BIOLOGICAL OVERVIEW

Nitric oxide (NO) is a diffusible, free-radical gas that serves as a multifunctional messenger affecting many diverse aspects of mammalian physiology, such as regulation of vascular tone, macrophage-mediated cytotoxicity, and cell-cell interactions in the nervous system. For example, during the development of the rat central nervous system and the process of neuronal cell differentiation, NO carries messages crucial to synaptogenesis and apoptosis. Nitric oxide synthase (NOS), the enzyme that produces NO, is expressed transiently in the developing rat brain, as one might expect, given its role in neural development (Bredt, 1994). NOS has been found to trigger the switch to the state of growth arrest that occurs during the differentiation of neuronal cells (Peunova, 1995). NO also appears to be involved in long-term potentiation in hippocampal tissue, according to an experimental model that posits neural plasticity in the learning process (Arancio, 1996).

NOS catalyzes the formation of nitric oxide from the terminal guanidino nitrogen of arginine, with the consequent production of citrulline. There are several homologous but separate NOS genes in mammals. Endothelial NOS and neuronal NOS enzymes are produced constitutively and are Calmodulin dependent, with activity regulated by calcium. The calcium dependence of nNOS suggests that nNOS is activated by neural activity. Macrophage NOS is expressed at low levels under basal conditions, is induced by gamma-interferon and lipopolysaccharide, and does not depend on calcium for activity. Inducible NOS is an important component of a host's resistence to infection, especially to parasites such as malaria.

What is the target of NO in signal transduction? NO directly activates the soluble form of guanylate cyclase, the heme protein that catalyzes the formation of cyclic GMP. In addition, NO binds with high affinity to heme moieties in other proteins. NO also affects enzymes that contain iron-sulfur groups, thus affecting iron metabolism. The immediate target of cyclic GMP is PKG (cGMP-dependent protein kinase), known to be the mediator of smooth muscle relaxation and other affector functions of NO signaling (Lincoln, 1995).

What role is played in normal development by developmentally regulated expression of NOS? Drosophila NOS is expressed at high levels in developing imaginal discs. Injection of specific NOS inhibitors into developing larvae at the end of the third instar (several hours before metamorphosis), results in an enlargement of appendages and other structures of the fly body. The changes include (1) hypertrophy of the femur, tibia, and segments of the tarsus; (2) overgrowth of the tissues originating from the genital disc; (3) an increase in the overall surface of the wings; (4) overgrowth of cells of tergites and sternites; (5) hypertrophy of the humerous; (6) occasional duplications of some areas of the eye; (7) occasional malformation of genital structures, legs, and eyes; and (8) occasional ectopic fomation of misplaced body structures. The changes most often affect and are most profound in the legs of adults, where the diameter of certain segments increases 3-4 times (Kuzin, 1996).

The opposite effects occur with the ectopic expression of a mouse NOS transgene. Flies were transformed with a NOS transgene in which NOS was placed under control of a heat-shock promoter. Transgenic larvae were heat-shocked within 1 hour after pupariation to induce ectopic expression of NOS before the final cell divisions took place. This treatment results in a reduction in limb size; distal segments in legs are affected most frequently and to the greatest degree. The segments of the adult leg most often affected by the overexpression of NOS are those that are not affected by the NOS inhibitors and whose precursors exhibit particularly low levels of enzyme staining in the early prepupal stages. It is concluded that ectopic expression of NOS at the late stages of larval development results in a decrease in cell proliferation and a reduction in the size of the adult fly's structures (Kuzin, 1996).

After inhibition of NOS in the eye disc, there is a consistent increase in the number of cells in S phase, but the resultant adult eye usually appears normal. What is the basis for this paradoxical finding? It could be that the apparently normal eye phenotype occurs as a result of programmed cell death, which counteracts excessive cell proliferation induced by NOS inhibition and restores the normal number of cells in the eye during morphogenesis.

To test this hypothesis, p35, a baculovirus inhibitor of apoptosis (programmed cell death) was expressed under the control of a multimerized Glass-binding site, obtained from the Drosophila Rhodopsin1 promoter. This promoter directs expression of the transgene in all cells, both in and posterior to the morphogenetic furrow in the eye imaginal disc. When NOS is inhibited in transgenic larvae, the eyes show numerous changes, reflecting the excessive proliferation of various cell types in the developing eye. After NOS inhibition in transgenic flies, the number of ommatidia increases from a nearly invariant complement of 750 (in wild-type flies and untreated transgenic flies) to nearly 820. This, together with an elevated number of cells per ommatidium, causes an increase in the overall size of the eye. In transgenic flies, the number of secondary and tertiary pigment cells is increased from 12 to 25 cells per sample area as a result of suppressed programmed cell death. Furthermore, the number of bristles in these NOS inhibited transgenic flies is increased in some areas of the eye to five or six in many ommatidia, where normally one would expect only four (Kuzin, 1996).

How general is the phenomenon of NO-mediated growth arrest in organism development? A strong elevation of NOS activity in the developing cerebral cortical plate and hippocampus of prenatal rats at days 15-19 correlates with the time course of cessation of precursor cell proliferation, tight growth arrest, and cell differentiation; notably, NOS activity goes down after proliferation of committed neuronal precursors is completed (Bredt, 1994 and Blottner, 1995). NOS levels are also transiently increased in developing lungs, bones, blood vessels and nervous system (Blottner, 1995, and Wetts, 1995). Elsewhere, NOS activity is greatly elevated in regenerating tissues when cessation of cell division is crucial for prevention of unregulated cell growth (Roskams, 1994, Blottner, 1995 and Hortelano, 1995). In all these cases, a transient elevation of NOS activity might trigger a switch from proliferation to growth arrest and differentiation, thus contributing to the proper morphogenesis of tissues and organs (Kuzin, 1996).

A nitric oxide/cyclic GMP (cGMP) signaling pathway is thought to play an important role in mammalian vasodilation during hypoxia. Drosophila utilizes components of this pathway to respond to hypoxia. Hypoxic exposure rapidly induces exploratory behavior in larvae and arrests the cell cycle. Exposure of larvae to 1% oxygen (about 5% of normal atmospheric levels) produces a striking behavioral response. Typically, larvae feed on yeast paste with only their posterior ends protruding from the food. This leaves the two posterior spiracles, the openings of the larval tracheal system, exposed to the outside atmosphere. Within seconds of oxygen deprivation, larvae stop feeding, back up slightly, and then exit the yeast. Subsequently, over about a 2 min period, larval motility increases but is largely confined to the surface of the yeast. In a third phase of the response, the larvae leave the yeast and wandered onto a clean agar surface, occasionally escaping the petri dish. After 15 min of 1% oxygen, nearly 75% of the larvae have left the yeast. Prolonged exposure to 1% oxygen (greater than 30 min) results in almost a complete cessation of motility. It is unclear whether the ultimate cessation of motility constitutes another phase of the behavioral response or a depletion of energy reserves (Wingrove, 1999).

These behavioral and cellular responses are diminished by an inhibitor of NO synthase and by a polymorphism affecting a form of cGMP-dependent protein kinase. Osborne (1997) identified the molecular basis for a behavioral polymorphism. Sitters (for^s) are less motile than rovers (for^R) while on food. This difference in foraging behavior is due to the allele state at a single locus, for. The gene dg2, which encodes one of two cGMP protein-dependent kinases (PKG) in Drosophila, is located at the for locus. The for^s stock contains an undefined polymorphism that eliminates expression of at least one dg2 transcript, and these stocks have slightly reduced PKG activity (Osborne, 1997). Thus, a partial reduction in PKG results in a behavioral change detected as decreased movement of larvae on a food source. It was thought that roving behavior might be related to the vigorous exploratory behavior observed during hypoxia. Larvae on a food source (yeast) are likely to suffer a reduction in oxygen availability due to metabolic competition with yeast, and roving might be a response to the reduced oxygen availability. Accordingly, the failure of for^s larvae to rove might result from a reduced behavioral response to hypoxia. The influence of the for alleles on the response to hypoxia were examined. Third instar larvae raised under noncrowded conditions were transferred onto a pile of yeast paste located at the center of a grape agar plate. Larvae were allowed to settle for approximately 10 to 15 min. Under these conditions, no obvious differences in behavior were observed between for^R and for^s larvae. Dishes were sealed and gassed with 1% oxygen through a hole in the lid. The initial response to hypoxia, cessation of feeding and movement to the surface, occurs in both stocks, although for^s larvae respond more slowly. These results indicate that PKG activity contributes to the exploratory response to hypoxia; however, during a prolonged mild hypoxia, for^s larvae eventually left the food (Wingrove, 1999).

NO activates PKG in mammals where it has recognized roles in the response to hypoxia. It was asked whether NO might similarly mediate responses to hypoxia in Drosophila. To do this, the ability of an inhibitor of nitric oxide synthase (NOS) to blunt the behavioral response to hypoxia was examined. Early third instar larvae that had been fed either L-NAME, an inhibitor of NOS, or D-NAME, the inactive isomer of L-NAME, were tested for behavioral responses to hypoxia. After 10 min of 1% oxygen, only 18% of the larvae fed L-NAME clear the yeast pile, compared to 53% of the larvae fed D-NAME. This suggests that NOS activity contributes to the induction of exploratory behavior by hypoxia. An inducible NOS transgene was used to increase NOS activity. Larvae carrying heat shock-inducible NOS (hs-iNOS) were either heat shocked for 20 min at 37°C to induce iNOS expression or left at 25°C as a control. As a second control, wild-type larvae lacking the transgene were subjected to a similar regime. After 60 min of recovery, oxygen was reduced to 10% for 15 min. This modest hypoxia induces only a low percentage of roving in the controls. In contrast, 60% of the larvae in the heat-shocked iNOS stock clear the yeast. Thus, induced iNOS makes the larvae hypersensitive to reductions in oxygen levels. These results suggest that NO and PKG play an important role in the ability of larvae to respond to hypoxia (Wingrove, 1999).

Regions of the larvae that specialize in responding to hypoxia should express significant levels of the activities involved. Two methods were used to define possible foci of function: a histochemical stain (diaphorase staining) for NOS activity, and DAF2/DA, a fluorescein derivative that increases in fluorescence when bound by NO. Diaphorase staining detects higher NOS activity in the imaginal discs. In addition, the CNS (optic lobes and ventral nerve cord; and sections of the gut stained well for diaphorase. The highest amount of diaphorase staining is observed in the spiricular glands of the posterior spiracles and the spiricular pouch of the anterior spiracles. DAF2/DA also stains the pouch of tissue surrounding the anterior spiracles as well as neuronal-like processes within the pouch. This staining near the openings of the tracheal system is interesting because the location is consistent with a possible role in governing the opening (eversion) of the spiracles to increase access to oxygen (Wingrove, 1999).

The behavioral responses observed presumably allow larvae to escape local hypoxia. Embryos, however, that are nonmotile and larvae that are exposed to a more general hypoxia do not have the option to escape. The capacity of embryos and larvae to endure prolonged hypoxia was tested. Early syncytial embryos are quite sensitive to hypoxia, but the ability to survive periods of hypoxia improves during cellularization and gastrulation. Eight-hr-old embryos can survive hypoxia for up to 8 days and larvae can survive for several days if protected from dehydration. Perhaps NO and PKG contribute to the ability of larvae and embryos to endure hypoxia. To test for an involvement of PKG, for^s or for^R embryos (stage 14) were abruptly exposed to severe hypoxia for different periods and the hatching and eclosion was followed under normoxic conditions. Both types of embryos survived 2 to 6 hr of hypoxia; however, while for^s embryos survive 12 hr of hypoxia poorly, for^R embryos survived well. It is concluded that PKG contributes to the survival of embryos exposed to hypoxia. The involvement of NO and PKG in the survival of larvae was also tested. When for^s larvae are kept hypoxic for 6 hr, they show little sign of motility upon return to normoxic conditions, and they rarely pupate or eclose. In contrast, for^R larvae rapidly reacquire motility after 6 hr of hypoxia, and 80% eclose. Larvae fed L-NAME also show diminished viability after hypoxia. It should be noted that the larvae in these experiments were subjected to severe and immediate hypoxia. When less severe conditions are imposed, little difference between for^s and for^R larvae is observed. Thus, it appears that NO and PKG contribute to the ability of Drosophila to endure suddenly imposed hypoxia (Wingrove, 1999).

Drosophila eggs ordinarily require about 24 hr to hatch. Hypoxic embryos, however, can arrest for days and then resume development. Presumably, surviving this arrest requires coordinate blocks to many embryonic processes, including cell proliferation. The influence of hypoxia on the well-defined cell cycles of the Drosophila embryo were examined. Cellularized embryos undergoing S phase of cell cycle 15 (3.5 hr old, stage 8 embryos) were placed in media equilibrated with either 1% or 22% oxygen via a bubbler. After 5 min, BrdU was added, and 5 min later the embryos were fixed and stained for BrdU incorporation, a measure of DNA replication. Hypoxia blocks BrdU incorporation. Since S phase length in cycle 15 (45 min) is longer than the treatment, the arrest indicates that the ongoing S phase is blocked. This block is reversed upon 5 min of reoxygenation. Different cells of the embryo progress through cell cycle 15 according to a strict developmental schedule, giving rise to stage-specific stereotyped patterns of S phase cells. Hypoxia does not appear to disrupt these patterns; upon reoxygenation, S phase resumes in the same domains as when it was blocked. Thus, hypoxia induces a rapid and reversible arrest of S phase (Wingrove, 1999).

Whereas severe hypoxia can arrest development, low oxygen levels can modify it. Development of the tracheal system shares many similarities with the development of the mammalian circulatory system. In both, the terminal branches elongate and ramify as growth increases oxygen demand and creates slight hypoxia. Perhaps NO and PKG contribute to this response to low oxygen. If NO and PKG stimulate tracheal ramification in low oxygen, it might be expected that terminal ramifications will decrease if these components are inhibited and increase if their expression is stimulated. Consistent with this, for^s larvae, which are genetically compromised for PKG activity, and L-NAME-treated larvae, in which NOS activity is inhibited, appear to have less frequent and shorter terminal ramifications. In contrast, induction of iNOS appears to have the opposite effect, increasing the number and the length of terminal branches. These results provide one example in which NO and PKG modulate an aspect of development (Wingrove, 1999).

These results implicate the involvement of NO and PKG in the response to hypoxia in Drosophila. This suggests that the responses observed are induced by a regulatory pathway resembling the one characterized in mammals. In the mammalian nervous system, this signaling process is triggered by calcium release, which stimulates the calcium/calmodulin-dependent activity of neuronal NOS. This results in NO production, which then stimulates cGMP synthesis in adjacent cells containing soluble guanylyl cyclase. In insects, this type of pathway is thought to play a specific role in signaling between specialized cells of the nervous system. Perhaps a similar signaling pathway is utilized in the more global responses observed during hypoxia (Wingrove, 1999 and references).

By virtue of its ability to stimulate vasodilation, NO has been implicated in the mammalian response to hypoxia. There is, however, no known mechanism by which hypoxia can induce a rapid (transcription-independent) increase in NOS activity. Furthermore, the synthesis of NO by NOS requires oxygen as a substrate, and rates of NO synthesis by NOS have been shown to decline under hypoxia. Nonetheless, these results suggest that NOS makes a positive contribution in the response to hypoxia. Perhaps there is sufficient oxygen to support NOS-catalyzed generation of NO during hypoxia. Alternatively, the activity of NOS prior to hypoxia might contribute to the release of NO during hypoxia. The NO produced by NOS under normoxic conditions rapidly reacts with molecular oxygen, resulting in the accumulation of NO2-. It has been suggested that hypoxia stimulates the release of NO from preformed stores of NO2-, which may be established by the earlier action of NOS. It has recently been shown that mammalian xanthine oxidase reduces NO3-/NO2- to NO under hypoxic conditions. Thus, under hypoxic conditions, accumulated NO2- might be reduced through the action of xanthine oxidase or other related enzymatic activities, resulting in the release of NO. Such a mechanism might have a general role in sensing oxygen levels (Wingrove, 1999 and references).

While these findings implicate NO and PKG in the response to hypoxia, their exact roles in the process are unknown. By analogy to the known roles of NO and PKG, they might act as direct transducers of the signal or they might function as facilitators of a separately transduced signal. A model is presently favored in which NO and PKG are involved in a signal tranduction pathway that is related to but perhaps different from the recognized NO signal transduction pathway (Wingrove, 1999 and references).

PROTEIN STRUCTURE

Amino Acids - 1350

Structural Domains

Comparison of the amino acid sequence of Drosophila NOS with sequences of mammalian NOSs reveals that DNOS is 43% identical to neural NOS, 40% identical to endothelial NOS and 39% identical to macrophage NOS. Structural motifs are conserved. The C-terminal half of the DNOS protein contains regions of high similarity corresponding to the presumptive FMN-,FAD-, and NADPH- binding sites. Amino acids thought to be important for making contacts with FAD and NADPH in mammalian NOSs are conserved in DNOS. The middle section of DNOS, between residues 215 and 746 show the highest similarity to mammalian NOSs. This region contains the presumptive heme- and calmodulin- binding sites in mammalian NOS enzymes. The heme-binding site is centered around cysteine C-415 of neural NOS, a residue important for heme binding. The region located between residues 643 to 671 of the fly protein contains the characteristics of a calmodulin-binding domain. DNOS has a protein kinase A consensus site at S-287 in a position similar to that of mammalian NOSs. The 214-amino acid N-terminal region of DNOS show no obvious similarity to its equivalent portion of nNOS. This region of DNOS contains an almost uninterrupted homopolymeric stretch of 24 glutamine residues (Regulski, 1995).

A 5.0 kb dNOS transcript is expressed preferentially in adult fly heads but not bodies. nNOS genes from mice and humans produce two alternatively spliced transcripts, the shorter of which yields a protein containing a 105-amino acid in-frame deletion (residues 504-608 in mouse or rat nNOS). Drosophila head mRNA reveals one species corresponding to the vertebrate long form, and a second species corresponding to the vertebrate short form. The two species demonstrate an alternative splicing pattern of NOS, conserved between vertebrates and flies (Regulski, 1995).

The nitric oxide synthase oxygenase domain (NOSox) oxidizes arginine to synthesize the cellular signal and defensive cytotoxin nitric oxide (NO). Crystal structures determined for cytokine-inducible NOSox reveal an unusual fold and heme environment for stabilization of activated oxygen intermediates, which are key elements for catalysis. A winged beta sheet engenders a curved alpha-beta domain resembling a baseball catcher's mitt with heme clasped in the palm. The location of exposed hydrophobic residues and the results of mutational analysis place the dimer interface adjacent to the heme-binding pocket. Juxtaposed hydrophobic O2- and polar L-arginine-binding sites occupied by imidazole and aminoguanidine, respectively, provide a template for designing dual-function inhibitors and imply substrate-assisted catalysis (Crane, 1997).

date revised: 15 Sept 99

Home page: The Interactive Fly © 1995, 1996 Thomas B. Brody, Ph.D.

The Interactive Fly resides on the
Society for Developmental Biology's Web server.