Cytokine

Cytokines are mostly 6 to 30 kDa proteins covering the interleukins, interferons, the tumour necrosis factor superfamily, chemokines and colony-stimulating factors. They bind receptors with subnanomolar affinity and act at picomolar concentrations, largely through JAK-STAT or NF-kB signalling, over short distances and with half-lives measured in minutes. Two properties define the network: pleiotropy, where one cytokine has different effects on different cell types, and redundancy, where several cytokines produce the same effect.

Cytokine blockade has produced some of the most successful drugs ever made: etanercept and infliximab against tumour necrosis factor in 1998, adalimumab in 2002, tocilizumab against the interleukin-6 receptor, and anakinra against interleukin-1. The failures are equally instructive: the same anti-TNF agents that transformed rheumatoid arthritis did not improve survival in sepsis trials, worsen outcomes in heart failure, and reactivate latent tuberculosis.

Redundancy makes a cytokine's importance disease-specific, so a node that is rate-limiting in one condition is bypassed in another and elevation in a tissue is no evidence that blocking it will help. Kinetics constrain measurement the same way: with half-lives of minutes and mainly local action, a single plasma value is a leaked snapshot of a fast process rather than a stable characteristic of a person.

Lowering inflammatory cytokines is the most overworked phrase in peptide marketing. It usually refers to supernatant from a cell line or an endotoxin-challenged rodent, measured hours later at compound concentrations far above anything a human plasma level could reach. Multiplex panels compound the problem, since absolute values for one sample can differ several-fold between kits, so a cytokine number quoted without its method cannot be compared with anything.