Bacterial Biosynthesis

    Cell Reports 14:1611–1620.

    https://doi.org/10.1016/j.celrep.2016.01.051

    • PubMed
    • Google Scholar
54. 1. Wang MC
    2. Oakley HD
    3. Carr CE
    4. Sowa JN
    5. Ruvkun G

    (2014) Gene pathways that delay Caenorhabditis elegans reproductive senescence

    PLOS Genetics 10:e1004752.

    https://doi.org/10.1371/journal.pgen.1004752

    • PubMed
    • Google Scholar
55. 1. Zhang F
    2. Berg M
    3. Dierking K
    4. Félix MA
    5. Shapira M
    6. Samuel BS
    7. Schulenburg H

    (2017) Caenorhabditis elegans as a Model for Microbiome Research

    Frontiers in Microbiology 8:485.

    https://doi.org/10.3389/fmicb.2017.00485

    • PubMed
    • Google Scholar

Acceptance summary:

The reviewers and editors feel that the approach and techniques presented are exciting, with important future applications – especially given progress on extension the approach to e.g. B. subtilis. You convincingly demonstrate the utility and value of your novel optogenetic approach for investigating bacterial metabolism.

Decision letter after peer review:

Thank you for submitting your article "Optogenetic control of gut bacterial metabolism to promote longevity" for consideration by eLife. Your article has been reviewed by three peer reviewers, one of whom is a member of our Board of Reviewing Editors, and the evaluation has been overseen by Jessica Tyler as the Senior Editor. The following individual involved in review of your submission has agreed to reveal their identity: William Mair (Reviewer #2).

The reviewers have discussed the reviews with one another and the Reviewing Editor has drafted this decision to help you prepare a revised submission. We all feel that you when you resubmit the revised paper, it should be as a "Tools and Resources" paper rather than an article, given the methodological nature of your study.

We would like to draw your attention to changes in our revision policy that we have made in response to COVID-19 (https://elifesciences.org/articles/57162). Specifically, when editors judge that a submitted work as a whole belongs in eLife but that some conclusions require a modest amount of additional new data, as they do with your paper, we are asking that the manuscript be revised to either limit claims to those supported by data in hand, or to explicitly state that the relevant conclusions require additional supporting data.

Our expectation is that the authors will eventually carry out the additional experiments and report on how they affect the relevant conclusions either in a preprint on bioRxiv or medRxiv, or if appropriate, as a Research Advance in eLife, either of which would be linked to the original paper.

Summary:

Hartsough et al. present a method to optogenetically control the production of colanic acid (CA) from a genetically engineered E. coli strain. Using this E. coli strain as a feed for C. elegans, the authors show CA production directly in the animal gut and its local effects in altering mitochondrial dynamics in the gut. The authors corroborate findings from a previous publication from Han et al., 2017, showing effects on C. elegans lifespan and intestinal mitochondrial network.

The concept of modifying host biology via optogenetic control of gut flora is novel and would be fitting for the scope of eLife. However, the reviewers had some concerns regarding some of the claims regarding biological insights as presented. Given the current COVID-19 situation, we feel that asking for additional experiments and novel biological findings would be beyond the scope of a revision, but the reviewers suggested that additional data or edits may help clarify how this work is distinct from Han et al., 2017, why an optogenetic approach is better than the standard genetic approaches and what the strength and weaknesses of this approach are.

Essential revisions:

1) One of the key claims made by the authors is that the optogenetic approach allows for more fine scale control of metabolite production and further that they "investigate how the lifespan extending effect of CA relates to CA levels". However, aside from one ex vivo assay (Results, Figure 2B/C), the amount of CA being secreted (or taken up by the host) is not reported. The authors claim, but don't actually show a relationship of lifespan with CA level, just green-light intensity. The relative CA production levels of the engineered strains under green light vs. the over-producing mutants should ideally be quantified and both compared to realistic exposure levels in vivo of conventionally grown C. elegans. If this cannot be done without additional experiments, the authors should be careful to temper their claims here.

2) A related concern is how do the authors know that the effects are not mediated by CA produced and secreted on plates. They attempt to answer this question using levamisole paralyzed worms – but the mitochondrial morphology of paralyzed worms seems very different (hyper fragmented, Figure 3D) form non-paralyzed – as the authors themselves concede. I agree that there appears to be some rescue effect in MVK29 exposed to green light – but it would be preferable to show benefits of in situ expression without the overlaying levamisole detriment.

3) The mitochondrial fragmentation results are obviously motivated by the findings in Han et al., 2017, but are not well introduced/explained. In the fifth paragraph of the Results it is stated that exposure to light-induced CA secretion increases fragmentation. But in the next paragraph that light-induced CA secretion decreases mitochondrial fragmentation caused by levamisole treatment. The interpretation of these conflicting results-and their relevance to other systems lacking stress induced mitochondrial fragmentation-is unclear and should be more clearly discussed.

4) In addition to local effects in the gut, were the authors able to see effects in distal tissues? An experiment investigating effects of light-induced CA production in the gut, affecting body wall muscle mitochondrial network similar to Han et al., 2017, could strengthen the current manuscript and show expected systemic effects of CA and not just local.

5) This work relies on a non-standard gut microbe for C. elegans and requires that the animals be paralyzed to maintain colonization through measurement. This is addressed only in passing in the Discussion, but altogether still gives pause. Yes, E. coli is "this important Gram-positive model bacterium," but it isn't in fact a very important bacterium to this host. Limitations and open questions that should be made explicitly are: Whether host uptake of metabolites produced by a symbiont would differ from that of a food source, whether/how this approach could be extended to bacteria other than E. coli and how such interactions might play out in a community context and whether / how optogenetics could be used to study this question.

6) As a standard C. elegans feeding strain, OP50 is naturally lon deficient. Can the authors comment on the levels of CA production from their light-induced methods as compared to the regular OP50? If no such data is available, this should still be discussed.

7) The authors make a claim regarding the relative magnitude of lifespan effects of purified CA vs. CA produced using their optogenetic approach. However, it appears that claim is based on comparison of lifespan effects with previously published results. Such comparisons are risky because differences in lab, feeding conditions, and strain. It would also have been preferable to have a CA "fed" control and it would further have been interesting to carry out a study of optogenetic induction in the presence of exogenous CA. In the absence of a CA fed control, the authors should be careful with their claims regarding effect size.

8) There were some concerns over the exclusive use of GFP with a mito targeting sequence as the reporter because this only labels mitochondria if they are functionally importing the GFP. In addition there seems to be a lot of noise from intestinal gut granules and aggregates in Figure 3B. These data could be improved if an additional label of mitochondria was used – either a membrane marker or dye, in order to ensure the changes seen are reflecting changes to mitochondrial network morphology and not GFP aggregation, or non-mitochondrial localized GFP signal. The authors should be careful to make these limitations clear.

9) Lifespan of control and treated animals in general appear shorter than would be expected. E.g. in Figure 4, the lifespan of worms fed the E. coli background strain JW1935-1 is considerably short. Can the authors comment on any previous studies with C. elegans lifespan on this strain and if this is expected? Are there other optogenetic background strains available?

Essential revisions:

1) One of the key claims made by the authors is that the optogenetic approach allows for more fine scale control of metabolite production and further that they "investigate how the lifespan extending effect of CA relates to CA levels". However, aside from one ex vivo assay (Results, Figure 2B/C), the amount of CA being secreted (or taken up by the host) is not reported. The authors claim, but don't actually show a relationship of lifespan with CA level, just green-light intensity. The relative CA production levels of the engineered strains under green light vs. the over-producing mutants should ideally be quantified and both compared to realistic exposure levels in vivo of conventionally grown C. elegans. If this cannot be done without additional experiments, the authors should be careful to temper their claims here.

We appreciate the reviewer’s comment on the comparison of CA levels among different conditions. Technically, it is currently impossible to measure CA levels in bacteria grow on solid agar plates or in the worm gut, or in the host cells. Thus, our conclusion is based on the ex vivo assay using liquid culture and the positive relationship between green-light intensity and the amount of CA being secreted. Based on the reviewer’s suggestion, we have revised the text to explicitly state this relationship. In addition, we have included new results presenting a direct comparison of CA levels between the Δlon mutant and the light inducible strain (new Figure 4A).

2) A related concern is how do the authors know that the effects are not mediated by CA produced and secreted on plates. They attempt to answer this question using levamisole paralyzed worms – but the mitochondrial morphology of paralyzed worms seems very different (hyper fragmented, Figure 3D) form non-paralyzed – as the authors themselves concede. I agree that there appears to be some rescue effect in MVK29 exposed to green light – but it would be preferable to show benefits of in situ expression without the overlaying levamisole detriment.

We understand the reviewer’s concern on the possible confounding effect of levamisole. We agree that levamisole is an imperfect means by which to paralyze the worms, resulting in undesired changes to host mitochondrial morphology. Unfortunately, host paralysis and cessation of host peristaltic activity is required to measure the effect of CA production in vivo; otherwise, bacteria would be expelled before sufficient amount of CA accumulated to affect mitochondrial morphology in the host. We have tried several alternative means of immobilizing the host for this experiment, and only levamisole is sufficient to stop bacteria expulsion from the gut and keeps worms alive after 6 hours of the treatment.

Regarding the concern that CA may be produced on the plate instead of in the host, in the context of Figure 3 (new Figure 3E and F), note that neither bacteria nor worms on the plate are exposed to green light (which induces CA production). Worms are only exposed to green light after they have been removed from the plate and paralyzed. Thus, we conclude that the effect on mitochondrial morphology is mediated by CA produced from bacteria inside the host gut. In the context of Figure 4, we expect that CA produced from bacteria on the plate and in the host gut both contribute to the lifespan-extending effect. Currently, we cannot experimentally separate these two processes. In the revised manuscript, we have included new diagram figure panels to illustrate the experimental design (new Figure 3A and D) and improved the language to introduce the experimental design more clearly and increase the textual clarity of the related claims.

3) The mitochondrial fragmentation results are obviously motivated by the findings in Han et al., 2017, but are not well introduced/explained. In the fifth paragraph of the Results it is stated that exposure to light-induced CA secretion increases fragmentation. But in the next paragraph that light-induced CA secretion decreases mitochondrial fragmentation caused by levamisole treatment. The interpretation of these conflicting results-and their relevance to other systems lacking stress induced mitochondrial fragmentation-is unclear and should be more clearly discussed.

Thank the reviewer for pointing out this concern. We have revised the manuscript to (1) better introduce the previously published findings, (2) explain the new results more clearly, and (3) discuss the potential difference between these two conditions.

4) In addition to local effects in the gut, were the authors able to see effects in distal tissues? An experiment investigating effects of light-induced CA production in the gut, affecting body wall muscle mitochondrial network similar to Han et al., 2017, could strengthen the current manuscript and show expected systemic effects of CA and not just local.

As the reviewer suggested, we have examined the effect of CA on mitochondrial morphology in the muscle using day-2-old young adults. We found that light-induced CA production from the gut bacteria does not affect muscular mitochondria. We conclude that (1) CA acts locally in intestinal cells, and (2) the protective effect of CA against age-associated fragmentation in muscular mitochondria that we observed previously (Han et al., 2017) is likely associated with the lifespan extending effect of CA, which may be secondary to the change in the intestine. We thank the reviewer for his/her suggestion, which help us distinguish the local effect of CA in the intestine and its systemic effect on other tissues.

5) This work relies on a non-standard gut microbe for C. elegans and requires that the animals be paralyzed to maintain colonization through measurement. This is addressed only in passing in the Discussion, but altogether still gives pause. Yes, E. coli is "this important Gram-positive model bacterium," but it isn't in fact a very important bacterium to this host. Limitations and open questions that should be made explicitly are: Whether host uptake of metabolites produced by a symbiont would differ from that of a food source, whether/how this approach could be extended to bacteria other than E. coli and how such interactions might play out in a community context and whether / how optogenetics could be used to study this question.

We appreciate the insightful suggestions from the reviewer to discuss the application of this new method in natural endosymbiont bacteria beyond E. coli. In the

Discussion, we note that we have recently demonstrated optogenetic control of gene expression in Bacillus subtilis, and suggest its possible application in studying Gram-positive bacterial-host interactions and longevity. Based on the reviewer’s suggestion, we have included additional discussions on 1) the difference in metabolite uptake from a symbiont and a food source, 2) the application of the E. coli optogenetic approach in host organisms where E. coli is a natural symbiont, 3) the development of new optogenetic approaches in different bacterial systems.

6) As a standard C. elegans feeding strain, OP50 is naturally lon deficient. Can the authors comment on the levels of CA production from their light-induced methods as compared to the regular OP50? If no such data is available, this should still be discussed.

Thank the reviewer for the suggestion. We haven’t directly compared to the level of CA from OP50 and the light-inducible bacterial strain. In the Han et al., 2017 paper, we reported that OP50 shows a 50% higher CA level than BW25113 and Δlon shows a 4-fold higher CA level than BW25113. In the current work, we found that the CA levels from Δlon and the light-inducible strain are similar (new Figure 4A). Based on this comparison, we conclude that the level of CA production from OP50 is lower than that from the light-inducible strain. We have included these information (subsection “Optogenetic control of CA production”).

7) The authors make a claim regarding the relative magnitude of lifespan effects of purified CA vs. CA produced using their optogenetic approach. However, it appears that claim is based on comparison of lifespan effects with previously published results. Such comparisons are risky because differences in lab, feeding conditions, and strain. It would also have been preferable to have a CA "fed" control and it would further have been interesting to carry out a study of optogenetic induction in the presence of exogenous CA. In the absence of a CA fed control, the authors should be careful with their claims regarding effect size.

We appreciate the reviewer’s comment. In our experiments (Figure 4), we have chosen the Δlon mutant as a positive control and a comparative reference for the effect of CA produced using the optogenetic approach, since the lifespan-extending effect of Δlon is stronger than CA supplementation and CA supplementation does not enhance the lifespan extension by the loss of lon (as shown in the Han et al., 2017 paper). In Figure 4 and Supplementary file 1C, all the lifespan assays were conducted in parallel for direct comparison. We found that the lifespan extension caused by the light-inducible strain is comparable to that caused by Δlon. Based on light-inducible = Δlon (in this work) and Δlon > CA (Han et al., 2017), we concluded that light-inducible > CA. We agree with the reviewer that we should be more careful with our claims, since we did not directly include CA supplementation as a control. We have revised text in the Introduction and Results to state this claim more precisely.

8) There were some concerns over the exclusive use of GFP with a mito targeting sequence as the reporter because this only labels mitochondria if they are functionally importing the GFP. In addition there seems to be a lot of noise from intestinal gut granules and aggregates in Figure 3B. These data could be improved if an additional label of mitochondria was used – either a membrane marker or dye, in order to ensure the changes seen are reflecting changes to mitochondrial network morphology and not GFP aggregation, or non-mitochondrial localized GFP signal. The authors should be careful to make these limitations clear.

We appreciate the reviewer’s suggestion, and we have conducted this experiment using another RFP reporter that localized to the mitochondrial membrane, in which the Nterminal 55 amino acids sequence of C. elegans tomm-20 is fused with RFP. Using this new reporter, we have validated the results using mito-GFP (new Figure 3C, E and Figure 3—figure supplement 1).

9) Lifespan of control and treated animals in general appear shorter than would be expected. E.g. in figure 4, the lifespan of worms fed the E. coli background strain JW1935-1 is considerably short. Can the authors comment on any previous studies with C. elegans lifespan on this strain and if this is expected? Are there other optogenetic background strains available?

Thank the reviewer for pointing this out. We are sorry that we did not explain this experimental design more clearly. The lifespan assay was conducted at 26 degrees using the sqt-3(e2117) temperature sensitive mutant that reproduces normal but is embryonically lethal at 26 degrees. In this way, we can perform the experiment at the restrictive temperature to avoid animal transfer that causes white light exposure, while not using floxuridine/FUDR that interrupts normal reproduction and alters bacterial metabolism. We have used this method previously in the Han et al., 2017 paper, and the mean lifespan is 8.7 ± 0.1 days. Thus, the lifespan, ~6.5 days, shown in Figure 4 is slightly shorter. We think this is not due to the E. coli strain background given that the lifespan of Δlon is also shorter, ~ 9 days (this work) vs. ~11 days (in Han et al., 2017). The possible explanation is that light exposure may shorten the lifespan of C. elegans (PMID: 29500338). In the revised manuscript, we have explained the experimental design more clearly (subsection “Lifespan experiments”) and included the temperature information in figure panels (new Figure 4B-D).

Author details

1. Lucas A Hartsough

   Department of Bioengineering, Houston, United States

   Contribution

   Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Visualization, Methodology, Writing - original draft, Writing - review and editing

   Competing interests

   No competing interests declared

2. Mooncheol Park

   Huffington Center on Aging, Houston, United States

   Contribution

   Data curation, Formal analysis, Validation, Investigation, Visualization, Methodology, Writing - review and editing

   Competing interests

   No competing interests declared

3. Matthew V Kotlajich

   Department of Bioengineering, Houston, United States

   Contribution

   Conceptualization, Data curation, Methodology

   Competing interests

   No competing interests declared

4. John Tyler Lazar

   Department of Chemical and Biomolecular Engineering, Houston, United States

   Contribution

   Data curation, Formal analysis, Validation, Investigation, Methodology

   Competing interests

   No competing interests declared

5. Bing Han

   Huffington Center on Aging, Houston, United States

   Present address

   Children's Hospital & Institutes of Biomedical Sciences, Fudan University, Shanghai, China

   Contribution

   Conceptualization, Validation, Investigation, Visualization, Methodology

   Competing interests

   No competing interests declared

6. Chih-Chun J Lin

   1. Huffington Center on Aging, Houston, United States
   2. Department of Molecular & Human Genetics, Baylor College of Medicine, Houston, United States

   Present address

   Department of Molecular Biology and Genetics, Cornell University, Ithaca, United States

   Contribution

   Data curation, Formal analysis, Investigation, Visualization, Methodology

   Competing interests

   No competing interests declared

7. Elena Musteata

   Systems, Synthetic, and Physical Biology Program, Rice University, Houston, United States

   Contribution

   Validation

   Competing interests

   No competing interests declared

8. Lauren Gambill

   Systems, Synthetic, and Physical Biology Program, Rice University, Houston, United States

   Contribution

   Investigation

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-1490-2449

9. Meng C Wang

   1. Huffington Center on Aging, Houston, United States
   2. Department of Molecular & Human Genetics, Baylor College of Medicine, Houston, United States
   3. Howard Hughes Medical Institute, Houston, United States

   Contribution

   Conceptualization, Data curation, Formal analysis, Supervision, Funding acquisition, Writing - original draft, Writing - review and editing

   For correspondence

   wmeng@bcm.edu

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-5898-6007

10. Jeffrey J Tabor

    1. Department of Bioengineering, Houston, United States
    2. Systems, Synthetic, and Physical Biology Program, Rice University, Houston, United States
    3. Department of Biosciences, Houston, United States

    Contribution

    Conceptualization, Supervision, Funding acquisition, Investigation, Writing - original draft, Writing - review and editing

    For correspondence

    jeff.tabor@gmail.com

    Competing interests

    No competing interests declared

• 6,238

  Page views

• 875

  Downloads

• 38

  Citations

Article citation count generated by polling the highest count across the following sources: Crossref, Scopus, PubMed Central.