Question 1

What are the side effects of LL-37?

Accepted Answer

None of the studies behind this page administered LL-37 to a person and then recorded what happened, so there is no list of human side effects to report. What the record contains instead is laboratory and animal evidence that the peptide is far from biologically inert. A 2024 study stated the problem directly: LL-37 is active against microorganisms, but it is also toxic to the body's own cells. A 2017 review of defence peptides across vertebrate animals reached a similar balance, cautioning that these molecules may also contribute to excessive inflammation and tumour development. Those two themes organise the rest of this page: direct damage to cells, and inflammation that goes further than it should. Each finding below identifies the model it came from, because a result in a culture dish or a mouse is a warning signal to monitor rather than a measured human side effect.

Question 2

What are the side effects of LL-37 treatment?

Accepted Answer

Treatment, in the sense of a deliberately administered dose, appears only in animal experiments among these sources, and the harm information from them is limited. The most informative example used the mouse version of the peptide, called CRAMP, which comes from a similar gene but is a separate molecule. Given to mice immediately before circulation returned to a blocked heart artery, CRAMP made the resulting cardiac damage worse. A 2003 rat study injected a fragment of hCAP18, the precursor protein LL-37 is cut from, into a vein, and its abstract reports the effect on eye inflammation without any mention of adverse effects. A 2011 mouse study applied synthetic LL-37 to wounds and reported only measures of tissue repair. So the record contains one harmful result in a cardiac model and two experiments that were not designed to look for harm. That is not the same as a reassuring safety finding. The amounts used in each experiment are set out on the LL-37 dosage page.

Question 3

Can LL-37 damage human cells?

Accepted Answer

Yes, in laboratory culture, and at a concentration the body itself can reach. A 2024 study stimulated a line of human immune cells with vitamin D and measured how much hCAP18 and LL-37 they subsequently produced. The quantity corresponded to roughly one micromolar, a unit describing how concentrated the peptide was in the surrounding fluid. The researchers then exposed human osteoblast-like cells, which behave like bone-forming cells, to synthetic LL-37 at that same concentration, and fewer of them survived. The immune cells were less sensitive than the bone cells, so the toxicity was not uniform across cell types. The authors interpreted this as potentially relevant to bone damage driven by the body's own LL-37, which is a different question from a dose taken deliberately. It still matters for anyone who assumes a naturally occurring peptide cannot injure the surrounding tissue. Cell survival in a flask is a crude measurement, however, and it does not tell you what a living body would show.

Question 4

Can LL-37 make skin inflammation worse?

Accepted Answer

In psoriasis, the research identifies LL-37 as part of the problem rather than the solution. A 2020 review describes cathelicidin and other defence peptides being released in psoriatic skin and activating the innate immune system in ways that generate inflammation, which is how they participate in causing the disease. One mechanism involves the body's own genetic material. The review explains that these peptides help fragments of a person's own DNA and RNA attach to immune receptors that would ordinarily ignore them. Another mechanism involves neutrophils, the white blood cells that arrive first at any injury. They release sticky extracellular webs of DNA and protein, LL-37 included, and in psoriatic skin those webs steer T cells toward a type that sustains the disease. All of this concerns the LL-37 a body produces in its own skin. It is not an experiment on applying or injecting more, but it is the reason a skin claim on the LL-37 benefits page cannot simply be accepted.

Question 5

Could LL-37 trigger rosacea-like redness?

Accepted Answer

Rosacea is the clearest case where LL-37 is used to create disease deliberately. A 2016 review lists elevated LL-37 in the skin of rosacea patients among the immune alterations of the condition, alongside increased vitamin D3 in the outer epidermal layer. Together, the review suggests, these alterations may make the skin more sensitive to external and internal triggers. Laboratories have converted that association into an experimental tool. A 2020 study employed a mouse model in which rosacea-like skin disease is induced with LL-37, and in those animals sleep deprivation made the rosacea-like features worse. That is a mouse experiment, and its abstract does not specify the amount or the route of administration. But a peptide that researchers rely on to produce a rosacea-like rash in mice is not one to assume is gentle on human facial skin. Whether a cosmetic or injected preparation does the same in a person has not been asked in the studies covered here, which is a gap rather than a reassurance.

Question 6

Can LL-37 set off allergy-like reactions?

Accepted Answer

Mast cells are the immune cells that release histamine during an allergic reaction, and LL-37 can activate them. A 2004 review lists among its actions that it causes mast cells to discharge their granules, which is the step that releases histamine and related inflammatory chemicals. A 2020 study describes one pathway for this. Mast cells carry a receptor called MRGPRX2, and the cathelicidin LL-37 is one of the molecules that activate it. The same receptor responds to certain approved medicines, and it has been implicated in allergy-like reactions to them in humans, as well as in the chronic inflammation of asthma, hives and rosacea. This type of reaction is described as pseudo-allergic because it does not require the antibodies of a true allergy. The study was investigating a plant compound that could block the receptor, rather than evaluating LL-37 as a product. It identifies a plausible route to flushing or hives, not a measured frequency of either.

Question 7

Does LL-37 affect the heart?

Accepted Answer

The one cardiac study among these sources points toward harm, and it is a mouse study. Researchers tied off a coronary artery in mice for 45 minutes and then restored blood flow, which imitates a heart attack followed by reopened circulation. Mice genetically engineered without their own cathelicidin gene developed less cardiac inflammation, a smaller area of dead muscle and lower blood levels of a heart damage marker. Administering the mouse peptide before the injury had the opposite effect, as described in the earlier section on deliberate dosing. The same team then exposed human white blood cells in culture to the human peptide, and LL-37 prompted those cells to process and release IL-1 beta, a signalling molecule that drives inflammation. Their conclusion was that cathelicidin from white blood cells plays a damaging role in the heart during the early period after this kind of injury. That is a single mouse model of one event. It does not show that LL-37 harms a healthy heart, but it does indicate that the peptide is not neutral in cardiac tissue.

Question 8

Has LL-37 been tied to severe skin reactions to medicines?

Accepted Answer

It has, in the most dangerous medication reactions the skin can experience. Stevens-Johnson syndrome and toxic epidermal necrolysis are rare, life-threatening reactions to medications in which large areas of the outer skin separate and detach. A 2021 study investigated how they begin. Neutrophils in early lesions released their DNA webs along with LL-37, and that LL-37 caused keratinocytes, the main cells of the outer skin, to produce a receptor called FPR1. Keratinocytes carrying the receptor became vulnerable to a programmed form of cell death, and the dying cells then released additional LL-37 of their own, a cycle that probably amplified the damage. The authors did not observe this cycle in milder drug eruptions, so it appears specific to the severe forms. The LL-37 in this sequence is the body's own, released inside the lesions. It does not show that LL-37 from outside the body causes these reactions, but it places the peptide at the centre of one of the most serious skin injuries in medicine.

Question 9

Is there a cancer question with LL-37?

Accepted Answer

There is an unresolved one, and it cuts both ways. A 2017 review of defence peptides across vertebrate animals concluded that this family may both promote and inhibit cancer growth. On one side, several of these peptides destroy cancer cells in laboratory experiments. On the other, the review notes that human cathelicidin is overexpressed in several types of cancer, and that its level correlates with tumour growth. Overexpressed means the tumour tissue produces more of it than normal tissue would. A correlation of that kind does not establish that LL-37 causes cancer, since a growing tumour could increase the peptide rather than the reverse. It does mean the question has a genuine basis in the scientific literature rather than being an alarmist invention. None of the studies behind this page followed people given LL-37 over any period, so whether additional peptide changes cancer risk in either direction remains an open question.

Question 10

What does LL-37 interact with?

Accepted Answer

The documented interactions fall into three kinds, and none of them involves a person taking LL-37 alongside something else. The first kind changes how much LL-37 the body makes. In human immune cells grown in a dish, vitamin D at a normal body level raised the activity of the gene behind LL-37 about 170-fold. Going the other way, healthy donors given five days of G-CSF, a medicine that boosts white cell production, ended up with white cells that completely lacked hCAP18, the protein LL-37 is cut from. The second kind is with antibiotics, in glassware. Against slimy bacterial films grown in the lab, pairing LL-37 with the antibiotic ciprofloxacin did better than either one alone in most of the pairs tested. The third kind blunts what LL-37 does: a plant compound called osthole damped the mast cell response that LL-37 and similar molecules set off. Drug interactions in the clinical sense, with blood concentrations and recorded adverse events in patients, are absent from this record.

Question 11

Does boosting the body's own LL-37 carry risks?

Accepted Answer

Several everyday inputs are said to raise how much LL-37 the body makes, and that route has its own cautions. A 2014 overview lists vitamin D, zinc, sunlight's ultraviolet B and a number of other factors as able to switch on the body's defence peptides. The rosacea work above found raised LL-37 alongside extra vitamin D3 in the outer skin, so more is not always better in skin that is already inflamed. Timing matters as well. In a 2020 mouse model of blood infection that spreads from the gut, mice given cholecalciferol, the everyday form of vitamin D3, before the infection died less often over the next seven days, as did mice given the active form of the vitamin. Mice given cholecalciferol after the infection had begun fared worse, with more deaths and harsher symptoms. That is one mouse model, and it says nothing about people directly, but it shows that pushing on this system is not a one-way bet.

Question 12

Who took part in the human studies, and who was left out?

Accepted Answer

Every participant in the human studies covered here was measured rather than given LL-37. The people who appear are adults with early kidney disease taking vitamin D for a year, healthy volunteers consuming a sugar substitute, healthy blood-cell donors receiving G-CSF, and patients providing gum fluid, saliva or skin samples. In the kidney trial, 46 adults with stage 2 or 3 disease received vitamin D or a placebo, and blood cathelicidin was measured at the start and again at 12 weeks. In the dietary trial, 19 healthy volunteers consumed 30 grams of isomalt or ordinary sugar daily for two four-week periods, and LL-37 activity in the lining of the rectum did not change. Excluded entirely is anyone given LL-37 itself, which means there is no information on pregnancy, older age, childhood or people taking other medications. Which groups ought to avoid it cannot be determined from this record, since none of its participants received the peptide.

Question 13

How long does the LL-37 safety record run?

Accepted Answer

For any animal actually receiving LL-37 or a closely related peptide, the record extends to hours. The rat eye study measured its results 24 hours after the injection. The mouse heart study examined the damage 3 and 24 hours after blood flow returned. The mouse infection experiment followed animals for seven days, although what those mice received was vitamin D rather than the peptide. In people, the longest measurement of the body's own LL-37 among these sources comes from the kidney trial, where blood cathelicidin was checked at 12 weeks within a year of vitamin D. The largest registry study tracking the precursor is VITAL Infection, which examines vitamin D and omega-3 supplements alongside infections and hCAP18, with 25,874 people enrolled. It is listed as active but not recruiting, and its results are not among these sources. What prolonged use of LL-37 as a product does to a person is, in short, unmeasured in this record.

Question 14

Why is the LL-37 harm record this thin?

Accepted Answer

Part of the explanation is historical. A 2000 review described these peptides as killing germs without harming the host, and for a molecule the body already makes, safety was easy to take for granted. Later research complicated that picture, as the sections above show. Another part of the answer is where development went next. A 2018 study set out to design hybrid peptides precisely to get past the harmful effects of the natural ones. A 2025 paper adds that natural antimicrobial peptides are degraded quickly and often harm cells without distinguishing bacteria from human cells. A 2017 review of the prospects for clinical use looks at LL-37 together with the versions built from it. So much of the effort moved on to modified molecules, and the original peptide was left with cellular and animal signals. A thin record is not evidence that LL-37 is harmless. It reflects the absence of the one experiment that would settle the question either way.

LL-37 side effects and safety data