Insect Cardioactive Peptides: II. Neurohormonal Control of Heart Activity by Two Cardioacceleratory Peptides In the Tobacco Hawkmoth, Manduca Sexta

The accompanying paper (Tublitz & Truman, 1985) described the distribution and some of the molecular properties of two cardioacceleratory peptides (CAPs) found in the central nervous system of the pharate adult stage of Manduca sexta. These peptides were defined by their elution characteristics on Sephadex G-15, by their sensitivity to proteolytic enzymes, and because they elicited an increase in rate when pulse applied to the isolated Manduca heart. The finding that the CAPs were primarily localized to the perivisceral organs (PVOs), the major neurohaemal release sites of the insect ventral nerve cord, was consistent with the hypothesis that the two peptides functioned as physiological modulators of cardiac activity. However, physiological data to support this hypothesis were lacking. This paper presents several lines of evidence that the two CAPs act as cardioregulatory neurohormones in Manduca sexta.

Gel filtration procedures and the rearing of the animals, Manduca sexta, used in the following experiments were described in a previous paper (Tublitz & Truman, 1985). CAP bioactivity was determined using the isolated Manduca heart bioassay (Tublitz & Truman, 1985).

In vivo heart recordings from intact pharate adult and adult Manduca males were accomplished using an impedance converter (Model 2991, Biocom Corporation). The basis of this device is a load-sensitive 50 kHz oscillator, which responds to a change in impedance at the recording electrodes by changing the amplitude of the oscillation. This change in amplitude is then converted into an output voltage (Miller, 1973).

Animals were anaesthetized with CO₂ and two holes were made with an insect pin into a descaled portion of the metathoracic medial scutellar plate. The recording leads from the impedance converter were inserted into the holes in the cuticle and fixed into place on either side of the heart using melted beeswax. The signal from the impedance converter was amplified and recorded onto a Gould 2200 pen recorder for later analysis.

After recovering from the CO₂ anaesthesia, animals with the implanted electrodes were placed into a cubic wire mesh container, measuring 15 cm on each side. This set-up effectively restrained the animal and allowed for in vivo heart recordings during eclosion and subsequent wing-spreading behaviour. Continuous recordings were taken for up to 72 h.

Haemolymph, taken from animals in various developmental stages, was utilized for several different experiments. In all cases, haemolymph was obtained by inverting a decapitated animal into an ice-cold glass vial. After collection, haemolymph was subjected to heating at 80 °C for 10 min followed by a 3 min spin at 9000xg in a Beckman microfuge. The supernatant was then removed and either bioassayed immediately or stored at −20°C for later testing.

For those experiments requiring gel filtration, the supernatant taken from the haemolymph of two to three animals was pooled, heat-treated, and immediately layered onto a 1·6x30cm chromatography column containing Sephadex G-15. The details of the chromatographic procedure were as described by Tublitz & Truman (1985). One and a half millilitre fractions were collected, lyophilized, rehydrated with Manduca saline and sequentially bioassayed for CAP bioactivity on the isolated Manduca heart.

Tissues were dissected out, pooled, heat-treated at 80°C for 10 min and homogenized in a small, ground glass vessel in 1·0 ml of 0·1 moll⁻¹ acetic acid. The homogenate was then subjected to a 3-min spin in a Beckman microfuge to pellet cellular debris. The supernatant was immediately layered onto a Sephadex G-15 column and the resulting fractions were assayed for CAP activity on the in vitro Manduca heart as previously described (Tublitz & Truman, 1985).

Intact pharate adults were implanted with impedance converter electrodes and in vivo heart recordings were taken as described above. Samples of G-15 purified CAP₁ or CAP₂ were dissolved in Manduca saline and injected into the abdomen using a 100μl Hamilton syringe. Injection volumes were usually 50 μ1 or less and never more than 100 μ1. All animals received a saline control injection prior to testing. Most animals received only a single test injection, and no individual was subjected to more than three sequential test injections Individuals received more than one test injection only if heart rate returned to and remained at basal levels for a 5-min period after each trial. All samples were tested during the anterior-going coordination mode of the in vivo heart (defined in the Results).

In vivo recordings of the Manduca heart from pharate adult and adult animals were obtained using an impedance converter and the results are summarized in Fig. 1. Prior to adult emergence, the behaviour of the heart was characterized by two, alternating modes of coordinated activity: a rapid beating mode during which the heartbeat frequency was approximately 38 beats min⁻¹ (range 31-43 beats min⁻¹), followed by a second, distinct mode when the heart was beating at a much slower rate (mean, 18 beats min⁻¹; range, 2-30 beats min⁻¹). Behavioural observations of the pharate adult heart indicated that these alternating activity modes corresponded to the phenomenon known as ‘heartbeat reversal’, common to many insects including lepidopterans (Malpighi, 1669; Wigglesworth, 1972; Wasserthal, 1976). The accelerated coordination mode occurred only when the metachronal wave of contraction of the heart proceeded in an anterior direction (i.e. the contractile wave was initiated at the posterior end of the heart tube and progressed anteriorly) and will subsequently be referred to as the anterior-going coordination mode of the heart. Each anterior-going mode typically lasted for several minutes, and terminated abruptly as the entire heart tube ceased active contractions for 5-15 s. After the brief quiescence, the heart proceeded into the second and slower mode, characterized not only by a decreased contraction rate but also by a posteriorly-directed contractile wave that commenced anteriorly and terminated in the last abdominal segment. This second period of cardiac behaviour will subsequently be referred to as the posterior-going coordination mode, and had a duration of 2-4 min (Fig. 1).

The heart abruptly changed its behaviour at the time of eclosion. Immediately after eclosion and for the next 90 min the heart behaviour was characterized by a sustained increase in heart rate, anterior-going beating, and a lack of heartbeat reversals. At eclosion the frequency of contractions increased suddenly, from a preeclosion average of 26 beats min⁻¹ (mean rate of both anterior and posterior coordination modes for the 30th min prior to eclosion) to a peak rate of 68 beats min⁻¹ during the first 60 s following adult emergence. Heart rate declined slowly over the subsequent 90 min, during which time the newly emerged animal proceeded to inflate its wings to their normal adult length followed by wing folding (Truman & Endo, 1974). Behavioural observations confirmed that at the start of eclosion the heart commenced beating in an anterior-going direction, which was maintained throughout eclosion and wing-spreading behaviour. After the successful completion of eclosion and wing spreading, the heart reverted back to its preemergence behaviour, complete with alternating anterior- and posterior-going coordination modes (Fig. 1).

Blood samples, taken from pharate adults or wing-spreading animals, were individually collected and bioassayed on the isolated Manduca heart to determine the presence or absence of cardioactive factors (see Fig. 2). Whole blood from pharate adults produced a marked decrease in heart rate. Of the twenty blood samples taken from pharate adults, all contained substantial amounts of inhibitory activity, varying from an 8% to a 35% decrease in frequency, with a mean decrease of 21%.

In contrast, blood taken from animals 5 min after wing spreading had begun showed significant levels of cardioacceleratory activity (Fig. 2). Each sample of wing-spreading blood produced an increase in heart rate (N =20), with a mean increase of 28% (range: 14-39%) over the basal, unstimulated rate of the isolated heart. These data suggested the presence of one or several cardioacceleratory factors in the blood of wing-spreading animals.

To analyse the molecular characteristics of the humoral cardioacceleratory factor(s) found in the blood during wing spreading, pooled blood from two to three wing-spreading animals was subjected to gel filtration on Sephadex G-15 and the resultant fractions bioassayed for cardioacceleratory activity. Chromatographic profiles of blood from wing-spreading animals such as the one depicted in Fig. 3 indicated the presence of three, discrete peaks of cardioactivity: two excitatory peaks and a single, inhibitory peak. The first peak of cardioacceleratory activity had a V_e/V_o value of 1·5 and appeared to be truncated by the inhibitory activity in later fractions. The second peak may also have been slightly masked by the inhibitory material and eluted with aV_e/V_ovalue of 2·4. The two cardioacceleratory peaka showed elution times similar to that of CAP₁ and CAP₂. The activity in each peak was stable to heating at 80°C for 10 min. The sum of cardioacceleratory activity present in the two excitatory peaks accounted for most, if not all, of the cardioacceleratory activity present in the whole blood of wing-spreading animals. Neither octopamine nor 5-hydroxytryptamine (5-HT) were detected in blood as determined on the bioassay following gel filtration.

The activity of the cardioinhibitory peak was quite potent and eluted on the Sephadex G-15 column after the first and prior to the second acceleratory peak with a V_e/V_o value of 1·8. The elution time of the inhibitory factor suggested a molecular weight of approximately 200-300 Da. The activity associated with this peak was stable to heating. The inhibitory peak did not co-elute with any substance found in the pharate adult VNC, and at present, its site of origin and release is unknown.

All three cardioactive peaks, the two cardioexcitatory and single cardioinhibitory, were present in each sample of wing-spreading blood that was analysed (N = 6).

Pharate adult blood was also subjected to gel filtration chromatography using the identical protocol to that employed with wing-spreading blood. Bioassay of pharate adult blood chromatographed on Sephadex G-15 showed the presence of a large, cardioinhibitory component and a single, minor cardioexcitatory peak. The activity in the inhibitory peak co-eluted with the peak found in blood from wing-spreading animals. The single acceleratory peak partially co-eluted with CAP₂ and the second excitatory peak in blood from wing-spreading animals. This peak was not always detectable, being present in only two-thirds of the samples tested (N = 6). In those samples in which it was present, the amount of activity in the blood of pharate adults varied from 2 to 9% when compared to the mean level of the factor found in wing-spreading blood.

Data in the previous section indicated that the CAPs appear in the blood during wing-spreading behaviour. Therefore, it was of interest to determine the time course of their appearance. Whole blood was taken from individuals at various times after adult eclosion, processed and bioassayed on the isolated Manduca heart (Fig4).

There was no measurable CAP activity in the blood prior to adult emergence. Within a few seconds after the successful completion of eclosion, however, CAP activity was detected (time = 0 in Fig. 4). Although these blood samples produced no change in the in vitro heart rate, it was indicative of the presence of significant levels of CAP activity since pre-emergent blood contained substantial cardioinhibitory activity. The CAP level rose dramatically during the next few minutes, with the peak blood titre occurring at 5 min after eclosion. This time coincided with the time of initiation of wing-spreading behaviour. The CAP titre declined slowly throughout the wing inflation period, so that by the time the wings were fully expanded and in their proper adult position, at 90 min post-emergence, the blood CAP levels were barely detectable, and their combined effect on heart rate was negligible.

The relationship between the onset of wing spreading and the release of the two CAPs into the blood was further examined by confining newly emerged adult Manduca in glass vials. This treatment delays wing-spreading behaviour until the insects are released from confinement (Truman & Endo, 1974). Blood was taken from these animals by decapitation at various times after confinement and treated as described above. Each of these blood samples produced a slowing of the isolated Manduca heart (Fig. 5), characteristic of the cardioinhibitory activity present in pharate adult blood. This decrease in heart rate averaged 21 ±4% (±S.E.M., N =30).

An additional group of animals was confined for 60 min, after which they were removed from the glass vials and allowed to commence wing-spreading behaviour.

Five minutes later, they were decapitated and their blood individually assayed for CAP activity. All animals showed significant levels of cardioacceleratory activity in their blood (Fig. 5) with a mean increase in heart rate of 24 ± 7% (±S.E.M., N= 10). These data indicate that the release of the CAPs into the blood is associated with wing-spreading behaviour and is delayed by factors that delay the onset of this behaviour.

Although the CAPs are released into the blood at the time of wing spreading, the above data do not show that this blood-borne, cardioacceleratory activity comes from the VNC. Utilizing gel filtration chromatographic techniques, the levels of the two CAPs in pharate adult VNCs were measured and contrasted to CAP levels in the VNC of day 1 adults that had emerged and successfully completed wing spreading during the previous day. Chromatographed material from the day 1 adults showed significantly lowered amounts of both CAPs compared to the levels found in the pharate adult VNC (Fig. 6). After adjusting for the log linear relationship of the dose-response curve (Tublitz & Truman, 1985), changes in the area under each peak indicated that the stored levels of CAP₁ and CAP₂ declined by 94% and 96%, respectively. VNCs from a second group of day 1 adults showed an 88% and 93% decrease of CAP₁ and CAP₂ levels, respectively. These results show that a massive depletion in the stored levels of the two CAPs occurs during the first 24 h after adult eclosion, a period that encompasses the time that CAP activity appears in the blood during wing-spreading behaviour.

Twenty abdominal nerve cords including PVOs were dissected from pharate adult males, blotted dry and pre-incubated in a low calcium, normal potassium saline (Table 1) for 1 h at 0°C to allow the cords time to equilibrate at the lowered calcium level. They were then transferred to 1·0 ml of a low calcium, high (6X) potassium saline (Table 1). After a 1-h incubation at 0°C, they were then placed in 1·0 ml of a normal calcium, high potassium saline (Table 1) for an additional hour at 0°C. Each of the two incubation solutions was then centrifuged at 1000×g to remove cellular debris, subjected to gel filtration and analysed for CAP₁ and CAP₂ activities on the isolated Manduca heart.

As shown in Fig. 7, significant levels of CAP bioactivity were released by the VNC only during incubation in normal calcium, high potassium saline. Two peaks of cardioexcitatory activity were released which co-eluted with the two CAPs. Neither peptide was detected in the low calcium, high potassium media. The amount of CAP₁ and CAP₂ recovered from the normal calcium, high potassium incubation was 2% and 3%, respectively, of the total amount of each peptide stored in the pharate adult VNC. The fact that high potassium stimulation evokes release of both CAPs in a calcium-dependent manner suggests that these peptides are co-released during the endogenous activation of the CAP system.

Pharate adults, previously implanted with impedance converter electrodes, were injected with varying concentrations of G-15 purified CAPi or CAP₂ and monitored for changes in heart rate. All samples were tested when the heart was in the anteriorgoing coordination mode. Injection of either peptide produced a dose-dependent increase in heart rate (Fig. 8) and higher doses tended to delay the onset of the posterior-going coordination mode. Cardioregulatory responses were not detected when amounts less than 0-05 ANC équivalents of either peptide were injected (1 ANC equivalent = the total amount of either peptide found in the abdominal portion of the ventral nerve cord of a single, pharate adult). Injection of 1 ANC equivalent of CAP] and CAP₂ induced a 22% or 39% increase, respectively, in the frequency of contraction of the in vivo heart. Control injections of the saline carrier did not elicit any change in heart rate. The dose-response curves of the in vivo heart to increasing concentrations of CAPi or CAP2 were similar to those seen when either crude extracts or partially purified samples were bioassayed on the isolated Manduca heart (Tublitz & Truman, 1985). Application of 1 ANC equivalent of CAP₂ to the in vitro heart produced a 37% increase in frequency.

Malpighi (1669) was the first to report heartbeat reversals in the silk moth Bombyx mori, a phenomenon now known to occur in many insect species (Wigglesworth, 1972). It has been most intensively studied in lepidopterans, where regular reversals have been observed in the pupal and imaginai stages in a number of different species (Gerould, 1924, 1929; Wasserthal, 1976). The data from the in vivo heart recordings presented in this paper show that periodic anterior- and posterior-going peristaltic contractions of the Manduca heart occur regularly at the end of metamorphosis and again in the adult. These results are in good agreement with those obtained in the butterfly Caligo brasiliensis (Wasserthal, 1975), and in the moths Bombyx mori (Yokohama, 1927), Attacus atlas (Wasserthal, 1975) and Mamestra brassicae (Queinnec & Campon, .1972), all of which show a periodic reversal of the contraction of the heart in both the pharate adult and the adult.

The observation that the Manduca heart radically changes its behaviour in vivo during adult emergence and subsequent wing inflation is crucial to the interpretation of the remainder of the data presented in this paper. At emergence the average rate of cardiac contractions increases by twofold (Fig. 1) and heartbeat reversals cease. These changes are maintained throughout the period in which the wings are inflated, and it is only after the successful completion of wing spreading that the heart returns to its pre-emergence pattern. Similar changes in the patterned behaviour of the heart have been described in other lepidopterans including the moths Attacus atlas and Mamestra brassicae (Wasserthal, 1975; Queinnec & Campon, 1972).

The regulation of the heart by circulating neurohormones has been extensively studied in a variety of insect species (see Raabe, 1982 for review). Several cardiotropic factors (defined as those substances that affect the heart) have been isolated from the cockroach corpus cardiaca (CC). The majority of these factors produce a marked cardioexcitation (Ralph, 1962; Gersch, 1974). Proctolin, when assayed on the in vitro cockroach heart, increases both the rate and amplitude of contractions, even at a concentration as low as 10⁻⁹moll⁻¹ (Miller, 1979). Several investigators have even been able to induce release of putative cardioregulatory substances via electrical stimulation of the nerves that innervate the CC (Gersch, 1974; Kater, 1968). Yet, fulfilling these criteria does not provide sufficient evidence to establish conclusively that any of these factors has a regulatory function in the normal physiology of an insect.

Several lines of evidence have been presented in this paper to support the hypothesis that the two CAPs act as cardioregulatory neurohormones during wingspreading behaviour in the newly emerged adult Manduca sexta. The results from the bioassays of whole blood (Fig. 2) and the gel filtration experiments (Fig. 3) clearly establish the presence of two cardioacceleratory factors in wing-spreading blood that co-elute with the two CAPs. There is a marked depletion in the stored level of each of the CAPs during the first 24h of adult life (Fig. 6), and this depletion is accompanied by the appearance of CAP bioactivity in the blood during the first 90 min after eclosion (Fig. 4). The high potassium release studies show that depolarization causes CAP release in a calcium-dependent manner as expected for a neurosecretory product (Fig. 7).

The rise and decline of the CAP blood titre is temporally correlated with the initiation and termination of wing-spreading behaviour and also with the changes in heart rate seen during that behaviour. The peak blood CAP concentration, occurring 5 min after emergence, precisely coincides with the onset of wing-spreading behaviour (Truman & Endo, 1974). CAP blood levels decline throughout the 90-min period of wing inflation, and are barely detectable by the time wing spreading has been completed (Fig. 4).

The release of the two CAPs can be postponed by preventing the onset of wingspreading behaviour (Fig. 5). This result suggests that the wing-spreading motor programme must be activated in order to ensure peptide release. The release of the insect hormone bursicon is also delayed by preventing wing-spreading behaviour (Truman, 1973) and, therefore, is apparently regulated by the same mechanisms that control CAP release. Besides these control mechanisms, bursicon and the CAPs also have several other features in common. All three peptides share the same neurohaemal release site (Taghert & Truman, 1982a) and are synthesized in the abdominal ganglion but by different cells (Taghert & Truman, 1982b; Tublitz, 1983).

Another criterion commonly used in establishing a hormonal role is to determine whether exogenous application of the putative hormone mimics the physiological response of the target tissue to the endogenously released substance. Injections of either CAP₁ or CAP2 (Fig. 8) at physiological concentrations produced increases in heart rate which were quantitatively similar to the changes seen in intact animals and animals behaving normally (Fig. 1).

These results indicate that the two CAPs are released into the haemolymph and are responsible for the increase in heart rate during wing spreading in newly emerged Manduca sexta.

It is clear that the CAPs perform an important cardioregulatory function in postemergent adult Manduca, yet they cannot account for all the changes made by the heart during this period. The in vivo heart recordings show that the maximal contraction rate of the heart occurs immediately upon the initiation of adult eclosion (Fig. 1). By the time blood CAP levels are at their highest, 5 min after eclosion (Fig. 4), the heartbeat frequency has already dropped substantially. Obviously, the CAPs are not responsible for this initial heart rate increase. The rapid onset of this response may also rule out other humoral factors and suggests that this response might be regulated by other means. One hypothesis is that the CNS is responsible for this fast response via direct neural innervation. Although the innervation of the lepidopteran heart is not well understood (Wasserthal & Wasserthal, 1977; Sanger & McCann, 1968a,b), the studies by Heinrich (1970, 1971) clearly show that the lepidopteran CNS plays an important role in cardioregulation. At present, however, there is no direct experimental evidence to support this hypothesis of direct neural control.

Expansion of the wings after adult emergence is accomplished by an increase in haemocoelic pressure allowing blood to be pumped into the veins of the uninflated but elastic wings. Haemolymph is pumped into the wings by the accessory pulsatile organs, peripheral structures located at the base of each wing (Wigglesworth, 1972). These peripheral pumping stations contract in synchrony with the heart, responding to changes in heartbeat frequency by correspondingly altering their own contraction rate. The contraction rate of the accessory pulsatile organs increases two-fold simultaneously with the elevation in heart rate during eclosion and wing spreading (Wasserthal, 1976; Moreau & Lavenseau, 1975). Based upon the evidence presented in this paper, one might predict that these accessory pulsatile organs also accelerate their rates of contraction in response to the high titres of the two CAPs during eclosion and wing spreading. One consequence of this prediction is that an elevation in contraction rate might facilitate wing inflation by decreasing the duration of wing-spreading behaviour. The relationship between the heart, accessory pulsatile organs and the rate of wing spreading remains to be investigated.

Successful completion of wing inflation is also facilitated by the absence of heartbeat reversals during wing-spreading behaviour. Throughout this period the heart contracts only in an anterior-going direction shunting blood into the thorax and preventing its drainage back into the abdomen. This increase in thoracic blood volume is coupled with the elevation in haemocoelic pressure to ensure that the wings are inflated.

We thank Drs Lynn M. Riddiford and A. O. D. Willows for their criticisms of an earlier draft of this manuscript. This work was supported by a NSF predoctoral fellowship and a NIH traineeship to NJT and by a NSF Grant 8020975 to JWT.