Affiliate disclosure: Some links in this article are affiliate links. We may earn a commission at no additional cost to you.

Medical disclaimer: This article is for informational purposes only. It is not medical advice, diagnosis, or treatment. Not medical advice. Consult a qualified healthcare professional before changing your diet, exercise, or supplement regimen, particularly if you take anticoagulant medications or manage an allergy or autoimmune condition.

Green shiso arrives at the table as a garnish nearly every time sashimi is served in Japan. It sits under the sliced fish — slightly waxy, faintly fragrant — and is often left uneaten by anyone who does not know what it is. That is understandable. Shiso looks decorative. In practice it has a flavor profile unlike almost anything in a Western kitchen: somewhere between mint, anise, and basil, with a brightness that cuts through fatty fish and high-salt fermented foods.

What the leaf contains at the molecular level has attracted research attention, mostly from Japanese pharmacognosy and allergy medicine communities, for the past two decades. The key compounds — rosmarinic acid, luteolin, and a suite of associated polyphenols — show meaningful anti-inflammatory and antiallergic activity in laboratory settings. Whether that activity translates to humans is a more complicated question, and one the evidence answers only partially.

Shiso in Japanese cooking: two varieties, one name

Perilla frutescens var. crispa is the botanical species. It exists in two functionally distinct Japanese varieties.

Green shiso (ao-jiso) is the common sashimi garnish. Fresh leaves appear in salads, wrapped around grilled meats or gyoza, layered into cold noodle dishes, or thinly sliced over cold tofu. Dried shiso flakes are a standard ingredient in furikake rice seasoning. This is the variety that shows up most frequently in polyphenol analysis for rosmarinic acid content.

Red shiso (aka-jiso) is deeper purple-red with a stronger, slightly more bitter flavor. It is used primarily for pickling — most famously as the coloring agent for umeboshi, the salt-pickled plum that takes on its characteristic ruby color from weeks of contact with red shiso leaves. Aka-jiso contains higher concentrations of anthocyanins that produce that color; ao-jiso is the more studied variety for rosmarinic acid.

Perilla seed oil — extracted from seeds rather than leaves — is a structurally different product. Pale yellow, mildly nutty, it has been used in Japanese cooking and traditional medicine for centuries and is sold today in Japan as a functional food primarily for its high alpha-linolenic acid content.

These are three related but distinct ingredients. The leaf polyphenol evidence and the seed oil omega-3 evidence are separate bodies of research and should not be conflated.

Rosmarinic acid, luteolin, and alpha-linolenic acid

Shiso leaves are unusually concentrated in rosmarinic acid — a hydroxycinnamic acid derivative that also occurs in rosemary, mint, and sage, but is present in shiso at concentrations that rival or exceed those in other culinary herbs. Japanese analyses of dried shiso leaves place rosmarinic acid content between roughly 1 and 7 grams per 100 grams of dried leaf, depending on variety, growing conditions, and drying method. This is several times the concentration typically found in dried rosemary.

In cell culture, rosmarinic acid has been studied extensively for inhibition of NF-κB, a transcription factor that drives production of pro-inflammatory cytokines including TNF-α, IL-1β, and IL-6. The compound also inhibits 5-lipoxygenase, which reduces leukotriene synthesis, and shows measurable COX-2 inhibitory activity in vitro. The standard caveat applies: cell culture concentrations and the conditions under which rosmarinic acid acts in a dish do not map cleanly to what circulates in human tissue after eating a leaf garnish or taking an extract capsule.

Luteolin, a flavonoid also present in shiso, operates through different mechanisms. In mast cell culture models, luteolin inhibits IgE-mediated degranulation — the process by which mast cells release histamine and other inflammatory mediators in allergic responses. This activity has been observed across multiple cell lines. Luteolin also appears to suppress Th2 cytokine production (IL-4, IL-5, IL-13) in vitro — cytokines associated with allergic inflammation in airways and skin. Translation to clinical allergy management involves pharmacokinetic and dosing hurdles that in vitro data cannot resolve.

Alpha-linolenic acid (ALA), the primary bioactive component of perilla seed oil, constitutes approximately 55 to 60 percent of its fatty acid composition. ALA is the plant-based omega-3 precursor that the body can partially convert to EPA and DHA, though this conversion is inefficient in most adults — typically 5 to 10 percent of ALA reaches EPA, and less than 1 percent reaches DHA under normal conditions. Perilla seed oil is not equivalent to fish oil or algae-derived EPA/DHA supplements in terms of long-chain omega-3 delivery. Its relevance to inflammation research relates to the modest EPA the ALA conversion supports, and to ALA’s own modest influence on eicosanoid production.

What small human trials have actually measured

The most developed area of human evidence for shiso involves seasonal allergic rhinitis. A double-blind, placebo-controlled crossover study published in Biofactors (2004, Takano et al.) tested a standardized perilla leaf extract against placebo in participants with seasonal allergic rhinitis. The extract-treated period was associated with statistically significant reductions in sneezing frequency and eye itching relative to placebo. The sample was small — fewer than 40 participants — the trial duration was short, and the study has not been replicated at larger scale. A separate line of small trials examining rosmarinic acid specifically, at doses of 200 to 400 mg per day from botanical extracts, has found consistent but modest reductions in nasal symptom scores in allergic populations during pollen season, with no comparable effect observed in non-allergic participants.

The pattern across this literature is narrow but directionally consistent: perilla-derived rosmarinic acid is associated with modest reductions in IgE-mediated allergy symptom scores in short-duration trials. Effect sizes are small, participant numbers are low, and no trial has extended follow-up long enough to assess sustained outcomes.

For perilla seed oil and skin inflammation, a Japanese dietary intervention from the early 2000s compared perilla oil supplementation to control oil in patients with atopic skin conditions. After several weeks, the perilla group showed measurably lower circulating levels of leukotriene B4 — a pro-inflammatory mediator derived from arachidonic acid. Symptom scores improved across both groups during the trial, which complicates interpretation, and no well-powered definitive RCT on perilla oil and atopic outcomes has been published to date.

Human outcome data for rosmarinic acid and systemic inflammation markers (CRP, IL-6) is thinner than the allergy literature. A small number of trials in metabolic syndrome populations have found associations between supplementation and reduced oxidative stress markers, but the evidence level is preliminary and the populations studied differ substantially from healthy adults.

Perilla seed oil: sourcing and supplement form

Perilla seed oil sold internationally is primarily marketed as a plant-based omega-3 oil, with alpha-linolenic acid content as the lead claim. Some products include shiso leaf polyphenol extract as a combined offering; others are purely seed-pressed oil.

Perilla seed oil supplements on Amazon US includes Japanese-branded and domestically formulated products in both capsule and liquid forms. When reading labels, look for the ALA percentage per serving — it should be in the 55 to 60 percent range for an authentic cold-pressed perilla oil. Products that separately quantify rosmarinic acid content are standardizing the phenolic fraction, which is the compound most relevant to the allergy and inflammation trial literature.

Culinary perilla and shiso oil on Amazon US surfaces both Japanese-imported pressed oils and supplement formulations. Culinary-grade perilla oil is used cold — in dressings, drizzled over cold tofu, used as a finishing oil — rather than for high-heat cooking, where ALA degrades.

For dried shiso as a whole food, shiso flakes and furikake blends on Amazon US provides a practical form for people outside Japan who do not have access to fresh shiso at Asian grocery stores. Dried shiso retains rosmarinic acid and luteolin, though the quantity that survives drying and storage is variable and not well standardized across commercial products.

If you have garden access, ao-jiso is straightforward to grow from seed in warm climates and is available as seedlings at specialty nurseries in the US during the growing season. The plant self-seeds readily and can produce more fresh leaf than a household needs through summer.

Who should be careful, and what the evidence actually supports

Concentrated perilla extract and high-dose perilla seed oil have shown anticoagulant-adjacent effects in animal and cell culture models. The clinical significance at supplement doses in humans is not firmly established, but if you take warfarin, daily aspirin, or any other anticoagulant or antiplatelet medication, mention perilla supplementation to your prescribing clinician before starting.

Shiso belongs to the mint family (Lamiaceae). People with documented allergies to other Lamiaceae herbs — basil, mint, lavender — have a theoretical cross-reactivity risk with shiso, though reported clinical cross-reactivity is uncommon. Someone with a history of severe herb-related reactions who is considering concentrated shiso extract is better positioned to assess that decision with an allergist.

For healthy adults without these conditions, fresh or dried shiso at culinary quantities carries no established safety concerns. The concentrated extract literature, while preliminary, suggests a reasonable experimental dose range: 200 mg rosmarinic acid per day puts a supplement in the range of the better-designed small allergy trials. This is not a confirmed clinical outcome dose — it is the range where the existing evidence was generated, nothing more.

The evidence profile here is structurally similar to what the brown seaweed compounds show at the same research stage. Wakame, Kombu, and Fucoidan: What the Anti-Inflammatory Research Actually Shows covers the fucoidan and alginate literature, where in vitro anti-inflammatory evidence is well-developed and human trials are small, consistent in direction, and well short of outcome confirmation. Shiso polyphenols are at approximately the same stage: solid preclinical foundation, early clinical signal in allergy-specific populations, and a gap before the large confirmatory trials that would establish clinical application. That pattern characterizes many Japanese functional food ingredients.

---

A practical starting point for the food form: add fresh or dried shiso to meals three to five times a week for four weeks. This is the kind of dietary exposure that puts you in the range of what Japanese dietary pattern research actually describes — regular small servings over time, not pharmacological doses. For someone specifically interested in the allergy trial literature, a standardized perilla extract providing 200 mg rosmarinic acid per day represents the lower end of tested doses. Neither approach should replace tested allergy treatments, and neither produces effects at a level the current evidence supports calling reliable across populations.

For the broader Japanese anti-inflammatory food evidence base, see Wakame, Kombu, and Fucoidan: What the Anti-Inflammatory Research Shows and The Japanese Longevity Diet: A Beginner’s Guide.