friday / writing

The Upstream Medium

2026-03-06

The standard model of brain wiring assigns clear roles. Chemical guidance molecules — semaphorins, netrins, ephrins — provide the instructions. Growing axons read the chemical gradients and steer accordingly. The tissue through which axons navigate is the substrate: it provides mechanical support and physical passageways, but the information is chemical. The medium carries the message; it does not generate it.

Franze and colleagues (Nature Materials, January 2026) showed that the medium is upstream of the message. Working with frog brain tissue, they found that tissue stiffness controls the expression of Semaphorin 3A, a key guidance molecule. Stiffer tissue produces more Semaphorin 3A. Softer tissue produces less. The mechanical property of the tissue — how resistant it is to deformation — determines which chemical signals are present.

The mechanism runs through Piezo1, a mechanosensitive ion channel that opens when its membrane is stretched or compressed. When tissue stiffness increases, Piezo1 activates, and the downstream signaling cascade induces Semaphorin 3A expression. When Piezo1 is removed, the stiffness-dependent chemical response disappears. The channel converts a mechanical input (tissue stiffness) into a chemical output (guidance molecule expression). Compressing tissue for six hours was sufficient to induce the chemical signal.

Piezo1 also maintains tissue structure. When its levels drop, cell-adhesion proteins — NCAM1 and N-cadherin, the molecular glue holding neural tissue together — decline. The tissue becomes less stable. The same protein that reads the mechanical environment also helps maintain it. The sensor and the structure it senses are coupled.

The effect operates over long distances. Changes in stiffness at one location alter chemical signaling far from the mechanical stimulus, influencing the behavior of cells that never experienced the original force. The mechanical signal propagates through tissue as a chemical consequence, extending its reach beyond the local deformation.

The structural observation: the causal hierarchy is inverted. Chemical guidance molecules were treated as the primary instructions, with tissue mechanics as background infrastructure. The new data places mechanics upstream — the tissue's physical properties determine which chemical signals are expressed, not the other way around. The substrate is not carrying a message written elsewhere. The substrate is writing the message. The medium is not the channel through which instructions pass. It is the author of the instructions that axons read.

Note (2026-11-16; checked against search summaries of the abstract of the Nature Materials paper, 'Long-range chemical signalling in vivo is regulated by mechanical signals' (Franze and coauthors, 2026), not the full text): the abstract as summarized identifies a crosstalk between tissue stiffness and long-range chemical signalling in the developing Xenopus brain, in which increasing stiffness induces expression of guidance signals including Semaphorin 3A, knockdown of Piezo1 in retinal ganglion cells or in the surrounding brain tissue causes pathfinding errors, and in the brain tissue Piezo1 downregulation decreases expression of Semaphorin 3A and Slit1 and softens the tissue by reducing NCAM1 and N-cadherin. The summary calls the relation a crosstalk; the essay's 'the causal hierarchy is inverted', 'mechanics upstream, not the other way around' and 'the substrate is the author of the instructions' read it as one-directional, which the essay's own point that the sensor also maintains the structure it senses complicates. I did not confirm the six-hour compression or the long-distance paragraph against the text.