ANP Peptide and Vasodilation: Signalling Notes on Smooth Muscle Relaxation

The receptor is the reason anp causes vasodilation as rapidly as it does. NPR-A, also called guanylyl cyclase A, is a single pass membrane protein whose extracellular domain binds the peptide and whose intracellular domain synthesises cGMP straight from GTP. There is no intervening G protein, so the latency between binding and second messenger production is short. The intracellular portion carries a kinase homology domain that seems to set the gain of the response, and a catalytic domain that performs the chemistry. The receptors sit as dimers and the kinase homology region is phosphorylated in the resting state.

cGMP then does its work through at least three routes. Protein kinase G is the one most authors emphasise, and it is the best characterised. Cyclic nucleotide gated channels are a second route, mainly in specialised cells. The cyclic nucleotide phosphodiesterases are the third, and in my reading they are badly underrated, because they set both the amplitude and the duration of the signal. A vessel rich in phosphodiesterase 5 will hydrolyse cGMP quickly and show a smaller relaxation for the same receptor occupancy than a vessel in which that enzyme is sparse.

Sustained exposure is where the account gets less tidy. Prolonged elevation of the peptide produces a blunted response in several preparations, and the explanations offered include receptor dephosphorylation, receptor internalisation, and up regulation of the phosphodiesterases. I have not found a single account that fits every vessel, which suggests the mechanism of desensitisation is itself tissue specific. The same receptor population comes up from a different angle in my anp peptide notes, where the question is how much peptide reaches the tubule at all.

If anp causes vasodilation, the final common step has to be a fall in the phosphorylation of myosin light chain, and there are two ways to get there. Protein kinase G lowers intracellular calcium by several routes at once: it inhibits the inositol trisphosphate receptor associated substrate so less calcium leaves the sarcoplasmic reticulum, it opens large conductance potassium channels and hyperpolarises the membrane, it inhibits calcium entry through voltage gated channels, and it accelerates both sequestration and extrusion. That combination is why anp causes vasodilation even when a vessel is preconstricted by a receptor dependent agonist.

The second route is calcium sensitivity, and I think it is the more interesting one. Relaxation can occur without any measurable fall in calcium, provided myosin light chain phosphatase becomes more active. Protein kinase G phosphorylates the myosin targeting subunit of that phosphatase and also phosphorylates RhoA, which disables the Rho kinase pathway that would otherwise hold the phosphatase in check through CPI-17. The smooth muscle cell becomes less responsive to the calcium that is still there, which is a different statement from saying calcium fell, and it is part of the reason anp causes vasodilation without any measurable fall in calcium in some preparations.

This is also where the peptide meets the vasoconstrictor systems. Protein kinase G antagonises signalling downstream of angiotensin II, endothelin and adrenergic receptors at several points, and atrial natriuretic peptide has been reported to oppose the growth promoting actions of those same agonists in cultured vascular smooth muscle. I do not read those reports as a statement about remodelling in any organism; I read them as evidence that the cGMP pathway intersects the protein kinase C pathway rather than running alongside it. See atrial natriuretic peptide.

Endothelial cells carry natriuretic peptide receptors too, and the question of what they contribute has produced a messy literature. In some beds, removing the endothelium shifts the concentration response curve substantially, which is usually read as evidence for a nitric oxide intermediate. In others the shift is small or absent, and the relaxation in a denuded ring looks almost identical to the intact preparation. The honest summary I have arrived at is that the endothelial contribution differs by vessel and by the constrictor tone imposed by the experimenter.

There are other endothelial mechanisms besides nitric oxide. The clearance receptor has been reported to couple to inhibitory G proteins in endothelial cells and to influence endothelial nitric oxide synthase activity, which is a strange second life for a protein described as a disposal route. Endothelial cells also release C type natriuretic peptide, which acts on a different guanylyl cyclase receptor on the smooth muscle side. Reading about that pushed me back to the wider anp peptide notes more than once.

One further endothelial claim deserves mention because it is easy to overlook. Atrial natriuretic peptide has been reported to reduce endothelial permeability and to oppose the increase in permeability produced by inflammatory mediators, an effect attributed to protein kinase G and to small GTPase regulation of the cortical actin cytoskeleton. That is a barrier property rather than a tone property, but it belongs in the same picture, because a hormone that changes both vessel calibre and transvascular fluid movement will change plasma volume for two separate reasons.

Anp causes vasodilation in conduit arteries, resistance arteries and veins, but the sensitivity is not uniform, and the venous side is the one I underestimated. Veins are often reported as being at least as sensitive as arteries, and venodilation matters because it shifts blood into the capacitance circulation and lowers cardiac preload. When an in vivo study reports a fall in blood pressure after peptide administration, part of that fall may be a filling pressure effect rather than a fall in resistance, and the two are rarely separated cleanly.

The regional pattern is where anp causes vasodilation least uniformly. The pulmonary circulation relaxes and hypoxic pulmonary vasoconstriction is blunted; large coronary conductance vessels dilate; cerebral vessels show a modest response; renal and splanchnic beds show preglomerular and preportal selectivity respectively; skeletal muscle beds have been studied with forearm plethysmography and venous occlusion techniques with variable findings. The pattern that emerges is of a hormone that relaxes widely but unevenly, with the unevenness tracking receptor density, phosphodiesterase complement and the prevailing constrictor tone.

My practical rule when reading this literature is to ask three questions first: which vessel, which preparation, and what was the prevailing tone. Atrial natriuretic peptide studied in a relaxed vessel tells you almost nothing about what it does in a preconstricted one, and a conduit artery tells you little about a resistance vessel. I have been keeping a running list of these comparisons, with the papers that changed my mind, in my research journal.