Mechanism of the Insect Enzyme, Tyramine β-Monooxygenase, Reveals Differences from the Mammalian Enzyme, Dopamine β-Monooxygenase*

Tyramine β-monooxygenase (TβM) catalyzes the synthesis of the neurotransmitter, octopamine, in insects. Kinetic and isotope effect studies have been carried out to determine the kinetic mechanism of TβM for comparison with the homologous mammalian enzymes, dopamine β-monooxygenase and peptidylglycine α-hydroxylating monooxygenase. A new and distinctive feature of TβM is very strong substrate inhibition that is dependent on the level of the co-substrate, O2, and reductant as well as substrate deuteration. This has led to a model in which tyramine can bind to either the Cu(I) or Cu(II) forms of TβM, with substrate inhibition ameliorated at very high ascorbate levels. The rate of ascorbate reduction of the E-Cu(II) form of TβM is also reduced at high tyramine, leading us to propose the existence of a binding site for ascorbate to this class of enzymes. These findings may be relevant to the control of octopamine production in insect cells.

The copper hydroxylases are a unique class of enzymes that are found in eukaryotes and play a critical role in the biosynthesis of neurotransmitters and hormones. The most studied enzymes in this family are peptidylglycine ␣-hydroxylating monooxygenase (PHM) 3 and dopamine ␤-monooxygenase (D␤M) (1). PHM catalyzes the conversion of C-terminal glycine-extended peptides to their ␣-hydroxylated products, the first step in the amidation of peptide hormones, required for a range of biological activities (2). D␤M catalyzes the hydroxylation of dopamine to yield norepinephrine and, thus, is vital for the regulation of these neurotransmitters (3)(4)(5). More recently, a third member of this enzyme family was identified, tyramine ␤-monooxygenase (6). T␤M is the insect homolog of D␤M, sharing 39% identity and 55% similarity with the mammalian enzyme. T␤M similarly catalyzes the hydroxylation of tyramine at the ␤-carbon position (Scheme 1). Although tyramine plays no role in mammalian physiology, the product of T␤M, octopamine, has been shown to act as a neurotransmitter in invertebrates, regulating physiological functions such as neuromuscular transmission, behavioral development, and ovulation (7)(8)(9). One advantage to studies with T␤M is the much more facile expression system for this enzyme in relation to D␤M (10), which makes it possible to pursue structure/function relationships. In this paper, we report a detailed analysis of the kinetic behavior of the wild type T␤M, an essential first step in understanding this complex enzyme system.
All three of the copper hydroxylases employ two noncoupled, mononuclear Cu centers for oxygenase activity, termed Cu M and Cu H . Structural information pertaining to the active site of the enzymes is derived primarily from the crystal structure for PHM and extended X-ray absorption fine structure studies with both D␤M and PHM (11)(12)(13). Each copper site assumes a unique coordination environment and a distinct mechanistic function. Cu M serves as the site of dioxygen binding and activation, whereas the Cu H site functions as an electron transfer site in the reaction mechanism (1,14). The Cu M site is coordinated by two histidine residues, a weakly bound methionine, and one or two water molecules, depending on the oxidation state. In the oxidized state, Cu H is ligated by three histidine residues and a water molecule in a tetrahedral geometry. Loss of a water molecule at Cu H accompanies reduction of the copper center, as for the Cu M site. The ligands to both copper centers are fully conserved among all three enzymes. Although little structural information is available for T␤M, the sequence and EPR spectra support an active site ligation identical to D␤M and PHM (10).
A mechanism for substrate hydroxylation by this class of enzymes has been proposed based on extensive kinetic characterization of D␤M and PHM (Scheme 2) (1). Product formation is coupled to the 2e Ϫ oxidation of substrate and the 4e Ϫ reduction of dioxygen to water, with one atom of O 2 incorporated into the final product. The copper atoms supply the two remaining electrons for this process. Ascorbate is required as a co-substrate to regenerate the reduced Cu(I) form of enzyme during turnover, with 2 mol of ascorbate converted to semidehydroascorbate during the one-electron reduction of each Cu(II) site. In the absence of direct evidence for an ascorbate binding site, an outer sphere electron transfer to both Cu M and Cu H has appeared to be the most likely mechanism for the reductive process. In the oxidative half-reaction, dioxygen reacts with the reduced Cu M center. The initial product of this reaction is believed to be a Cu(II)-superoxide species that subsequently abstracts a hydrogen atom from the bound substrate (15,16). A long range electron transfer from Cu H to Cu M is required to complete the reaction cycle, leading to formation of the hydroxylated product (17) and reoxidation of both copper centers.
Based on the high sequence similarity and initial reactivity studies (10), the reaction mechanisms of D␤M and T␤M were predicted to be identical. The detailed kinetic characterization of T␤M presented herein, including an examination of the dioxygen and ascorbate dependence in the oxygenase reaction and kinetic isotope effects, reveals several significant differences between the mechanisms of the insect T␤M and the mammalian D␤M. Substrate inhibition observed in the reaction mechanism of T␤M implies tighter regulation of neurotransmitter levels by the insect enzyme. The dependence of the rate of T␤M reduction by ascorbate on substrate level suggests, for the first time, a site for reductant binding within this class of enzymes.

EXPERIMENTAL PROCEDURES
Materials-Drosophila S2 cell growth medium was obtained from Invitrogen. Chromatography medium was purchased from GE Healthcare, except Talon affinity resin, which was purchased from BD Biosciences. 4-hydroxy-␤,␤-[2-2 H 2 ]phenethylamine hydrochloride (dideuterated tyramine) was synthesized as previously described (18), and the deuterium content was verified by NMR. Catalase was purchased from Roche Applied Science. All other reagents were purchased from Sigma.
Enzyme Preparation and Purification-T␤M was expressed in Drosophila Schneider 2 (S2) cells (Invitrogen) according to methods described previously (10). The protein was purified using anion exchange, His tag affinity, and size exclusion chromatography methods as reported (10), with the following modification. The T␤M-containing DEAE fractions were dialyzed against 10 mM KP i ϩ 0.25 M NaCl, pH 8.0, prior to binding to the Talon resin. Protein concentrations were determined by UV absorbance at 280 nm (⑀ 280 ϭ 1.34 ml mg Ϫ1 cm Ϫ1 ) using a monomer mass of 68,580 Da. Purified T␤M (at 3-5 mg/ml) was frozen in 50 -100-l aliquots in liquid nitrogen and stored at Ϫ80°C until further use.
Enzyme Kinetic Assays-The rates of oxygen consumption by T␤M were measured on a Yellow Springs Instruments model 5300 biological oxygen electrode. Assay conditions were similar to those described previously in kinetic assays with D␤M (19). Reaction mixtures (1 ml) contained 50 mM potassium phosphate (pH 6), 0.1 M KCl (to maintain ionic strength), 2 M CuSO 4 , and 40 g/ml catalase, T ϭ 35°C. Assay mixtures for the measurement of enzyme specific activities contained 250 M tyramine, 10 mM ascorbate, 0.16 M T␤M, and 222 M oxygen (air-saturated). An average specific activity for T␤M was determined for each set of experiments and used to standardize data collected over numerous days. The concentration of T␤M in the assay solutions varied from 0.1 to 0.4 M. The concentrations of tyramine (0.025-10 mM), ascorbate (2-200 mM) and dioxygen (11 M to 1.067 mM) were otherwise varied as indicated under "Results." The dioxygen concentration in assay solutions was varied by stirring reaction mixtures under a mixed O 2 /N 2 atmosphere of the appropriate proportion for ϳ10 min. The amount of dioxygen in solution was recorded as a voltage by the oxygen electrode and converted to units of molarity based on the known concentration of dissolved oxygen in air-saturated water (222 M at 35°C) or O 2 -saturated water (1067 M at 35°C). All reactions were initiated by the addition of enzyme. Initial velocities of dioxygen consumption by the enzyme were corrected for the ascorbate background rate and fit to the Michaelis-Menten equation, using the program Kaleidagraph. In cases where substrate inhibition was observed, the data were fitted to Equation 1, which includes a term for the inhibition constant, K i (20).
The values for k cat and K m and K i and errors in these values were derived from the resultant fits. The error in k cat /K m or KIE values was obtained through error propagation according to standard methods (21).

RESULTS
Initial kinetic studies indicated that the reaction of T␤M with tyramine followed simple Michaelis-Menten kinetics; a k cat of 1 s Ϫ1 and an apparent K m,tyr of 161 M were reported (10). Ferrocyanide was used as the reductant in these experiments. Ferrocyanide is an alternate electron donor for the copper monooxygenases and provides a convenient spectroscopic assay for determining enzyme activities (19). However, ascorbate is the anticipated in vivo reductant as reported for D␤M and PHM (2,19). In the present studies, all kinetics of T␤M were pursued using ascorbate as the reductant. This allows the direct comparison of kinetic data obtained with T␤M with values previously reported in the D␤M and PHM reactions. The use of ascorbate also provides a frame of reference for understanding the physiological properties of T␤M.
Initial velocities were measured by the rate of O 2 consumption by T␤M at varied tyramine (25 M to 10 mM) and 10 mM ascorbate in air-saturated solutions at 35°C. From the fit of the data, the apparent K m for tyramine was determined as 87.6 Ϯ 12.7 M, comparable with the value reported previously for the ferrocyanide assays (10) and to the K m obtained in previous activity assays of T␤M-containing homogenates from different organisms, using ascorbate (8). The maximum turnover rate under these conditions was calculated as 3.91 Ϯ 0.22 s Ϫ1 , ϳ4-fold greater than the initial value reported.
A surprising result was the decrease in rates observed at tyramine concentrations of Ͼ250 M (Fig. 1A), indicative of substrate inhibition. This behavior was not evident in the preliminary studies using ferrocyanide; nor has substrate inhibition ever been demonstrated in kinetic studies with the analogous enzymes, D␤M or PHM. The fit of the initial rate data to Equation 1 yielded an apparent substrate inhibition constant, K i,tyr , of 3.5 Ϯ 0.56 mM.
The effect of fumarate on the reaction also was investigated, since fumarate acts as an anion activator in the D␤M reaction mechanism (22). Consistent with the initial findings (10), fumarate had no effect, within experimental error, on either k cat , K m,tyr , or K i,tyr and was subsequently omitted from all assay solutions.
Ascorbate Dependence-The early kinetic characterization demonstrated a higher substrate specificity for T␤M in comparison with D␤M (10). This fact and the comparatively low observed k cat value (D␤M ϭ 12.7 Ϯ 0.5 s Ϫ1 (dopamine); PHM ϭ 39.1 Ϯ 0.5 s Ϫ1 (hippuric acid) (18,23)) prompted us to investigate the ascorbate dependence on the T␤M reaction rates. Initial rates of oxygen consumption by the enzyme were measured as a function of tyramine, at fixed ascorbate concentrations ranging from 2 to 200 mM ( Fig. 1B and Table 1). The rate of turnover, k obs , increases nearly 10-fold in this range, from a value of 1.3 s Ϫ1 at low ascorbate to a maximum value of 12 s Ϫ1 at 100 mM ascorbate. The latter value more closely approximates the substrate turnover rates determined for D␤M. The effect of increasing ascorbate on k obs and K m,tyr results in a minor descending trend in the apparent k cat /K m,tyr (Table 1). However, with the exception of the data at 2 mM ascorbate, k cat /K m,tyr appears to be independent of ascorbate concentrations, indicative of a ping-pong mechanism, where ascorbate and tyramine interact with different forms of enzyme, Cu(II) and Cu(I), respectively. The slightly inflated value at 2 mM ascorbate probably reflects the poorer signal/noise ratio of the rate data at such low ascorbate concentrations.  Noteworthy is the effect of ascorbate on substrate inhibition. Increasing ascorbate attenuates the observed inhibition at high tyramine concentrations ( Table 1). The tyramine K i (app) varies from ϳ1-2 mM at low ascorbate to 10 mM at high ascorbate. Since the role of ascorbate is to regenerate the reduced form of the enzyme during turnover, the effect of ascorbate on K i,tyr implies that the inhibitory pathway is minimized when more enzyme is present in the reduced state.
Initial velocity data were obtained in experiments with varied ascorbate (2-200 mM) at fixed tyramine concentrations ranging from 50 to 700 M. The resultant fit of the data to the Michaelis-Menten equation provided an estimated limiting K m for ascorbate of Ն16 mM ( Fig. 2 and Table 2). This value represents a lower approximation due to difficulties in attaining levels of saturating tyramine, as a result of the substrate inhibition. However, substrate inhibition is negligible in the chosen range of tyramine concentrations. The inhibitory region is already reached at 1 mM tyramine, regardless of the ascorbate concentration. Therefore, the K m,asc value obtained from these experiments provides a reasonable estimate.
A small decrease in rates also is observed at high ascorbate concentrations. The kinetic data obtained at 300 and 700 M tyramine can be fit easily to the Michaelis-Menten equation. However, at lower tyramine, the data are best fit to Equation 1 containing the additional term for K i . Thus, although elevated ascorbate mitigates the observed inhibition by tyramine, the reductant itself can act as an inhibitor. The inhibition by ascorbate is not very strong, however, with a K i,asc of Ͼ1 M even at low tyramine. A more significant trend is evident in the apparent k cat /K m,asc , which decreases by a factor of 2.5 in the range of tyramine concentrations examined, showing that high levels of tyramine interfere with the second order reaction between the Cu(II) form of T␤M and reductant. A plot of K m /k cat,asc versus tyramine concentration yields a good correlation (r ϭ 0.995) and an estimate for K i,tyr ϭ 0.43 mM in the limit of very low ascorbate.
Oxygen Dependence-The kinetics of the reaction with tyramine at several fixed dioxygen concentrations in the range of 5-100% O 2 subsequently were examined (cf. Fig. 3 for data in the presence of 10 mM ascorbate). These experiments allowed us to ascertain the dependence of the apparent kinetic parameters k cat , K m,tyr , and k cat /K m,tyr on dioxygen and to establish a K m for O 2 . The full range of oxygen-dependent kinetic studies focused on two different ascorbate concentrations, 10 and 50 mM (Table 3), since the dependence of the reaction rates on the reductant concentration already had been established.
The apparent k cat was obtained from the fit of the initial velocity data as a function of tyramine, at each concentration of    for data collected with tyramine as the substrate) (Fig. 4) and appears to be independent of ascorbate. This value is comparable with the K m ,O 2 obtained for D␤M (18). Increasing dioxygen results in a regular increase in the apparent k cat /K m,tyr , consistent with a ternary complex mechanism. Contrary to the effect of ascorbate, dioxygen amplifies substrate inhibition. K i,tyr varies from 8 mM at 5% O 2 (60 M) to 1 mM at 98% O 2 (850 M) in reactions with 10 mM ascorbate. A 6-fold reduction in K i,tyr is observed in experiments using 50 mM ascorbate. This result is consistent with the premise that elevated ascorbate concentrations diminish tyramine inhibition due to a decrease in the amount of oxidized enzyme present. In contrast to ascorbate, higher oxygen concentrations favor formation of the oxidized, Cu(II) form of the enzyme and result in greater inhibition via formation of the abortive, E-Cu(II)-tyramine complex.
Kinetic Studies with Tyramine-d 2 -To obtain further insight into the kinetic mechanism of T␤M, initial velocities were measured as a function of dioxygen and dideuterated tyramine. A range of dideuterated tyramine concentrations (25 M to 10 mM) and oxygen concentrations (60 M to ϳ1040 M) were analogous to those in experiments with protiated tyramine. Data were collected at 10 and 50 mM ascorbate ( Table 4). Replots of the data provided the limiting D k cat ϭ 2.58 Ϯ 0.64 (10 mM ascorbate) and 2.75 Ϯ 0.5 (50 mM ascorbate). A value of 1.85 Ϯ 0.5 was calculated for D k cat /K m,tyr (50 mM ascorbate, 97% O 2 ). The substrate inhibition made it difficult to obtain D k cat / K m,O 2 with any great accuracy, but values appear to be in the range of values obtained for D k cat /K m,tyr . The deuterium isotope effect on k cat appears somewhat larger than the D k cat /K m value. These trends are different from D␤M, where D k cat Ͻ D k cat /K m in the absence of the anionic activator (22). An isotope effect on K i,tyr is evident as well. A decrease in K i is observed as a result of increasing oxygen in reaction mixtures containing deuterated tyramine, similar to the effect observed using protiated tyramine. However, the K i value for tyramine-d 2 displays a narrower range than K i,tyr . Under conditions where dioxygen is increased by ϳ16-fold (ϳ60 M to 1 mM), K i,tyr-d 2 varies from 3 to 2 mM at 10 mM ascorbate and from 7 to 3 mM at 50 mM ascorbate. This represents a smaller decrease in K i,tyr-d 2 , compared with the 6 -8-fold decrease in K i,tyr observed under analogous conditions. The isotope effect on K i can be interpreted in the context of a reduced steady state concentration of the Cu(II) form of enzyme, due to the slower abstraction of deuterium from the labeled substrate and, hence, an elevated level of the E-Cu(I)-tyr-O 2 complex.

DISCUSSION
Oxidative Mechanism-The T␤M oxidative half-reaction seems to proceed in accord with the mechanisms established for the mammalian enzymes, as determined from isotope effect studies. An isotope effect on k cat of ϳ2.7 indicates that hydrogen abstraction from the substrate is partially rate-limiting, as demonstrated for D␤M and PHM (23,24). Values for D k cat / K m,tyr appear to be virtually independent of ascorbate, which again points to an irreversible step separating reduction of the enzyme and the subsequent chemistry. A plot of the dependence of D k cat /K m,tyr on dioxygen is shown in Fig. 5 as a compilation of data at 50 mM ascorbate. Although the errors are fairly large, there is a decreasing trend as the O 2 concentration increases from ϳ60 M to 1 mM. The fact that D k cat /K m,tyr remains above unity at the highest oxygen concentration is consistent with an ability of tyramine to dissociate from the E-Cu(II)-tyr-O 2 complex (21). The intercepts of the 1/v versus 1/O 2 reciprocal plots at varied tyramine intersect to the left of the 1/v axis (Fig. 6 for 50 mM ascorbate data), in support of a random mechanism where either tyramine or O 2 can bind first to the reduced enzyme. It would appear that there is no obligatory binding of substrate before O 2 in the T␤M reaction. The observed dependence of D k cat /K m,tyr on O 2 also is consistent with a random mechanism. The full expression for the isotope effect in a random kinetic mechanism is illustrated in Scheme 3 and Equation 2, where k 5H /k 5D ϭ Int represents the intrinsic isotope effect (25).
The full expression simplifies according to Equations 3 and 4 in the limits of low and high oxygen, respectively.
The larger value for k 5H /(k 4 ϩ k 4Ј ) in Equation 3 versus k 5H /k 4Ј in Equation 4 leads to the observed decrease in D k cat /K m in Fig.  5. The oxidative mechanism of T␤M is contrary to the oxidative mechanism of PHM, which is equilibrium ordered with respect to binding of substrates (23), and more closely resembles the reaction of D␤M with substrates in the absence of anion activators (21). The isotope effect on k cat /K m,tyr in the limit of very low oxygen levels remains quite small in relation to previously determined intrinsic isotope effects for D␤M and PHM, 10.9 Ϯ 1.9 (dopamine) and 10.6 Ϯ 0.8 (hippuric acid), respectively. This indicates that the barrier for dissociation of bound tyramine is kinetically significant relative to that for C-H abstraction, again consistent with previous studies of D␤M and PHM.
Reductive Mechanism and Tyramine Inhibition-Based on the present data, we now describe a mechanism for T␤M, which exhibits several notable differences from the mechanisms established for the related mammalian enzymes, D␤M and PHM. The most remarkable distinction is the substrate inhibition observed for the reaction of T␤M with tyramine. As mentioned earlier, this behavior was not observed in the hydroxylation reactions of either D␤M or PHM. The substrates bind solely to the reduced form of the latter enzymes; reaction of the substrate-bound reduced enzyme complex with dioxygen subsequently leads to product formation. In contrast to the mammalian systems, our results support a mechanism in which tyramine also is able to bind to the oxidized form of T␤M. The interaction of tyramine with the oxidized form of the enzyme leads to an inhibitory complex, notated as E-Cu(II)-tyr.
The ability of tyramine to form an unproductive complex with the oxidized enzyme explains the effect of ascorbate on substrate inhibition. As the concentration of the reductant is increased, a greater amount of enzyme is present in the reduced form, such that binding of tyramine preferentially leads to turnover rather than formation of the E-Cu(II)-tyr complex. This is reflected in the large K i,tyr values observed at elevated ascorbate ( Table 1).
The dioxygen dependence and the isotope effect on substrate inhibition further support our conclusions. Contrary to the effect of ascorbate, high dioxygen concentrations increase the steady state level of oxidized enzyme, and formation of the E-Cu(II)-tyr complex becomes more probable. Lower K i,tyr values, consequently, are observed at high oxygen (Table 3). In the reaction with dideuterated tyramine, hydrogen abstraction becomes more rate-limiting, which again increases the steady state amount of reduced enzyme present. Under these conditions, the interaction of tyramine with the oxidized   Mechanism of the Insect Enzyme, Tyramine ␤-Monooxygenase FEBRUARY 8, 2008 • VOLUME 283 • NUMBER 6 enzyme is reduced relative to the protio-substrate, leading to a more narrow range of K i,tyr as the oxygen is varied (Table 4). In effect, any condition that alters the steady state amount of reduced or oxidized enzyme is manifest in the substrate inhibition.
Furthermore, it appears that tyramine competes with ascorbate for interaction with the oxidized form of T␤M. The notion of two competing pathways available to E ox is corroborated by the effect of substrate on k cat /K m,asc ( Table 2). In a ping-pong mechanism, as established for D␤M and PHM, the values for k cat /K m,asc are independent of the substrate concentration. Our data show a tyramine-dependent variation in k cat /K m,asc for the reaction of T␤M and indicate that tyramine interferes with the function of the reductant. The competition by tyramine may arise from two possible origins; either tyramine binding interferes with the formation of a reduced T␤M-ascorbate complex, or bound tyramine reduces the rate of outer sphere electron transfer from ascorbate. Both of these effects are likely to be focused at the Cu M site, given the absence of a substrate binding site near Cu H in PHM (26). The ability of tyramine to obstruct outer sphere electron transfer to the Cu sites could only occur as a result of changes in the redox properties of the Cu M site upon binding of substrate. This scenario seems unlikely, given that the substrate does not coordinate directly to the metal; nor are changes in the copper coordination geometry observed upon binding of substrate to PHM or D␤M. We therefore present the alternate explanation of an ascorbate binding site in T␤M. This concept is highly novel, given the absence of any evidence for direct binding of ascorbate to the mammalian enzymes. Ascorbate binding near Cu M would facilitate electron transfer to the oxygen-reactive metal center. However, a new mechanism is required to account for the reduction of both Cu M and Cu H during the reductive half-reaction. One explanation would be an outer sphere electron transfer from a second ascorbate molecule to the Cu H site. An alternative mechanism involves an electron transfer from Cu M to Cu H during the reductive half-reaction; studies on D␤M indicate a similar redox potential at both copper sites (27) such that an electron would be expected to equilibrate between the two sites if the rate of electron transfer were sufficiently fast. This corresponds to the reverse scenario of the electron transfer step requisite in the oxidative mechanism, with the caveat of a larger driving force for Cu H 3 Cu M electron transfer during substrate hydroxylation. Our results provide the first indication of a possible bidirectional electron exchange between the two copper centers.
Although tyramine interferes with ascorbate for the oxidized form of the enzyme, ascorbate does not seem to interact appreciably with the reduced form of T␤M. The apparent values for k cat /K m,tyr are fairly constant over the range of ascorbate concentrations evaluated (Table 1). Minor ascorbate inhibition is observed only at low tyramine concentrations. Given a K i,asc of greater than 1 M, any interference by ascorbate is irrelevant under physiological conditions. The oxidative half-reaction is therefore largely independent of ascorbate, as expected for a ping-pong mechanism. The lack of appreciable ascorbate interaction with the reduced T␤M signifies two unique active site conformations, such that ascorbate is precluded from interact-ing with any of the reactive species generated subsequent to formation of E-Cu(I)-tyr-O 2 .
Conclusions-The initial kinetic characterization of T␤M (10) demonstrated a much higher substrate specificity compared with D␤M. It is surprising, in light of this fact, that tyramine is able to bind to both the Cu(II) and Cu(I) form of T␤M. The ability of the enzyme to discriminate against dopamine would have indicated a far more constrained active site. Prior studies have demonstrated that a change in coordination geometry accompanies redox changes at both copper centers of D␤M and PHM (12). The differing substrate affinities of the reduced and oxidized T␤M suggest further extended conformational changes of the protein scaffold. Despite a high sequence homology between T␤M and D␤M, the disparate regions of the core may be the key to the observed differences in the reactivity of the enzymes.
The high K m value for ascorbate (Ͼ16 mM) seems remarkable, given that considerably lower concentrations of reductant (Ͻ10 mM) are required to attain maximal turnover rates for both D␤M and PHM (28). Ascorbate concentrations in vesicles, such as the chromaffin granules and synaptic vesicles where D␤M is localized, can be as high as 20 mM. It is not known definitively whether T␤M is expressed in similar vesicles of the invertebrate nervous system. Octopamine, however, is associated with dense core vesicles of dorsal unpaired median neurons in locusts (29,30), and studies also have shown that T␤M is localized in octopaminergic neurons/cell bodies (of honeybees and lobster) (8,9,31). A recent study demonstrated that octopamine-containing cells of the Drosophila reproductive system, including vesicles of the ovarian sheath, were immunoreactive for T␤M. It is therefore likely that T␤M is expressed in vesicles, similar to D␤M, consistent with the requirement for high ascorbate concentrations.
The high K m,asc observed for T␤M is not unprecedented. Two distinct ascorbate K m values previously were described for D␤M (69 M and 37.2 mM) (28); the affinity of the enzyme for the reductant was highly dependent on pH and the presence of fumarate. Although the results of this earlier study were not fully understood, one interpretation for the varying ascorbate effects involved competition between ascorbate and fumarate for a binding site within the enzyme. Fumarate has no bearing on the reaction of T␤M with tyramine, and much higher levels of ascorbate are required to observe activity by the insect enzyme, excluding the possibility of a low secondary K m value. However, the current evidence suggesting an ascorbate binding site in T␤M may warrant a reinvestigation of the role of the reductant in the mammalian systems as well.
Finally, we note that relatively minor changes in the levels of all three substrates (ascorbate, tyramine, and dioxygen) appear to have dramatic effects on the observed rate constant. Recent studies have demonstrated that tyramine and octopamine have antagonistic effects and suggest that behavioral regulation may depend on the balance of these two hormones (7,32). Thus, T␤M may be expected to function under k cat /K m conditions for all three substrates in vivo, making the enzyme exquisitely sensitive to small shifts in cellular conditions. Overall, the T␤M kinetics data imply tighter regulation of neurotransmitter levels by the insect enzyme than in the mammalian homologue.