A switch, not a supply

The instinct with most things you inject is arithmetic: a level drops, you top it back up, and more-often is more-restored. That model fits a nutrient or a hormone you are replacing. GHK-Cu does not work that way. It is a copper-carrying tripeptide — Gly-His-Lys with a Cu²⁺ ion — and what it does when it reaches a fibroblast is not feed the cell but instruct it: flip a transcriptional program on. The dose is a switch thrown, not a reservoir refilled.

That single distinction reframes the whole question of how often to use it. A supply you keep topped up. A program you let run — and the program GHK-Cu starts is one your body already owns and already knows how to finish.

The sequence GHK-Cu launches is the wound-repair program, and its defining feature is that it is phased — it builds, then remodels, then stops. First activation: fibroblasts switch into repair mode. Then synthesis: collagen, elastin, and the glycosaminoglycan ground substance are laid down. Then the phase that matters most for this argument — remodeling, run by the matrix metalloproteinases, the enzymes that cut, reorganize, and let newly deposited matrix cross-link and mature into something with the right architecture. Then resolution: enzyme levels fall back, activation quiets, the tissue settles.

GHK-Cu sits directly on the remodeling machinery. It stimulates MMP-2 expression in dermal fibroblast cultures (Siméon et al., Life Sciences 2000), and it modulates MMP expression and activation in wounds (Siméon et al., J. Invest. Dermatol. 1999). Those two findings are the well-cited core of the mechanism, and they are what make the phasing consequential: the drug does not just start construction, it turns up the crew that tears down and rebuilds. A crew you never send home is not the same as a finished building.

Dose GHK-Cu every day, indefinitely, and the switch is never released. The tissue is held in a standing build-and-remodel state: MMP turnover that keeps churning, matrix that is perpetually being reorganized rather than allowed to settle, cross-link, and mature. Low-grade activation becomes the resting condition instead of a passing phase. That is not more repair — it is repair that never reaches its own conclusion.

A break is what lets the last phase happen. During the off-weeks the program runs to resolution: enzymes fall back to baseline, new matrix finishes maturing, the tissue returns to homeostasis. When the next cycle starts, the fresh signal lands on rested tissue and can drive the full sequence again — which is the mechanistic reason cyclers tend to report a consistent effect cycle over cycle, where continuous users more often describe the effect plateauing or thinning out. The win isn’t dosing more; it’s letting each program finish before starting the next.

A schedule you will see repeated across the research-peptide literature is roughly six to eight weeks on, two to four weeks off, then resume. It is worth being precise about where that comes from, because the site’s creed — bullish on the science, sceptical on the page — applies most exactly here. Those week counts are not the output of a randomized head-to-head of intermittent versus continuous dosing in humans. That trial does not exist. The numbers are the mechanism above plus observed practice — a reasonable range, not a settled law. Treat the logic of cycling as well-grounded and the specific numbers as a convention.

• The MMP mechanism. Well cited — GHK-Cu up-regulating and modulating matrix metalloproteinases in fibroblasts and wounds rests on named primary literature you can check above.
• The phasing. Well established — wound healing runs build → remodel → resolve as a matter of basic tissue biology, independent of GHK-Cu.
• The exact weeks-on / weeks-off. A convention, not a finding — no controlled human trial has compared schedules. The range is inference from mechanism plus practice.

One more distinction the mechanism forces. A cosmetic topical at surface concentrations tolerates near-continuous use far better: the exposure is low and it decays across the depth of the dermis, so the signal reaching any given fibroblast is a fraction of the applied dose. An injectable — or a high-concentration topical — delivers the full signaling dose straight into tissue. The stronger and more direct the signal, the more the program is at risk of being pinned open, and the more the break earns its place. The break matters in proportion to how loudly you are throwing the switch.

None of this is a dosing prescription — it is the reasoning under one. If you want to see how a cycle segments a course in practice, the cycle planner renders on-and-off weeks against a supply estimate; and the GHK-Cu monograph holds the reference detail on structure, copper delivery, and where the skin evidence is strong versus where the systemic claims outrun it. The practical takeaway is the one the mechanism keeps pointing at: cycle it, and let each remodeling program run to completion.