Extract of Perilla frutescens Enriched for Rosmarinic Acid, a Polyphenolic Phytochemical, Inhibits Seasonal Allergic Rhinoconjunctivitis in Humans

 

  1. Hirohisa Takano*,1,
  2. Naomi Osakabe,
  3. Chiaki Sanbongi,
  4. Rie Yanagisawa,
  5. Ken-Ichiro Inoue*,
  6. Akiko Yasuda,
  7. Midori Natsume,
  8. Seigo Baba,
  9. Ei-Ichiro Ichiishi* and
  10. Toshikazu Yoshikawa*

  1. *First Department of Medicine, Kyoto Prefectural University of Medicine, Kyoto 602-0841, Japan; Pathophysiology Research Team, National Institute for Environmental Studies, Tsukuba 305-0053, Japan; and Health and Bioscience Laboratories, Meiji Seika Kaisha, Ltd., Sakado 350-2889, Japan.
  1. 1Pathophysiology Research Team, National Institute for Environmental Studies, 16-2 Onogawa, Tsukuba 305-0053, Japan. E-mail:
    htakano{at}nies.go.jp

Abstract

Extract of Perilla frutescens enriched for rosmarinic acid, a polyphenolic phytochemical, suppresses allergic immunoglobulin responses and inflammation
caused by polymorphonuclear leukocytes (PMNL) in mice. However, few placebo-controlled clinical trials have examined the efficacy
and safety of polyphenolic phytochemicals for treatment of allergic inflammatory diseases in humans. The present study determined
whether oral supplementation with rosmarinic acid is an effective intervention for patients with seasonal allergic rhinoconjunctivitis
(SAR). In this 21-day, randomized, double-blind, age-matched, placebo-controlled parallel group study, patients with mild
SAR were treated daily with extract of Perilla frutescens enriched for rosmarinic acid (200 mg [n = 10] or 50 mg [n = 9]) or placebo (n = 10). Patients recorded symptoms daily in a diary. Profiles of infiltrating cells and concentrations of eotaxin, IL-1β,
IL-8, and histamine were measured in nasal lavage fluid. Serum IgE concentrations and routine blood tests were also examined.
As compared with placebo supplementation, supplementation with extract of Perilla frutescens enriched for rosmarinic acid resulted in a significant increase in responder rates for itchy nose, watery eyes, itchy eyes,
and total symptoms (P < 0.05). Active treatment significantly decreased the numbers of neutrophils and eosinophils in nasal lavage fluid (P < 0.05 vs. placebo). Patients reported no adverse events, and no significant abnormalities were detected in routine blood
tests. In conclusion, extract of Perilla frutescens enriched for rosmarinic acid can be an effective intervention for mild SAR at least partly through inhibition of PMNL infiltration
into the nostrils. Use of this alternative treatment for SAR might reduce treatment costs for allergic diseases.

  • rosmarinic acid
  • seasonal allergic rhinoconjunctivitis
  • polymorphonuclear leukocytes
  • alternative medicine
  • randomized, double-blind, age-matched, placebo-controlled parallel group study
  • extract of Perilla frutescens

During the latter half of the 20th century, the incidence of allergic rhinitis increased worldwide, leading to dramatic escalation
in direct and indirect treatment costs (1). In 1996, direct and indirect costs for allergic rhinitis were estimated at $7 billion and $3 billion, respectively. Despite
steady advances in conventional therapies for allergy symptoms, phytochemicals and herbal products have been widely used by
consumers as alternatives to prescription drugs without definitive clinical evidence (2). In the United States, the prevalence of alternative therapy use for allergies rose from 8.7% in 1990 to 16.6% in 1997 (3).

Among a variety of phytochemicals, polyphenolic phytochemicals have been shown experimentally to inhibit inflammatory and
vascular processes (47). Furthermore, a randomized, blinded, crossover investigation of the cocoa polyphenol, procyanidin, has shown that it can
favorably alter eicosanoid (prostacyclin) synthesis in humans, providing a possible mechanism by which it could inhibit atherogenic
disorders (8). In addition, another well-controlled clinical trial has demonstrated that a polyphenol-rich cocoa beverage suppresses ADP-
and epinephrine-stimulated platelet activation (9). In contrast, few placebo-controlled clinical trials have examined the efficacy or safety of polyphenolic phytochemicals
on allergic inflammatory diseases in humans. A randomized, double-blind, placebo-controlled study of grape seed extract has
shown no significant difference between active and placebo groups in rhinitis quality of life assessments, symptom disease
scores, or requirements for rescue antihistamine treatment (10).

Rosmarinic acid, a polyphenolic phytochemical, exists in a variety of medicinal species within the plant genus Lamiaceae,
such as basil, sage, rosemary, mint, and Perilla frutescens, a popular garnish in Japan. Oral supplementation with rosmarinic acid in Perilla decoction reportedly suppresses allergic
reaction in mice (11).

The present study was undertaken to investigate whether extract of Perilla frutescens enriched for rosmarinic acid is a beneficial intervention for patients with seasonal allergic rhinoconjunctivitis (SAR).
Furthermore, we assessed its effects on mediator release and on polymorphonuclear leukocyte (PMNL; neutrophils and eosinophils)
infiltration into nasal lavage fluid.

Materials and Methods

Subjects.

We enrolled 30 patients aged 21 to 53 years with mild SAR to Japanese cedar (Cryptomeria japonica) pollen. All had a known medical history of allergic rhinoconjunctivitis only during the pollen season for at least the previous
2 years and had a positive serum test for Japanese cedar pollen specific for IgE of Class 2 or greater. Patients were excluded
if they had received any drugs or had an active respiratory infection within the previous 3 weeks or during the study period.
Exclusion criteria also included a history of drug allergies, presence of other medical conditions, an established history
of wheezing or asthma, and use of immunotherapy or nasal surgery. The enrolled patients lived within 10 km of the same company
and commuted there daily except on weekends. Participants were fully informed regarding the experimental procedures, and written
consent was obtained. Safety was monitored by clinical history, physical examinations, and routine blood tests, including
hepatic and renal function tests and complete blood counts.

Study Design.

The study used a randomized, double-blind, age-matched, placebo-controlled parallel group design. The protocol was approved
by the ethics committees of the Kyoto Prefectural University of Medicine. The study was carried out in accordance with the
Declaration of Helsinki. The trial was carried out from late February to late March. During the study period, the mean concentration
of Japanese cedar pollen was 2.48 million grains/m3 (SD = 2.41 million; Fig. 1).

View larger version:


Figure 1.

The pollen counts during the study period. At Days 0, 3, and 21 (arrows), daily symptoms, tablet intake, and adverse events
were collected and confirmed, and nasal washes were performed. All patients were tested on the same day in the morning at
the same location.

Ten patients each were randomized to the rosmarinic acid (200 mg ) group, the rosmarinic acid (50) mg group, and to the placebo
group. At Day 0, patients were instructed to record their daily symptoms and medication on a diary card. For 21 days, each
patient took tablets daily after breakfast. Tablets were indistinguishable between groups and contained rosmarinic acid from
Perilla frutescens extracts at a daily dose of 200, 50, or 0 mg. At Days 0, 3, and 21, daily symptoms, tablet intake, and adverse events were
collected and confirmed, and nasal washes were performed. All patients were tested on the same day in the morning at the same
location in the company. Cellular profiles and concentrations of eotaxin, IL-1β, IL-8, and histamine were measured in nasal
lavage fluid. Routine laboratory examinations and specific IgE concentrations were measured in blood samples. At Day 21, patient
evaluations of global symptoms were assessed using a self-report health-related quality of life questionnaire.

Tablets.

All test tablets were prepared by Meiji Seika Kaisha, Ltd. In brief, fresh Perilla leaves were extracted with 1.0% w/v citric
acid at 90°C for 30 mins. The extract was filtered and freeze-dried. The concentration of polyphenolic substances including
rosmarinic acid in this extract was determined by high performance liquid chromatography (HPLC) using a Develosil HG-5 column
(Nomura Chemical Co., Ltd., Aichi, Japan) with solvents A (0.1% v/v trifluoroacetic acid [TFA] in distilled water) and B (0.1%
v/v TFA in acetonitrile) under the following conditions: 10%–50% linear gradient of A in B; flow rate =0.8 ml/min; and detection
= 280 nm (Fig. 2). The concentration of rosmarinic acid was 20% w/w. The concentrations of luteolin, luteolin-7-O-glucoside, protocatechuic
acid, and caffeic acid were below the detection limit. The test tablets were made with this Perilla extract, lactose, and
cellulose powder. The placebo tablets were made in the same manner as the test tablets but without Perilla extract. Flavor
and pigment were added to the placebo tablets in order to match those added to the test tablets.

View larger version:


Figure 2.

HPLC profile of the tablet containing the extract of Perilla frutescens. The concentration of polyphenolic substances including rosmarinic acid was determined by HPLC using a Develosil HG-5 column
with solvents A (0.1% v/v trifluoroacetic acid [TFA] in distilled water) and B (0.1% v/v TFA in acetonitrile) under the following
conditions: 10%–50% linear gradient of A in B; flow rate, 0.8 ml/min; detection, 280 nm.

Diary Card and Patient Evaluations of Global Symptoms.

We used a self-report health-related quality of life questionnaire and a diary card on the basis of the previous reports (12, 13) with modification. Patients recorded their symptoms daily using a score from 0 to 3, where 0 = symptom not present; 1 =
symptom present, no discomfort; 2 = some discomfort; and 3 = marked discomfort. This scoring system was used to assess sneezing,
rhinorrhea, stuffy nose, itchy nose, watery eyes, itchy eyes, and total symptoms.

A self-report health-related quality of life questionnaire was used to measure patients’ assessment of global symptoms. In
the questionnaire, patients rated their relief from symptoms on a 3-point scale: no relief or worse (1), partial (2), or complete (3) relief from global symptoms.

Nasal Lavage.

Nasal lavages were performed with isotonic sterile saline preheated to 37°C. Specifically, 3 ml of saline was instilled in
each nostril. After 15 secs, the lavage fluid was collected. The average volume retrieved was 95% of the instillation volume
(6.0 ml); the amounts did not differ by treatment. The lavage fluid was centrifuged at 300 g for 10 mins, and the total cell count was determined on a fresh fluid specimen. Differential cell counts were assessed on
cytologic preparations stained with Diff-Quik (International Reagent, Kobe, Japan). A total of 500 cells per preparation were
counted under oil immersion microscopy (14). The supernatants were stored at −80°C for the measurements of proinflammatory mediators.

Measurement of Proinflammatory Mediators.

Eotaxin (Biosource, Camarillo, CA); IL-1β(Genzyme, Minneapolis, MN); IL-8 (Genzyme); and histamine (IBL, Hamburg, Germany)
were measured in nasal lavage fluid supernatants using commercially available ELISA kits according to the manufacturer’s instructions.
Specific IgE concentrations were measured using Pharmacia CAP RAST (Uppsala, Sweden) according to the manufacturer’s recommended
protocol.

Routine Blood Tests.

We estimated complete blood cell counts, hepatic and renal function, and total protein; performed proteinogram analysis; and
measured concentrations of electrolytes, lipids, uric acid, and creatine phosphokinase in blood samples.

Statistical Methods.

Data were analyzed using SPSS for Windows 7.5.1. (SPSS Japan, Inc., Tokyo, Japan). The responder rate was defined as an improvement
of at least two points in the total symptom score and one point in each symptom score between baseline (Day 0) and Days 3
or 21. The chi-square test was used to evaluate differences in improvement between groups. An analysis of covariance (ANCOVA)
was used for all other efficacy variables. The number of cells in total nasal lavage fluid was analyzed after log transformation.

Results

The participants in the three groups were well-matched in age, gender, and duration of SAR (Table 1). One patient in the rosmarinic acid (50 mg) group was excluded from the study because of an acute respiratory infection
during the study period. Twenty-nine patients were finally enrolled in the study. The rosmarinic acid (200 mg) group consisted
of 10 patients (six men and four women, mean age 33.1 ± 6.3 years); the rosmarinic acid (50 mg) group of nine patients (five
men and four women, mean age 32.2 ± 6.6 years); and the placebo group of 10 patients (five men and five women, mean age 33.0
± 9.3 years). There were no significant differences in baseline total symptom scores between the groups: the rosmarinic acid
(200 mg) group, 2.3 ± 3.9; the rosmarinic acid (50 mg) group, 3.8 ± 3.2; and the placebo group, 3.9 ± 7.0. There were also
no significant differences in the baseline symptom scores for sneezing, rhinorrhea, stuffy nose, itchy nose, watery eyes,
and itchy eyes among the three groups. No rescue medication was required for any patient, and no patient reported any adverse
event.

View this table:


Table 1.

Baseline Characteristics of Patientsa

Changes in the symptom scores during the study period were not significantly different among the groups. Patients’ evaluations
of global symptoms using the health-related quality of life questionnaire, however, showed that only 30% of placebo-supplemented
participants improved after treatment (Fig. 3). In contrast, 55.6% and 70% of the patients reported global symptom relief in the rosmarinic acid (50 and 200 mg) groups,
respectively (P = 0.05; placebo vs. rosmarinic acid at a daily dose of 200 mg).

View larger version:


Figure 3.

Patients’ evaluations of global symptoms using the health-related quality of life questionnaire. Patients’ assessments of
global symptoms at study endpoint relative to prestudy levels are shown for each group. The chi-square test was used to analyze
the difference in variation of quality of life between groups.

Compared with placebo supplementation, daily oral supplementation with the extract containing rosmarinic acid at its daily
dose of 50 or 200 mg resulted in significant increases in responder rates that were based on symptom score reductions for
itchy nose, watery eyes, and itchy eyes on Day 21 (P < 0.05; Table 2). The 50-mg dose significantly increased the responder rate on the total symptom score on Day 3 (P < 0.05 vs. placebo), and tended to increase it on Day 21. The responder rates for the total symptom score were also greater
in the rosmarinic acid (200 mg) group than in the placebo group on Days 3 and 21; however, these increases did not achieve
statistical significance.

View this table:


Table 2.

Responder Rates on Symptom Scoresa

On Day 3, the numbers of PMNL and neutrophils in nasal lavage fluid were significantly decreased by the extract at both doses
of rosmarinic acid (P < 0.05 vs. placebo; Table 3). The numbers of eosinophils in nasal lavage fluid were significantly decreased by the daily dose of 200 mg of rosmarinic
acid on Day 3 (P < 0.05 vs. placebo) and showed a nonsignificant decrease in the rosmarinic acid (50 mg) group. On Day 21, the numbers of
these cells were smaller in the rosmarinic acid groups than in the placebo group, but these changes were not statistically
significant.

View this table:


Table 3.

Numbers of Inflammatory Cells in Nasal Lavage Fluida

The levels of eotaxin, IL-1β, IL-8, and histamine in nasal lavage fluid supernatants were not significantly different among
the three treatment groups (Table 4). There were also no significant differences in serum pollen-specific IgE concentrations among the groups (Table 4).

View this table:


Table 4.

Inflammatory Mediators in Nasal Lavage Fluid or Seruma

On Day 21, no significant abnormalities were detected by routine blood tests including complete blood cell counts, hepatic
and renal function tests, total protein and proteinograms, electrolytes, lipids, uric acid, and concentrations of creatine
phosphokinase (Table 5).

View this table:


Table 5.

Routine Laboratory Testsa

Discussion

The present randomized, double-blind, age-matched, placebo-controlled parallel group study provides clinical evidence that
extract of Perilla frutescens enriched for rosmarinic acid is an effective intervention for mild SAR. The effects may be mediated, at least in part, through
the inhibition of PMNL infiltration into the nostrils.

In previous experimental studies, polyphenolic phytochemicals including rosmarinic acid have been shown to inhibit IgE responses
(11, 15) and inflammation characterized by PMNL infiltration (5, 6, 16, 17). Furthermore, Makino et al. (11) have reported that Perilla decoction suppresses the PCA reaction in mice. Since the degree of inhibition by Perilla decoction
and rosmarinic acid were nearly equal, they postulated that the anti-allergic effect of Perilla decoction is primarily due
to rosmarinic acid (11). In another report, polyphenol strictinin isolated from tea leaves has inhibited IgE production in B cells (15). Green tea polyphenols have also been shown to attenuate bowel inflammation caused by autoimmunity as demonstrated by lower
histologic scores and wet colon weights (5). Finally, it has been reported that virgin olive oil with a higher content of polyphenolic compounds protects experimental
animals from carrageenan-induced inflammation and adjuvant arthritis (6). However, there have been no clinical reports that oral supplementation with polyphenolic phytochemicals is an effective
intervention for SAR. To our knowledge, the current study is the first report demonstrating the efficacy of phytochemicals
for patients with SAR.

Although a previous randomized, double-blind, placebo-controlled clinical trial suggests that a Chinese herbal formulation
is partially effective in treatment of perennial allergic rhinitis (18), the mode of action has not been elucidated. In our investigation, amelioration of total symptoms on Day 3 showed consistent
correlation with decreased numbers of PMNL in nasal lavage fluid. Since allergic rhinitis is inflammation characterized by
PMNL such as eosinophils and neutrophils, the beneficial effects of extract of Perilla frutescens enriched for rosmarinic acid on SAR may depend, at least in part, on the inhibition of PMNL infiltration into nostrils.

It is well recognized that PMNL are activated by proinflammatory cytokines and chemokines including IL-1β, IL-8, and eotaxin.
IL-8 and eotaxin are chemotactic for neutrophils and eosinophils, respectively. In fact, our recent experimental study has
shown that extract of Perilla frutescens enriched for rosmarinic acid inhibits murine airway inflammation induced by diesel exhaust particles possibly through inhibition
of the enhanced local expression of proinflammatory cytokines and chemokines such as IL-1βand IL-8 (17). In the present clinical study, however, the levels of the proinflammatory cytokines and chemokines were not proportional
to the numbers of PMNL in the nasal lavage fluid. In general, allergic inflammation consists of multiple steps, including
immediate and late allergic responses where a variety of proinflammatory cytokines and chemokines as well as chemical mediators
are locally expressed at different times after allergen exposure and play important roles in the pathogenesis. Thus, the timing
and the dose of allergen exposure can affect the local expression of these proinflammatory mediators. In the present study,
however, the enrolled patients’ daily exposure (timing and the dose of allergen) was not strictly matched. Future time-course
studies using controlled provocation with known allergens at an identical dose could further elucidate the role of these mediators
in the beneficial molecular actions of rosmarinic acid. Similarly, future measurement of these mediators in eye wash fluid
or tears might provide further information about how rosmarinic acid acts, since this extract predominantly inhibited eye,
rather than nose, symptoms. Furthermore, the role of the other mediators such as complements, platelet activating factor,
and leukotrienes, which are reportedly chemotactic for PMNL, should be examined.

In our study, extract of Perilla frutescens enriched for rosmarinic acid did not affect pollen-specific IgE concentrations in the serum, which is not consistent with
the findings of a previous animal study (11). The discrepancy might be explained in part by differences in species, allergen, or period of rosmarinic acid administration.
Our patients received the extract after past sensitization with pollen, whereas the experimental animals received it before
sensitization with ovalbumin. Our data suggest that the beneficial action of this extract in SAR patients is mediated by inhibition
of PMNL-dependent inflammation, rather than by an effect on pollen-specific IgE production. Concentration of the pollen-specific
IgE present in the nose or eye, which can directly induce the release of proinflammatory mediators in the local site, might
be used to elucidate the mechanism of rosmarinic acid action.

In the present study, routine laboratory tests including complete blood cell counts, hepatic and renal function tests, and
creatine phosphokinase showed safety of rosmarinic acid at a daily dose of 200 mg for 21 days.

In conclusion, extract of Perilla frutescens enriched for rosmarinic acid can be an effective intervention for patients with mild SAR that may act through inhibition
of PMNL-dependent inflammation. The present preliminary study provides clinical evidence that this alternative treatment for
SAR might reduce treatment costs for allergic diseases. A more detailed study involving a larger number of patients treated
for a longer period of time will be needed to draw more definite conclusions.

Acknowledgments

We thank T. Tsuda at Josai University for providing information on pollen concentrations.

Footnotes

  • This work was supported by a grant from Meiji Seika Kaisha, Ltd.

  • Received September 8, 2003.
  • Accepted December 8, 2003.

References

  1. CrossRefMedline
  2. CrossRefMedline
  3. CrossRefMedline
  4. Abstract/FREE Full Text
  5. CrossRefMedline
  6. Abstract/FREE Full Text
  7. Abstract/FREE Full Text
  8. Abstract/FREE Full Text
  9. Medline
  10. CrossRefMedline
  11. Abstract/FREE Full Text
  12. CrossRefMedline
  13. Abstract/FREE Full Text
  14. CrossRefMedline
  15. CrossRefMedline
  16. CrossRefMedline
  17. Medline