Question 1

What amounts of LL-37 did the published studies use?

Accepted Answer

Among the studies behind this page, the amounts given on purpose fall into three groups, and none of them involves a person. In a living animal, the one stated amount of a human cathelicidin peptide comes from a 2003 rat study, which injected 1, 10 or 100 micrograms of an hCAP18 fragment into a vein. In animal tissue, a 2020 mouse mastitis study used synthetic LL-37 and synthetic mouse cathelicidin at a concentration of 8 micromolar. In cells and bacteria grown in a dish, the figures spread widely: about one micromolar of synthetic LL-37 reduced the survival of human bone-forming cells, while eradicating a Pseudomonas biofilm required the peptides at 640 milligrams per litre or more. A 2011 mouse wound study applied LL-37 to the skin without its abstract stating an amount. Micrograms per animal, micromolar and milligrams per litre are different units measuring different things. They cannot be lined up on a single scale, let alone a human one.

Question 2

Who received those amounts?

Accepted Answer

Laboratory animals and isolated tissue, each selected to answer one narrow question. The rat dose comparison used male Lewis rats whose eye inflammation had been triggered by a bacterial toxin. The wound work used mice given dexamethasone, a steroid that deliberately slows healing. The heart study gave the mouse peptide, CRAMP, to adult male mice of a standard laboratory strain immediately before blood flow returned to a blocked artery. Two studies of peptides engineered from LL-37 went further: one applied an ointment to samples of human skin kept outside the body, and one used goats with a spinal disc infection. None of these subjects was a healthy adult human, and none received LL-37 by a route a person could reproduce at home. Each model was chosen to make one particular effect easy to observe, which is a reason to read the amounts as properties of the experiment rather than of the peptide.

Question 3

How was LL-37 administered in the studies?

Accepted Answer

Four different approaches appear, and only one of them is an injection. The rat study delivered the precursor fragment into a vein, straight after the toxin had been injected into a footpad. The 2011 mouse study applied synthetic and laboratory-grown LL-37 to the surface of wounds. A 2005 study took a different path entirely, using a modified virus to carry the gene for active human cathelicidin into cultured skin cells, which made those cells better at killing bacteria. And SAAP-148, a peptide derived from LL-37, was mixed into a hypromellose ointment, a gel base used for skin preparations. The human trials that tracked cathelicidin gave vitamin D by mouth, not LL-37 by any route. So none of these routes has been investigated in a person for the peptide itself, and the routes that exist were chosen to suit an animal model rather than a patient.

Question 4

How often was it given, and for how long?

Accepted Answer

Once, in nearly every case: each rat received a single injection, and its eyes were examined 24 hours later. The heart study gave the mouse peptide once, just before blood flow returned, and assessed the damage after 3 or 24 hours. SAAP-148, the derived peptide, was applied in a single treatment lasting four hours. None of these sources describes LL-37 given repeatedly, on a schedule, over days or weeks. The only extended schedules belong to vitamin D, and in a 2013 kidney disease trial adults took 50,000 IU of cholecalciferol weekly for 12 weeks, then every other week for a further 40 weeks. That is a vitamin D schedule, designed to correct vitamin D status, and measuring cathelicidin along the way does not convert it into an LL-37 regimen. A pattern of single exposures measured within a day says nothing about what repeated use would do.

Question 5

What was the lowest amount tested, and did it do anything?

Accepted Answer

The rat study is the only one among these sources that compared several stated amounts of a human cathelicidin peptide in a living animal, and its result has a distinctive pattern. At 1 microgram, the hCAP18 fragment did not suppress the eye inflammation compared with rats given the toxin alone. At 10 and 100 micrograms it held back the rise in inflammatory cells and several inflammatory signals. The higher amount added nothing further, because the effect of 10 micrograms was as strong as that of 100. In that one rat model, then, there was a floor below which nothing happened and a ceiling above which more did not help. It is tempting to derive a starting amount from that pattern, and it would be a mistake. The animals were rats, the peptide was a fragment rather than complete LL-37, and the inflammation was provoked experimentally by a bacterial toxin in a way that no human illness reproduces exactly.

Question 6

Are the body's own LL-37 levels a dose?

Accepted Answer

No, and confusing the two is the most common error in reading this literature. The body produces LL-37 continuously, and a 2011 paper notes it is found at different concentrations in many different cells, tissues and body fluids. Studies that measure those levels, such as registry entries on LL-37 around dental implants or in smokers given vitamin D3, record how much a person already carries. That is a reading, not an amount anyone received, and the difference between the two can be considerable. A 2012 review of gum disease found that most defence peptides in mouth fluids sit below the concentrations that work in laboratory tests, and suggested that local levels must be higher where it counts or that the peptides do other jobs there. A natural level is not a harmless ceiling either: the roughly one micromolar produced by vitamin D-stimulated immune cells was enough to damage bone-forming cells, as the LL-37 safety page explains.

Question 7

Does a vitamin D amount count as an LL-37 dose?

Accepted Answer

Not in any measurable or practical sense. Vitamin D switches on the gene for the LL-37 precursor, and a 2024 cell study put numbers on the protein side: 50 nanomolar vitamin D raised hCAP18 and LL-37 protein three- to fourfold in cultured human immune cells. But a vitamin D amount does not tell you how much LL-37 an individual eventually produces. In the 2013 kidney trial, high-dose vitamin D lowered one blood marker of inflammation, MCP-1, by 12 weeks. The authors report that the other markers of inflammation and immunity did not change, and blood cathelicidin was among the things measured. A skin study used calcipotriol, a vitamin D-like compound, to restore hCAP18 in skin cells from patients with a blistering disease. In each case the amount administered is an amount of vitamin D or a related compound, selected for its own purpose, and translating it into an LL-37 dose would require a conversion these studies did not attempt.

Question 8

Do mouse amounts apply to human LL-37?

Accepted Answer

Not directly, because a mouse study often gives the mouse molecule. Humans and mice each have only one cathelicidin, a 2014 review notes, but they are different peptides: LL-37 in people and CRAMP in mice. A 2001 study describes them as coming from similar genes, with similar shapes, similar ranges of activity and a similar distribution across tissues. Similar, however, is not identical. When a paper says cathelicidin was given to mice, it usually means CRAMP, as in the 2020 heart study, where the mouse peptide worsened the injury. A few studies did evaluate both peptides side by side. The 2020 mastitis work used each at 8 micromolar, and both limited bacterial invasion of mouse breast tissue. Even then, the figure describes a concentration around cells in a mouse rather than a quantity for a person, and the species label has to travel with every number.

Question 9

Do amounts for peptides made from LL-37 apply to LL-37?

Accepted Answer

No, because several research groups have used LL-37 as a starting point and then modified it, and their amounts belong to the new molecule. SAAP-148 was built from LL-37 to have stronger antimicrobial activity than the parent, so the same amount of each would not behave alike. FFRKSKEK is an eight-unit segment of LL-37 that only switches on under ultrasound, and its antibacterial effect was measured after 15 minutes of ultrasound exposure. Ceragenins are synthetic mimics of these peptides rather than fragments, and a 2017 study found them toxic to human skin cells above 10 micrograms per millilitre while lower amounts encouraged cell movement. Every modification in sequence or structure alters how much is required and what toxicity appears. A figure from one of these studies is not evidence about the amount of LL-37 that would behave the same way.

Question 10

Why is a study amount not a protocol?

Accepted Answer

Because each number was selected to answer one question in one experimental setting, and the setting determines much of the outcome. A 2015 review of that year's peptide research stated it plainly: environmental conditions clearly influence how these peptides act. For LL-37, it describes clumping that depends on acidity, which changes what the peptide carries and where it releases it. A culture dish has a fixed volume, identified bacteria and nothing attempting to degrade the peptide. A body has enzymes, salts, proteins and shifting acidity, and a 2025 paper notes that antimicrobial peptides are often slow to act and can be rapidly degraded. A study amount also arrives tied to its endpoint: an eye cell count at 24 hours, or a biofilm count after a set time. A protocol would require amounts tested in people, repeated over time, with outcomes and adverse effects systematically recorded. The studies behind this page supply none of those steps, which is why their figures appear on this page as parameters rather than guidance.

Question 11

What did the human trials that tracked LL-37 actually give?

Accepted Answer

Something other than LL-37, on every occasion. The 2006 diet trial gave 19 healthy volunteers 30 grams of isomalt or sugar a day across two four-week periods, with rectal biopsies taken at the end of each phase. Healthy donors given G-CSF for five days, a medicine that drives white cell production, ended up with neutrophils lacking hCAP18 altogether. On the registry, the largest entry linked to the precursor is VITAL Infection, which gives vitamin D and omega-3 fatty acids and is listed as active but not recruiting. A completed study of periodontal care examines LL-37 as a biomarker while patients receive gum disease treatment. In every one of these, the administered dose belongs to a sugar substitute, a growth factor, a vitamin or a dental procedure. LL-37 is the thing measured afterwards, and that measurement is not an amount anyone took.

Question 12

Does the timing of a dose change the result?

Accepted Answer

In the animal work, timing changed the outcome as much as the amount did. In a 2020 mouse sepsis study, cholecalciferol given before the infection was followed by fewer deaths over seven days, while the same compound given after the infection had started led to more deaths and worse symptoms. The active form of vitamin D, administered afterwards, was beneficial. That study used vitamin D3 as an inducer of the mouse's own cathelicidin rather than giving the peptide, but it shows how much the clock matters in this system. The peptide experiments were timed just as tightly: the rat fragment went in immediately after the toxin, and the mouse heart peptide immediately before blood flow returned. Neither result indicates what would happen an hour earlier or a day later. A reader looking for when to take LL-37 will not find it in these timings, which describe the moment each experiment was built around.

Study	Setting	Amount studied
2003 rat study (hCAP18 fragment)	Living rat, into a vein	1, 10 or 100 micrograms
2020 mouse mastitis study	Animal tissue	8 micromolar
Human bone-forming cells	Dish	About one micromolar reduced survival
Pseudomonas biofilm	Dish	640 milligrams per litre or more to eradicate
2011 mouse wound study	Skin	Not stated in the abstract

LL-37 protocols in the published literature