Arthropod 5-HT2 receptors: a neurohormonal receptor in decapod crustaceans that displays agonist independent activity resulting from an evolutionary alteration to the DRY motif

Abstract

The stomatogastric nervous system (STNS) is a premiere model for studying modulation of motor pattern generation. Whereas the cellular and network responses to monoamines have been particularly well characterized electrophysiologically, the transduction mechanisms that link the different monoaminergic signals to specific intracellular responses are presently unknown in this system. To begin to elucidate monoaminergic signal transduction in pyloric neurons, we used a bioinformatics approach to predict the existence of 18 monoamine receptors in arthropods, 9 of which have been previously cloned in Drosophila and other insects. We then went on to use the two existing insect databases to clone and characterize the 10th putative arthropod receptor from the spiny lobster, Panulirus interruptus. This receptor is most homologous to the 5-HT2 subtype and shows a dose-dependent response to 5-HT but not to any of the other monoamines present in the STNS. Through a series of pharmacological experiments, we demonstrate that this newly described receptor, 5-HT2betaPan, couples with the traditional G(q) pathway when expressed in HEK293 cells, but not to G(s) or G(i/o). Moreover, it is constitutively active, because the highly conserved DRY motif in transmembrane region 3 has evolved into DRF. Site-directed mutagenesis that reverts the motif back to DRY abolishes this agonist-independent activity. We further demonstrate that this receptor most likely participates in the modulation of stomatogastric motor output, because it is found in neurites in the synaptic neuropil of the stomatogastric ganglion as well as in the axon terminals at identified pyloric neuromuscular junctions.

Figure 1.

Conservation of 5-HT receptors. The predicted protein sequences for the two arthropod 5-HT_2β receptors are aligned with those predicted for the human type 2C receptor, the arthropod 5-HT_2α paralog, and a lobster 5-HT type 1 receptor. Amino acids are numbered such that the start methionine is +1 in each sequence. The heavy black bars above the sequence approximate the seven transmembrane regions. Amino acids matching the consensus sequence are boxed. Amino acids discussed in the text or listed in Table 2 are highlighted. Highlighted circles represent N-linked glycosylation sites. The accession numbers are: 2Chum, NM_000868; 2αDro, NP_524223; 2βDro, NP_731257 plus NP_649805; 2βPan, AY550910; 1Pan, AY528822.

Figure 2.

The effect of biogenic amines on PKC translocation. HEK 5-HT_2βPan or nontransfected HEK cells were exposed to the indicated drug for 15 min. Cytosolic and membrane fractions were separated, and PKC-specific activity in each fraction was measured. The percentage of total PKC-specific activity associated with the cytosolic fractions (▪) versus membrane fractions (□) is indicated. Data are expressed as the mean ± SEM, and n = 3 for all experiments, except A and B, in which n = 8 for 0 5-HT. *p < 0.05, significantly different when compared with the same fraction in the absence of drug.

Figure 3.

The PLC inhibitor ET-18-OCH₃, but not PTX, blocks the 5HT_2βPan receptor-mediated translocation of PKC. After pretreatment with the indicated inhibitor, cells were incubated an additional 15 min with 5-HT (10^–3m). PKC activity was measured in the cytosolic and membrane fractions. Data are expressed as the mean ± SEM; n = 3 per condition. *p < 0.05, significantly different when compared with 0 5-HT.

Figure 4.

5-HT stimulates IP production in transfected cells expressing serotonin receptors but not in the parental HEK cell line_. Total PI and IP were assayed in nontransfected HEK (□), HEK 5-HT_2βPan (▪), and HEK 5-HT_2βPan (F171Y) (▦) cells. Three conditions are shown: control (no drug), 5-HT (60 min exposure to 10^–3 m 5-HT), and NaF (60 min exposure to the nonspecific trimeric G-protein activator 20 mm NaF). Results are expressed as a ratio of IP/(PI + IP). Data represent the mean ± SEM from three separate experiments. *Significantly different (p < 0.05) for drug versus control for a given cell line; **significantly different from HEK cells under the same condition (i.e., comparisons within each of the three conditions: control, 5-HT, NaF); ***significantly different from HEK 5-HT_2βPan (F171Y) under the same condition.

Figure 5.

cAMP levels do not change in response to 5-HT treatment. cAMP levels were measured in 5-HT_2βPan (▪) and nontransfected HEK (□) cells or cells exposed to 5-HT (10^–3 m), or forskolin (2.5 mm), or forskolin (2.5 mm) with 5-HT(10^–3 m), or forskolin (2.5 mm) with 5-HT(10^–3 m) after pretreatment with PTX. Results are expressed as picomoles of cAMP/per milligram of protein. Data are the mean ± SEM from three separate experiments. *p < 0.05, significantly different for drug versus control for a given cell line. There were no significant differences between any treatments containing forskolin within or between cell lines.

Figure 6.

The PMA response is potentiated in HEK 5-HT_2βPan cells. The PKC activator PMA (0.1 μm) was applied for 15, 30, or 60 min to each of three cell lines [HEK 5-HT_2βPan, ♦; HEK, ▪; HEK 5-HT_2βPan (F171Y), ▴]. The PKC-specific activity associated with the cytosolic and membrane fractions was measured, and the percentage of the total specific activity associated with a given fraction was determined. The percentage of PKC-specific activity associated with the membrane is plotted for each time point in each cell line. Each data point represents the mean ± SEM; n = 3 per data point.

Figure 7.

Restoration of the DRY motif disrupts the agonist-independent activity associated with the 5-HT_2βPan receptor but not the response to 5-HT. HEK 5-HT_2βPan (F171Y) cells were assayed for PKC translocation after a 15 min exposure to varying concentrations of 5-HT, as indicated. Data are expressed as the mean ± SEM from three separate experiments, except for the 0 5-HT data point, in which n = 5. *p < 0.05 when compared with activity in the same fraction in the absence of drug treatment.

Figure 8.

Anti-5HT₂βCrust specifically recognizes the 5-HT_2βPan receptor in lobster nervous tissue. A representative Western blot experiment is shown. The blot containing protein extracts from lobster nervous tissue was probed with an antibody against 5-HT_2βPan (5-HT_2βCrust) or the same antibody preabsorbed with the peptide antigen used to generate the antibody (preabsorbed). Molecular weight standards are indicated. The antibody produces a signal corresponding to the predicted size of the protein, which is lost on preabsorption.

Figure 9.

Most stomatogastric neurons express 5-HT_2βPan to varying degrees. A, B, A 5.45 μm confocal slice from different representative ganglia are shown. Scale bar (in A), 50 μm. A, Anterior left quadrant ∼20 μm from the dorsalmost aspect of the ganglion. This slice depicts mostly the peripheral layer showing eight neuronal cell bodies; however, the fine, synaptic neuropil-containing tufts of neurites from stomatogastric and modulatory input neurons can be seen in the bottom right corner, just above the scale bar. The arrow points to one of many profiles most likely representing transport of the receptor in fibers that are entering/leaving the ganglion through the peripheral layer. B, Optical slice representing posterior left quadrant of the ganglion ∼35 μm from the ventralmost aspect. All layers of the ganglion appear in this slice: the triangular course neuropil (CN) jutting out from the left, surrounded by several tufts of fine neuropil (fn), surrounded by the peripheral layer containing somata, surrounded by the perineural sheath. C, Confocal projection representing ∼1.5μm in depth, showing a single neuron double-labeled for the 5-HT_2βPan receptor (red) and the K⁺ channel shal 1.b (green). The right panel represents the merged image. Notice the ring of protein in the plasma membrane that is present in the K⁺ channel profile but absent in the receptor profile. In our hands, the nuclei always stained intensely in the double-label experiments (in both the receptor and channel profiles) but not in the single-label experiments (in both the receptor and channel profiles). We do not understand this technical artifact.

Figure 10.

Colocalization of 5-HT_2βPan receptors with synaptotagmin at identified NMJs. Double-label immunocytochemistry was performed on PD and PY muscles, using the antibodies indicated above the panels. Control represents the same double-label experiment, except that the anti-5HT₂βCrust antibody was preabsorbed. Each row displays representative NMJs from one experiment, in which identified muscles and control are as indicated. Arrows point to 5-HT_2βPan staining that is not localized to the synapse. The PD panels represent a projection of five confocal slices representing ∼5 μm in depth. Each set of PY and preabsorbed panels represent a single 1.0 μm confocal slice.