© 2004 by The Society for Integrative and Comparative Biology
Role of Nitric Oxide and Mitochondria in Control of Firefly Flash1
1 Department of Biology, University of Richmond, Richmond, Virginia 23173
2 Department of Biology, Tufts University, Medford, Massachusetts 02155
3 Department of Biology, Merrimack College, North Andover, Massachusetts 01845
 |
SYNOPSIS |
|---|
TOP
 SYNOPSIS INTRODUCTIONREFERENCES |
|---|
In light-producing cells (photocytes) of the firefly light organ, mitochondria are clustered in the cell periphery, positioned between the tracheolar air supply and the oxygen-requiring bioluminescent reactants which are sequestered in more centrally-localized peroxisomes. This relative positioning suggests that mitochondria could control oxygen availability for the light reaction. We hypothesized that active cellular respiration would make the interior regions of the photocytes relatively hypoxic, and that the "on" signal for production of bioluminescence might depend on inhibition of mitochondrial oxygen consumption, which would allow delivered oxygen to pass through the peripheral mitochondrial zone to reach peroxisomes deep in the cell interior. We published recently that exogenous NO induces bioluminescence in the intact firefly; that NO mediates octopamine-induced bioluminescence in the dissected lantern, and that nitric oxide synthase is abundant in cells of the tracheolar system of the light organ. Additional experiments showed that nitric oxide gas (NO) inhibits respiration in isolated lantern mitochondria. Inhibition is reversed by bright light, and this inhibition is relieved when the light is turned off. Altogether, the results support the idea that NO triggers light production by reversible inhibition of mitochondrial respiration in lantern cells, and probably in tracheolar cells as well. The data also suggest that the light of bioluminescence itself relieves NO inhibition thus contributing to rapid on/off switching. While other mechanisms may be in play, NO production that is directly related to neural input appears to have a key role in the oxygen gating that controls flash communication signals.
|
INTRODUCTION |
|---|
|
TOP
SYNOPSISINTRODUCTIONREFERENCES |
|---|
The beautiful phenomenon of firefly bioluminescence has no doubt been admired since antiquity. The flashes occur in precise and reproducible patterns and with spectra that are characteristic for different species. The obvious functional importance of flashing is mate location (Branham and Greenfield, 1996
), and recently the quality of individual flashes was shown to be a factor in sexual selection as well (Cratsley and Lewis, 2003). The biochemical basis of bioluminescence in fireflies involves the luciferase-catalyzed ATP-dependent activation of luciferin to a luciferyl-adenylate intermediate, which upon reaction with oxygen emits a photon (McElroy and DeLuca, 1978). While many behavioral and biochemical aspects of flashing are well understood, the on-off regulation of bioluminescence is still a fascinating puzzle into which a few pieces have finally begun to fit.
Discovering an on/off control mechanism for the bioluminescent reaction in the adult lantern has been particularly challenging, because controlling neurons do not synapse on lantern cells; instead they end on tracheolar cells that lie between the tracheolar airway and the photocytes. The lack of direct innervation of the light-producing cells suggested the existence of a chemical mediator that must diffuse rapidly and readily cross cell membranes in order to act on adjacent cells. We proposed nitric oxide (NO), a soluble gas, for this trans-cellular signaling role (Trimmer et al., 2001). Named "Molecule of the Year" in 1992 by Science (Culotta and Koshland, 1992), NO has a solid reputation as a "second messenger" capable of mediating neural and hormonal signals in a variety of tissues and systems throughout the animal kingdom, including insects (Trimmer et al., 2004).
In experiments with fireflies at several levels— whole animal, intact organ, cells, and organelles—we have accumulated data consistent with the idea that NO indeed is the mediator of neural signals that regulate bioluminescence in the adult. Most of the experimental results been published elsewhere (Trimmer et al., 2001; Aprille et al., 2002; Trimmer et al., 2004). The purpose of this work is to provide an integrated synopsis of the evidence to date that supports a role for NO in flash control. At the end of the paper, the relation of NO to other factors (such as gas vs. aqueous phase diffusion distances in the tracheoles) that may contribute to flash control is discussed briefly.
Evidence that NO is a signal mediator in flash control
Firstly, we observed that intact fireflies responded by flashing when NO was added as a gas (70 ppm) to a normoxic atmosphere (80% N2/20% O2) in a gas perfusion chamber (Trimmer et al., 2001). The result was most reproducible in headless fireflies possibly because neural inputs that regulate spiracle closing were eliminated, thus allowing the gas mixtures unrestricted access to the tracheolar system that supplies the lantern. Light production began when NO was introduced to the chamber, and stopped when NO was removed (Fig. 1). The addition of NO did not evoke light in an atmosphere of N2 only (no O2, not shown).
|
Secondly, lantern organs dissected with neural inputs removed but with the tracheal system intact produced light when the neurotransmitter octopamine was applied in the bathing medium. The bioluminescent response to octopamine was inhibited in the presence of the NO scavenger carboxy-PTIO (Fig. 2). No exogenous NO was used in these experiments, showing that NO was a naturally occurring mediator of the neural signal that normally produces flashing. We therefore concluded that NO synthase must be present in the lantern, and capable of being activated by an octopamine signal to generate NO.
|
In a third set of experiments, we showed in whole mounts of the intact adult lantern that NO synthase was present abundantly in the tracheolar end organ. Standard NADPH-diaphorase staining (Fig. 3) and immunocytochemistry (Fig. 4) were both employed to show that the enzyme was localized in cells of the tracheolar system that are interposed between neural inputs and the light-producing photocytes. In Figure 3, dark staining corresponds to NO synthase localization; some landmarks of lantern cellular anatomy are labeled as well to refresh the reader's appreciation of the elegant spatial arrangements that appear to be so important for flash control (discussed below). In Figure 4 the immunocytochemical localization of NO synthase is dramatically revealed by confocal microscopy. Figures 4A, B, C show representative 1 um planes as the focus was moved through the lantern in a dorsal to ventral direction. In addition to localization in cells of the tracheolar end organ, immunofluorescence can be seen along the apical region of photocytes, and along the radial arms of each rosette as well; the latter might represent NO synthase localization in the thin layer of tracheolar cell cytoplasm that surrounds the fine tracheoles projecting between the photocytes, or localization to the mitochondrial-rich periphery of the photocyte cytoplasm, or both. At higher resolution of NADPH diaphorase stained lanterns, NO synthase was also seen in the mitochondrial-rich apical regions of photocytes where they abut the center of the rosette (Trimmer et al., 2001
).
|
|
Evidence that the interaction of NO with lantern mitochondria is an oxygen-gating mechanism for flash control
The results so far showed that NO was most likely the trans-cellular mediator that connects neural signals impinging on the tracheolar cells to a light response in the photocytes. The next question was: How does NO do this? In other words, what is the biochemical mechanism of NO action that results in light production? The answer was suggested by the particular location of NO synthase in cells of the tracheolar end organ. The NO produced by these cells in response to neural signals accesses tracheolar cell mitochondria instantly, and needs only to diffuse over very short distances to reach mitochondria which are clustered peripherally in the photocytes (Trimmer et al., 2001
; Trimmer et al., 2004).
In many other systems, NO typically acts by activating guanylyl cyclase in a target cell; we detected no increase in cGMP in stimulated lanterns (Trimmer et al., 2001). NO is also known to have direct inhibitory effects on mitochondrial oxygen consumption (Fig. 5). In mammalian cells, NO at low concentrations reversibly inhibits mitochondrial respiration by interaction with cytochrome-c-oxidase; the enzyme inhibition was reported to be reversed by light in vitro although no physiological function for light-reversal was imagined (Brown, 1999). These precedent observations in mammalian mitochondria suggested the intriguing idea that in the firefly lantern, reversible NO inhibition of mitochondrial oxygen consumption could function as the on/off signal for light production, thus indirectly controlling oxygen access to pre-activated luciferin that is sequestered in the more centrally located peroxisomes. Such a mechanism is consistent with the now well-accepted view that oxygen gating is the key to flash control (Ghiradella, 1998; Timmins et al., 2001). There may be other oxygen gating mechanisms in play which control physical barriers for diffusional access of oxygen to the photocytes, perhaps at the level of spiracles or by regulation of fluid balance in the fine tracheoles (Timmins et al., 2001). But to the extent that any oxygen is delivered at all, however slowly, a mechanism is still needed to keep it away from bioluminescent reactants that are poised for oxidation in the peroxisomes. We envision mitochondrial oxygen consumption as the final control point for removal of any oxygen that actually reaches the photocyte when the airways are open. Oxygen gating to initiate the flash occurs when, upon neuronal signaling, NO is generated to inhibit oxygen extraction by mitochondria, allowing oxygen to pass through to oxidize the luciferin-adenylate intermediate. To test certain aspects of this idea, we performed experiments with mitochondria isolated from the firefly lantern in order to determine whether NO inhibits oxygen consumption and if it does, whether the inhibition is reversed by light.
|
We first had to develop a method for isolating lantern mitochondria by differential centrifugation (Aprille et al., 2002
). An electron micrograph of the resulting mitochondrial fraction of the cellular homogenate is shown in Figure 6; in these preparations, mitochondria were the dominant organelle and appeared to be intact. A standard polarographic assay was used to measure oxygen consumption by the isolated mitochondria (Modica-Napolitano et al., 1990). We determined that antimycin-a inhibited active respiration completely in the polarographic assays, providing assurance that the oxygen consumption observed was due to electron transport and not to some other oxygen-requiring reaction that might have been present in the crude preparation of isolated mitochondria. To introduce NO into the assay at the appropriate time, the NO donor drug NOC-7 was added to a final concentration of 4 uM. A Kodak 35 mM high-intensity slide projector (18.8 x 104 lux incident on surface of the water-jacketed assay chamber) was employed to test whether light could reverse NO inhibition of respiration.
|
Figure 7 shows a typical polarographic recording; time is on the vertical axis with the experiment starting at the bottom, and oxygen concentration in the assay chamber decreases from left to right on the horizontal axis. The rate of respiration is easily calculated from the recording as the change in oxygen concentration per minute. Respiration by lantern mitochondria in the presence of the respiratory substrate (5 mM each glutamate and malate) was very slow until ADP was added. Upon addition of ADP (1,000 nmoles, arrow a) maximal rates of coupled respiration ensued; the average rate was 118 nmoles O/min/mg protein (range 101–133, n = 4) for two separate preparations of mitochondria each tested 2–3 times. Note that ADP greatly stimulated respiration, indicating that oxidation was tightly coupled to the phosphorylation of ADP to ATP and confirming that the mitochondria were intact. Next, while oxygen consumption was still maximal, the NO donor NOC-7 was added (arrow b), resulting in complete inhibition of oxygen consumption. The inhibition was not instantaneous most likely because it takes a little time for the NOC-7 to generate a significant concentration of NO in the assay. Once inhibition was complete, bright light was shone on the assay chamber (arrow c); this caused oxygen consumption to resume indicating that NO inhibition was relieved. Repeated episodes of light on/light off (c–f in Fig. 7) confirmed that inhibition by NO was potent, yet easily reversed by bright light.
|
The effect of filtered light (Kodak #21 filter, cutoff <540 nm) was tested as well since the peak wavelength of firefly bioluminescence is in the 550–580 nm range (Seliger and McElroy, 1960
). Reversal of NO inhibition by light with the cutoff filter in place restored the oxygen consumption rate to 64% of initial control (data not shown), which was similar to that seen for white light (78% restoration in Fig. 7).
Summary of the interaction between NO and mitochondria in relation to flash control
Figure 8 summarizes the proposed role of NO in on/ off flash control. In the "Quiescent" or dark mode (upper half of Fig. 8), oxygen in the air delivered through the lantern tracheolar system dissolves in cellular fluids and is extracted by respiration in mitochondria (small ovals) clustered in the infolded regions of the tracheolar end cells and especially in the peripheral cytoplasm of photocytes. This renders the photocyte cytoplasm hypoxic, thus keeping oxygen away from the light-producing oxidation reaction sequestered in the centrally located peroxisomes. ATP produced by oxidative phosphorylation in the quiescent state is probably also necessary for formation of the activated luciferyl-adenylate intermediate (asterisk), which can accumulate in the absence of oxygen.
|
In the "Flash" mode (lower half of Fig. 8), the neurotransmitter octopamine activates lantern NO synthase (NOS). The NO inhibits mitochondrial oxygen consumption resulting in the pass-through of dissolved oxygen delivered by the tracheoles. The sudden increase in oxygen concentration in the photocyte interior triggers the production of light by oxidation of the accumulated luciferyl-adenylate intermediate in the peroxisomes.
Cessation of bioluminescence occurs when the neural signal ceases and NO production stops. There should be little latency, since NO is exceedingly labile and is rapidly degraded. Oxygen competition for NO bound to cytochrome c oxidase may also contribute to the rapid restoration of respiration (Brown, 1999). And finally, our evidence suggests that light from the flash itself may contribute to the off signal (dotted arrow in Fig. 8) by promoting degradation of the NO bound to the light-sensitive cytochrome (Brown, 1999), further hastening the restoration of oxygen consumption by mitochondria.
An integrated view of flash control
Most certainly there are other factors besides NO inhibition of respiration that contribute to regulation of firefly bioluminescence. The fine control needed for precise communication signals in the adult may be qualitatively different from the controls that account for transitions between a dormant lantern and one that is flashing. Table 1 attempts to order several possible levels of control. Air delivery via the tracheolar system can be controlled grossly by spiracle closing but usually only in response to threats or stress. The next possible control point might be the barrier imposed by oxygen diffusion through the fluid-filled fine tracheoles. In a recent report, Timmins et al. (2001) hypothesized as the result of an elegant analysis, that resorption of tracheolar fluids to increase the gas versus aqueous diffusion distance would result in faster oxygen delivery to the photocytes and could be a factor in flash control. In this model, a specific mechanism to connect neural signals to the regulation of tracheolar fluid remains to be discovered. We can suggest that NO, in addition to inhibiting mitochondrial oxygen consumption, might also regulate ion pumps in the tracheolar cell membranes through a local cGMP-dependent protein kinase. Regulation of ion pumps in this way has been described in mammalian kidney (Wang, 1997).
|
The model for oxygen gating by control of fluid in tracheoles depends mainly on the faster diffusion kinetics for oxygen in gas versus liquid (Timmins et al., 2001
). While a mechanism for alternate fluid resorption and filling in the tracheoles could certainly modulate oxygen delivery rates, it is not clear yet whether fluid regulation itself occurs during flashing, especially on a time scale that is consistent with repeated flashes that last only a second or so. In any case, gating of oxygen delivery by regulation of tracheolar fluid cannot account for selective distribution of oxygen to organelles within the photocyte. Some oxygen delivery to the photocytes must occur virtually all the time as long as spiracles are open, even when the animals are not flashing (during the day, for example). At the level of the photocytes it is logical that oxygen must be available for mitochondrial synthesis of ATP in quantities sufficient to activate luciferin, yet at the same time oxygen must be withheld from the oxidative bioluminescent reaction. Mitochondria provide a solution to this intracellular distribution problem—their peripheral location in the photocytes in areas adjacent to the tracheoles puts them in position to extract oxygen that would otherwise reach the peroxisomes, and they can automatically make ATP in the process.
Interestingly, there are mitochondrial-rich infolded membrane regions of the tracheolar end cells and tracheolar cells that surround tracheoles in the vicinity of their branch points (Ghiradella, 1998). By analogy to Mapighian tubule structure, these crypts have been invoked as a possible location of ion pumps that might osmotically regulate fluid level in the tracheoles (Timmins et al., 2001). We suggest that an alternative or at least additional function of these mitochondrial-rich crypts of the tracheolar end organ cells may be to serve as the first line of defense for extraction of oxygen from delivered air (see Fig. 8). Efficient extraction of oxygen by the crypt mitochondria would be enhanced by the large surface area for oxygen diffusion created by the convoluted membrane foldings. Because crypt mitochondria are in the same cells as are the sites of NO synthase, they would be subject to respiratory inhibition by NO even more immediately than photocyte mitochondria, thus participating along with photocyte mitochondria in NO-mediated oxygen gating within the lantern. The proposed control mechanisms are consistent with the fact that artificial hyperoxia induces bioluminescence (Buck, Case, and Hanson, 1963; Timmins et al., 2001) presumably because all the oxygen gating mechanisms discussed so far (mitochondrial respiration, tracheolar fluid barriers) can be swamped by mass action.
It is also interesting to think about the possible significance of the role of NO in explaining the different bioluminescent response of the larval and adult lantern organs to neural stimulation. The peripheral localization of mitochondria in adult photocytes appears to be the key for triggering bioluminescence in many peroxisomes at virtually the same time resulting in an intense flash. It remains to be seen whether NO is an operative component in control of larval lantern bioluminescence. If it is, the slower development of a glow in larval lanterns rather than a flash is consistent with the fact that peroxisomes and mitochondria are evenly intermixed throughout the cytoplasm (Ghiradella, 1998). These will be interesting ideas to follow up on, although the experiments are sure to be challenging.
We have presented a strong case for mitochondria as the final "gatekeepers" for oxygen access to the bioluminescent reaction in the adult firefly lantern. The reader is referred to Greenfield (2001) for an interesting critical analysis of our conclusions, and a discussion of some evolutionary implications. While many features of flash control remain to be explored, we do seem closer to answering the old question of how firefly communication is so precisely regulated.
|
ACKNOWLEDGMENTS |
|---|
We are grateful to David M. Dudsinski, Sara M. Lewis, Thomas Michel, Sanjev Qazi, and Ricardo M. Zayas for the collaboration that started this line of research. Thanks also to Deanna Ward and Roni Kingsley for assistance with electron microscopy in Figure 6; and to Robert Willson for confocal microscopy in Figure 4.
|
FOOTNOTES |
|---|
1 From the Symposium Flash Communication: Fireflies at Fifty presented at the Annual Meeting of the Society for Integrative and Comparative Biology, 4–8 January 2003, at Toronto, Canada.
2 E-mail: japrille{at}richmond.edu
|
REFERENCES |
|---|
|
TOP
SYNOPSISINTRODUCTIONREFERENCES |
|---|
Aprille, J. R., C. J. Lagace, S. M. Lewis, T. Michel, J. S. Modica-Napolitano, B. A. Trimmer, and R. M. Zayas. 2002. Mechanism of firefly flash control: Nitric oxide inhibition of oxygen consumption in lantern mitochondria is reversed by light. In P. E. Stanley and L. J. Kricka (eds.), Bioluminescence & chemiluminescence: Progress & current applications, pp. 25–28. World Scientific Publishing, Singapore.
Branham, M. A., and M. D. Greenfield. 1996. Flashing males win mate success. Nature, 381:745-746.[Medline]
Brown, G. C. 1999. Nitric oxide and mitochondrial respiration. Biochim. Biophys. Acta, 1411:351-369.[Medline]
Buck, J. B., J. F. Case, and F. E. Hanson. 1963. Control of flashing in fireflies. III. Peripheral excitation. Biol. Bull, 125:251-269.
Cratsley, C. K., and S. M. Lewis. 2003. Female preference for male courtship flashes in Photinus ignits fireflies. Behav. Ecol, 14:135-140.
Culotta, E., and D. E. Koshland Jr. 1992. NO news is good news. Science, 258:1862-1865.
Ghiradella, H. 1998. The anatomy of light production: The fine structure of the firefly lantern. In F. W. Harrison and M. Locke (eds.), Microscopic anatomy of invertebrates, Vol 11A, Insecta, pp. 363–381. Wiley-Liss, New York.
Greenfield, M. D. 2001. Missing link in firefly bioluminescence revealed: NO regulation of photocyte respiration. BioEssays, 23:992-995.[CrossRef][Web of Science][Medline]
McElroy, W. D., and M. DeLuca. 1978. Chemistry of firefly luminescence. In P. J. Herring (ed.), Bioluminescence in action, pp. 109–127. Academic Press, London.
Modica-Napolitano, J. S., J. L. Joyal, G. Ara, A. R. Oseroff, and J. R. Aprille. 1990. Mitochondrial toxicity of cationic photosensitizers for photochemotherapy. Cancer Research, 50:7876-7881.
Seliger, H. H., and W. D. McElroy. 1960. Spectral emission and quantum yield of firefly bioluminescence. Arch. Biochem. Biophys, 88:136-41.[CrossRef][Web of Science][Medline]
Timmins, G. S., F. J. Fracer, C. M. Wilmot, S. K. Jackson, and H. M. Swartz. 2001. Firefly flashing is controlled by gating oxygen to light emitting cells. J. Exp. Biol, 204:2795-2801.
Trimmer, B. A., J. R. Aprille, D. M. Dudzinski, C. J. Lagace, S. M. Lewis, T. Michel, S. Qazi, and R. M. Zayas. 2001. Nitric oxide and the control of firefly flashing. Science, 292:2486-2488.
Trimmer, B. A., J. R. Aprille, and J. S. Modica Napolitano. 2004. Nitric oxide signalling: Insect brains and photocytes. In C. Cooper, M. Wilson, and V. Darley-Usmar (eds.),. Biochemical Society Annual Symposium, Vol. 71, Free radicals: Enzymology, signalling and disease, pp. 65–83. Portland Press, London.
Wang, T. 1997. Nitric oxide regulates HCO3– and Na+ transport by a cGMP-mediated mechanism in the kidney proximal tubule. Am. J. Physiol. Renal Physiol, 272:242-248.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||