Search

We have previously identified two isoforms of Go alpha in membranes of N1E-115 neuroblastoma cells, using an antibody raised against the purified Go alpha subunit; one isoform of the Go alpha subunit (pI 5.80) is present in undifferentiated cells, whereas a more acidic isoform (pI 5.55) appears during differentiation [J. Neurochem. 54:1310-1320 (1990)]. Recently, the Go alpha gene has been shown to encode, by alternative splicing, two polypeptides, Go1 alpha and Go2 alpha, which differ only in their carboxyl-terminal part. To determine unambiguously whether the two Go alpha subunits detected in neuroblastoma cells were actually the products of different mRNAs, rabbit polyclonal antibodies were generated against synthetic peptides (amino acids 291-302) of both sequences. Specificity of the two affinity-purified antipeptide antibodies was assessed on Western blots by comparing their immunoreactivities with those of other G alpha antibodies. On a blotted mixture of purified brain guanine nucleotide-binding proteins, the anti-alpha o1 and anti-alpha o2 peptide antibodies only recognized the 39-kDa Go alpha subunit. Furthermore, the immunological recognition of brain membranes from 15-day-old mouse fetuses by antipeptide antibodies could be specifically blocked by addition of the corresponding antigen. When membrane proteins from differentiated neuroblastoma cells and mouse fetus brain were blotted after two-dimensional gel electrophoresis, the anti-alpha o1 and anti-alpha o2 peptide antibodies labeled a 39-kDa subunit focused at a pI value of 5.55 or 5.80, respectively. Study of the ontogenesis of both Go alpha subunits revealed the predominance of Go2 alpha in the frontal cortex at day 15 of gestation. Thereafter, there was a progressive decline of the Go2 alpha polypeptide to a very low level, concomitant with an increase in the Go1 alpha protein, which plateaued about 15 days after birth to a level 8 times higher than at gestational day 15. Similarly, on neuroblastoma cells, the Go2 alpha subunit was almost exclusively present in undifferentiated cells, and differentiation induced the appearance of the Go1 alpha subunit, with a reduction in the amount of Go2 alpha polypeptide. Thus, the evolution of the two Go alpha subunits during cell differentiation, unambiguously identified with specific antibodies, suggests that neuronal differentiation is responsible for the on/off switch of the expression of the Go alpha isoforms and indicates that Go1 alpha, rather than Go2 alpha, is involved in neurotransmission.

The peptide hormones bradykinin and kallidin (Lys-bradykinin), as well as their analogues [des-Arg9]-bradykinin, a selective B1 agonist, [des-Arg9,Leu8]-bradykinin, a selective B1 antagonist, and [Thi5,8,D-Phe7]-bradykinin and D-Arg0-[Hyp3,D-Phe7]-bradykinin, two selective B2 antagonists, induced rapid histamine release from purified rat peritoneal mast cells. In contrast, the N-terminal fragment bradykinin-(1-5) was inactive. These peptides also activate the GTPase activity of GTP-binding proteins (G proteins) (Go/Gi) purified from calf brain, with an order of potency identical to that observed on mast cells, [Thi5,8,D-Phe7]-bradykinin much greater than kallidin greater than bradykinin greater than D-Arg0-[Hyp3,D-Phe7]-bradykinin greater than [des-Arg9]-bradykinin greater than [des-Arg9,Leu8]-bradykinin greater than bradykinin-(1-5). This correlation suggested that G proteins are the targets of kinins in mast cells. Accordingly, the concomitant increase in inositol trisphosphates and release of histamine elicited by kinins were inhibited by pertussis toxin pretreatment of mast cells. The inhibitory effect of benzalkonium chloride showed that the G proteins involved belong to the Gi type. GTPase activity was measured in the supernatant of homogenized mast cells but not in the membranous fraction. This activity was stimulated by kinins and by the venom peptide mastoparan. The potency of peptides was similar to that observed with purified bovine G proteins. Sodium dodecyl sulfate-gel electrophoresis of mast cell supernatant revealed pertussis toxin-induced ADP-ribosylation of two proteins, in the Mr 41,000 and 40,000 range, i.e., similar to purified alpha-subunits of Gi1 and Gi2 or Gi3 subtypes. The data support the proposal that bradykinin and analogues act like mastoparan, substance P, and compound 48/80, interacting first with sialic acid residues of the cell surface and then with Gi-like proteins, inducing phospholipase C activation and intracellular calcium mobilization.

Bovine brain contains two GTP-binding proteins, Gi and Go, which are substrates for ADP ribosylation by pertussis toxin. The Gi protein mediates hormone and GTP inhibition of adenylate cyclase, but the function and the precise tissue distribution of Go are unknown. To immunologically probe the localization of Go, we have purified the Go alpha and G beta, gamma subunits of Go and have raised antibodies against them. The polyclonal anti-Go alpha antibodies obtained were very selective for Go alpha compared to Gi alpha or Gs alpha. The positive Go alpha and G beta, gamma immunoreactivities were investigated in different tissues of vertebrates and invertebrates on immunoblots after gel electrophoresis of the crude membranes. The anti-G beta, gamma antibodies recognized a 35-36-kDa protein in brain of vertebrates such as mammals (rat), avians (pigeon), amphibians (frog), fish (trout), and reptiles (turtle) but not in the invertebrates such as molluscs (snail) and insects (locust). With the anti-Go alpha antibodies a high level of immunoreactivity was detected at molecular weights of 39,000-40,000 in the brain of invertebrates as well as in the central nervous system of vertebrates. Moreover, ADP ribosylation with pertussis toxin occurred in the nervous system of invertebrates. These results suggest that the GTP-binding proteins of invertebrates either are devoid of G beta, gamma subunit or, more probably, possess immunologically different G beta, gamma subunits when compared to those of vertebrates. In the vertebrates, Go alpha immunoreactivity was also present in the peripheral nervous system in areas such as the superior cervical ganglia and sciatic nerve. When examined with the anti-Go alpha antibodies, the neuro-and adenohypophysis exhibited a similar immunoreactivity which was about 6 times lower than in brain. Our antibodies also recognized a 40-kDa protein in human adipocytes but at a concentration 17 times lower than that recognized in brain. Taken together, these data show that the Go alpha subunit is well conserved through evolution and, furthermore, confirm that Go alpha is not strictly limited to the nervous system. This suggests that the protein Go ensures a function required for neuronal activity but also present in some other non-nervous tissues.

5-Hydroxytryptamine (5-HT) stimulates the accumulation of inositol-trisphosphate in WRK1 cells, a cell line originating from a rat mammary tumor. 5-HT acts via a single receptor type for which it has an affinity constant estimated to be 1.27 microM. A series of agonists known to act at 5-HT2 receptors are partial agonists in this system and have a rank order of relative intrinsic efficacies corresponding to that seen in other systems possessing 5-HT2 receptors. There is an essentially linear occupancy-response relationship for 5-HT and other agonists indicating the absence of a strong amplification mechanism between receptor activation and inositol phosphate formation. The selective blockade of the 5-HT response by nanomolar concentrations of 5-HT2 selective antagonists but not by drugs acting at other 5-HT receptor subtypes suggest that the receptor in WRK1 cells is of the 5-HT2 type. Additionally, we demonstrate that in WRK1 membranes 5-HT acts via the 5-HT2 receptor to elicit a GTP dependent increase in the production of inositol-bisphosphate and inositol-trisphosphate. These properties of the WRK1 cell line indicate that it is a useful model with which to study the nature of 5-HT receptor coupling to the putative second messenger(s), the inositol phosphates.