In Vitro Anti-inflammatory Effects of Larch Turpentine, Turpentine Oil, Eucalyptus Oil, and Their Mixture as Contained in a Marketed Ointment

Abstract

An ointment containing larch turpentine, turpentine oil, and eucalyptus oil has been used for almost a century for the symptomatic treatment of mild, localized, purulent inflammations of the skin. Its clinical efficacy in the treatment of skin infections has been shown in clinical trials, but the mode of action of the active ingredients on inflammation is not known. We studied the anti-inflammatory properties of the active ingredients of the ointment and their mixture in a human monocyte cell model, in which the cells were stimulated with lipopolysaccharide and incubated with the test substances. The cytotoxic threshold of each test substance and the mixture was identified using the alamarBlue assay, and their anti-inflammatory activity was assessed by measuring the release of interleukins IL-1 β , IL-6, IL-8, monocyte chemoattractant protein-1, prostaglandin E ₂ , and TNF- α . Cell toxicity was observed at a mixture concentration of 10 µg/mL. All immunological assays were carried out at nontoxic concentrations. Larch turpentine decreased IL-1 β , monocyte chemoattractant protein-1, and prostaglandin E ₂ release at a concentration of 3.9 µg/mL and TNF- α at concentrations > 1.95 µg/mL, whereas eucalyptus oil and turpentine oil had no relevant inhibitory effects. The mixture dose-dependently inhibited IL-1 β , IL-6, monocyte chemoattractant protein-1, prostaglandin E ₂ , and TNF- α release at concentrations > 1 µg/mL. IL-8 release was only marginally affected. The anti-inflammatory activity of the herbal ingredients and their mixture was confirmed in this model. This effect seems to be mediated mainly by larch turpentine, with turpentine oil and eucalyptus oil exerting an additive or possibly synergistic function.

Introduction

Skin and soft tissue infections (SSTIs) such as abscesses, carbuncles, furuncles, folliculitis, impetigo, erysipelas, and cellulitis are often caused by Staphylococcus aureus and other Staphylococcus species, streptococci, Pseudomonas aeruginosa , and some yeasts ¹ ,  ² . Skin infections usually have self-limiting conditions, but in rare cases, they evolve into severe and sometimes life-threatening diseases ³ . Symptoms include pain, itching, swelling ² , and almost always inflammation ⁴ . Antibiotic or antiseptic topical treatments are commonly used for symptomatic relief and to avoid disease progression ⁵ . Polyhexanide, povidone iodine, octenidine, and chlorhexidine are recommended options, whereas topical antibiotics (e.g., fusidic acid, mupirocin, clindamycin) are rarely used ⁵ . Natural remedies are also available for topical use, but only scant information exists on their efficacy and mode of action.

Neutrophil recruitment and abscess formation are critical to fight bacterial skin infections ⁶ ,  ⁷ . Neutrophil recruitment is mediated by several factors, including proinflammatory cytokines such as the interleukins (ILs) IL-1 α , IL-1 β , IL-6, and TNF- α as well as chemokines, e.g., IL-8 ³ . The importance of proinflammatory cytokines has been confirmed for IL-1 β , IL-6, monocyte chemoattractant protein-1 (MCP-1) ( α / β ), and TNF- α in an S. aureus infection mouse model ⁸ ,  ⁹ . IL-1 β plays a key role in bacterial skin infections by mobilizing neutrophils from the bone marrow ³ , with the resulting abscess limiting infection and ultimately clearing bacteria from the tissue ⁹ . In a murine skin infection model Streptococcus pyogenes , cytolysins induced cyclooxygenase 2 ¹⁰ , and in the same model, the high levels of the proinflammatory prostaglandin E ₂ (PGE ₂ ) produced led to increased vascular permeability, dysregulation of the inflammatory response, and a decreased phagocytic bacterial killing ¹⁰ .

Induction of proinflammatory cytokines, chemokines, and prostaglandins is crucial to fight skin infections, but the overexpression of proinflammatory substances and the release of large amounts of proinflammatory cytokines such as interleukins and TNF- α are also hallmarks of often painful inflammation ⁴ . Local cytokine secretion and activation of immunocompetent cells in the infected tissue are followed by a systemic reaction characterized by an inflammatory and immune response, as well as cellular stress reactions ¹¹ ,  ¹² ,  ¹³ , carcinogenesis, cell survival, and apoptosis ¹⁴ ,  ¹⁵ .

Pull ointments containing ammonium bitumino-sulfonate or turpentine derivatives have been traditionally used to treat mild skin infections ¹⁶ . Ilon classic ointment (Cesra Arzneimittel GmbH & Co.K), for instance, has been used for more than 90 years (see Supporting Information, Section 1) for the symptomatic treatment of mild, localized, purulent inflammations of the skin, nodules, hair follicles, and sweat glands. It contains a lipophilic mixture of turpentine oil, larch turpentine, and eucalyptus oil as active pharmaceutical ingredients ( Table 1 ). Larch turpentine mainly consists of resin acids (approx. 50 to 65%; main component: larixyl acetate) and essential oils (approx. 15%; main component α -pinene) ¹⁷ . Turpentine oil consists of monoterpenes such as α -pinene and β -pinene (together approx. 95%), as well as a modest share of monoterpenes such as limonene and 3-carene ¹⁸ . Eucalyptus oil contains mainly 1,8-cineole (≥ 80%), with low concentrations of α -pinene, p -cymene, and other monoterpenes ¹⁹ .

Table 1  Components of the lipophilic ilon classic ointment used in the experiments.

	Concentration (v/v)
*Concentrations are companyʼs confidential information
Active pharmaceutical ingredients
Turpentine oil (Terebinthinae aetheroleum) from Pinus pinaster Aiton (Pinaceae), Ph. Eur.	7.2%
Larch turpentine (Terebinthina laricina) from Larix decidua Mill. (Pinaceae), Ph. Helv.	5.4%
Eucalyptus oil (Eucalypti aetheroleum) from Eucalyptus globulus Labill. (Myrtaceae), Ph. Eur.	1.2%
Ointment basis
Vaseline	72%
Other basis components*: beeswax, chlorophyll paste, stearin, oleic acid, polysorbate 20
Excipients *
Rosmarini aetheroleum from Rosmarinus officinalis L. (Lamiaceae)
Thymi typo thymolo aetheroleum from Thymus vulgaris L. (Lamiaceae)
Thymol

The clinical efficacy of ilon classic ointment in the treatment of bacterial skin infections has been shown in double-blind and open, randomized clinical trials ²⁰ ,  ²¹ , but the mode of action of the active pharmaceutical ingredients on inflammation is not well understood. Antiseptic and skin perfusion-enhancing properties have been postulated for larch turpentine ²² ; turpentine oil has been ascribed disinfectant, analgesic, antiparasitic ²³ , and antimicrobial activities ²⁴ ,  ²⁵ and the same properties have been described for eucalyptus oil ²⁴ ,  ²⁶ .

Lipopolysaccharides (LPS) from gram-negative bacteria represent a potent immune-activating stimulus, to which human monocytes respond by producing several inflammatory cytokines ²⁷ ,  ²⁸ ,  ²⁹ . Monocytes have already been used to study essential oils in in vitro inflammation models e.g. 30 . We thus decided to investigate the effects of larch turpentine, eucalyptus oil, and pine turpentine oil, the active pharmaceutical ingredients (APIs) of the ointment, as well as of their mixture on LPS-induced cytokine release (IL-1 β , IL-6, IL-8, MCP-1, PGE ₂ , and TNF- α ) in human monocytes by means of an established in vitro model of inflammation widely used for testing anti-inflammatory compounds and drugs ¹³ .

Results

The human monocyte viability determination needed to establish the maximum allowable nontoxic concentration to be used in the immunology assays showed that a concentration of up to 17 µg/mL of eucalyptus oil, 3.9 µg/mL of larch turpentine, 10.4 µg/mL of turpentine oil, and 5 µg/mL of the mixture had no effect on the viability of human monocytes ( Fig. 1 ). As expected, NaF at a concentration of 100 µg/mL reduced cell viability, on average, by 25%, at 250 µg/mL by 50%, and at 500 µg/mL by 75% (data not shown). Cell toxicity (percentage of viable cells < 80% of control) was observed at a mixture concentration of 10 µg/mL, which was thus the highest concentration of the mixture tested in the cytokine inhibition assay. The highest concentrations of the individual APIs were chosen according to their share in the mixture present in the marketed ointment (3.9, 5.2, and 0.9 µg/mL, for larch turpentine, turpentine oil, and eucalyptus oil, respectively). Cytotoxicity of the individual APIs ( Fig. 1 ) was observed well above the concentration used in the immunomodulation experiment.

Fig. 1.

 Effects of eucalyptus oil, larch turpentine, turpentine oil, and their mixture on cell viability (isolated monocytes). Results are the average of the measurements of four wells and are expressed as % of the cell viability of the respective controls. Capped bars: standard deviation; A. U.: arbitrary units. Cytotoxicity has been defined as a reduction of viable cells by 20% (i.e., number of viable cells are less than 80% of the control).

Larch turpentine decreased IL-1 β , MCP-1, and PGE ₂ release at the concentration of 3.9 µg/mL and TNF- α already at the concentration of 1.95 µg/mL but had no effect on IL-6 and IL-8. Turpentine oil and eucalyptus oil, at the tested concentrations, had no relevant effects on cytokine release, i.e., no inhibition to values lower than 80% of control was observed ( Table 2 ). The mixture dose-dependently inhibited all cytokines investigated at concentrations ≥ 1 µg/mL, except IL-8, on which only a concentration of 10 µg/mL was effective ( Fig. 2 ).

Table 2  Percentage of cytokine release (rounded to the nearest unit) by the different treatments [mean (standard deviation), results of six wells, two wells for each buffy coat obtained from three donors].

	IL-1 β	IL-6	IL-8	MCP-1	PGE 2	TNF- α
Bold: biologically significant activity (< 80% of control) detected. IL: interleukin, MCP: monocyte chemoattractant protein, PGE 2 : prostaglandin E2. Statistically significant differences from control: *p < 0.05; **p < 0.001 (ANOVA with Bonferroni correction)
Eucalyptus oil
0.9 µg/mL	92 (10.1)	100 (9.9)	93 (7.7)	101 (7.4)	99 (3.8)	100 (7.3)
0.45 µg/mL	96 (6.6)	101 (9.0)	99 (7.1)	101 (2.6)	94.6 (4.3)	99 (5.2)
0.09 µg/mL	95.4 (3.9)	103 (4.8)	100 (5.2)	102 (6.2)	100 (4.2)	100 (3.8)
0.045 µg/mL	96 (5.9)	99 (8.7)	93 (3.6)	98 (5.1)	98 (5.2)	102 (5.8)
0.009 µg/mL	96 (3.5)	102 (9.4)	97 (5.0)	99 (6.3)	97 (7.9)	101 (4.3)
Larch turpentine
3.9 µg/mL	59 (5.7) **	86 (4.6)**	98 (10.8)	72 (8.2) **	76 (5.9) **	41 (5.8) **
1.95 µg/mL	91 (7.5)*	88 (2.3)**	96 (8.3)	100 (5.3)	102 (7.3)	61 (6.2) **
0.39 µg/mL	96 (5.0)	97 (3.2)	97 (6.6)	102 (6.1)	99 (2.3)	96 (9.9)
0.195 µg/mL	101 (3.0)	99 (1.4)	100 (5.4)	98 (2.9)	101 (3.6)	98 (5.1)
0.039 µg/mL	100 (2.3)	101 (4.5)	102 (4.1)	100 (3.5)	101 (4.8)	97 (5.4)
Turpentine oil
5.2 µg/mL	99 (8.3)	91 (6.3)	94 (7.6)	83 (4.9)**	94 (5.3)	98 (8.0)
2.6 µg/mL	101 (5.3)	95 (7.4)	99 (4.0)	91 (3.9)	93 (5.1)	97 (5.3)
0.52 µg/mL	94 (4.8)	100 (5.8)	101 (8.7)	96 (7.0)	97 (3.5)	98 (4.9)
0.26 µg/mL	101 (3.6)	98 (2.8)	100 (3.8)	95 (9.4)	100 (6.8)	97 (3.4)
0.052 µg/mL	100 (2.5)	97 (6.9)	100 (9.5)	99 (4.4)	99 (6.5)	99 (4.0)

Fig. 2.

 Percentage of cytokine release (mean and standard deviation of measurements of six wells, two for each individual buffy coat obtained from three donors) after addition of the mixture of the three APIs to the cells. Results are expressed as % of the LPS-induced release in untreated cells. Capped bars: standard deviation. Markers for the first concentration showing statistically significant differences from control: **p = 0.01; ***p < 0.001 (ANOVA with Bonferroni correction).

Discussion

This study shows that in the experimental model used, a combination of larch turpentine, pine turpentine oil, and eucalyptus oil exhibits anti-inflammatory activity by decreasing the release of selected cytokines, chemokines, and PGE ₂ . Our results thus support the use of this combination, as contained in ilon classic ointment, for the symptomatic treatment of skin infections.

Monocytes play an important role in inflammation in general, but also in inflammatory skin disorders ³¹ . Macrophages at a wound site are mainly recruited from the blood and are crucial for the successful repair after tissue injury by helping resolve the inflammatory response ³² . In our model, eucalyptus oil and turpentine oil did not downregulate cytokine release. Larch turpentine, however, downregulated IL-1 β , IL-6, MCP-1, PGE ₂ , and TNF- α release at concentrations > 1.95 µg/mL, and the mixture was effective in decreasing the release of these cytokines already at concentrations > 1 µg/mL. The highest concentration (10 µg/mL) of the mixture inhibited the release of IL-8 ( Fig. 2 ), but a partial cytotoxic effect of the mixture at this concentration cannot be excluded.

The results suggest that the anti-inflammatory effect of the mixture is mediated predominantly by larch turpentine, with turpentine and eucalyptus oils providing an additional activity enhancement.

Previous studies reported anti-inflammatory activities for the same or similar active ingredients. For instance, an extract from eucalyptus leaves tested at concentrations of 60 to 480 µg/mL showed activity on IL-6 and TNF- α release in LPS-induced granulocytes ³³ , and eucalyptus oil at a concentration of 1800 µg/mL decreased IL-1 β , IL-6, and IL-8 mRNA expression in TNF- α -induced T24 human bladder epithelial cells ³⁴ . The maximum concentration of eucalyptus oil tested in our study, however, was only 0.9 µg/mL. We cannot exclude that higher concentrations could inhibit inflammatory mediators also in our model.

Pine turpentine oil had no activity, even at its maximum concentration tested. In other studies, however, a Pinus koraiensis Siebold & Zucc. extract similar to the pine turpentine oil used in this study exhibited anti-inflammatory activity towards IL-4 and IL-13 in a rat basophilic leukemia cell line ³⁵ and the Pinus pinaster extract contained in Pycnogenol showed anti-inflammatory effects towards IL-1 α , IL-6, and IFN- β in RAW 264.7 cells and towards IL-1 β , IL-6, and TNF- α in BV2 microglia cells ³⁶ . All studies carried out so far, however, used higher concentrations of pine turpentine oil. As the target of this study was to assess the immunological activity of the APIs and their mixture at the concentrations present in the ointment, we did not test higher doses of pine turpentine oil, thus we cannot exclude that higher amounts may yield positive results.

To our knowledge, no data on the anti-inflammatory potential of larch turpentine have so far been published. Larix decidua terpene-rich sawdust extracts showed anti-inflammatory effects towards cyclooxygenase (COX) COX-1, COX-2, and leukotriene B4 (LTB4) at a concentration of 20 µg/mL ³⁷ . A recent review concluded that different extracts of L. decidua may exhibit antimicrobial, antioxidant, and anti-inflammatory effects that can be beneficial in treating ulcerating wounds ³⁸ .

A strength of this study is the use of primary cells, which are more sensitive and maintain the physiological functions present in vivo much better than cultured cell lines ³⁹ .

Potential limitations include having carried out the cytotoxicity test only on basal (unstimulated) monocytes, not having tested the marketed formulation of the ointment, the use of monocytes from blood and not from the skin, and not having studied higher concentrations of turpentine oil and eucalyptus oil.

Viability testing is mainly carried out on basal monocyte cells and not on cells induced with LPS. We recently compared the effects of some compounds on basal-induced monocytes as compared to LPS-induced monocytes (data not shown) and we observed no reduced viability on LPS-induced monocytes compared to basal monocytes. We cannot exclude a negative influence of LPS in some models, but we did not observe any effects on viability in the human monocyte model used in this study.

We also did not test the pull ointment because of its lipophilic properties, particularly the high Vaseline content (72%), which makes in vitro testing almost impossible in an aqueous environment. The main thrust of this study, however, was to investigate the effects of the active pharmaceutical ingredients of the ointment and their potential additive/synergistic effects. The effectiveness of the ointment has been shown in clinical trials ²⁰ ,  ²¹ , directly confirming that the activity observed in this study translates into clinical efficacy, possibly enhanced by some of the excipients we did not test.

We also tested human monocytes originating from the blood and not from the skin. Future studies should develop models using dermal cells or skin models, as neutrophiles are known to play an important role in skin diseases ⁴⁰ .

Finally, the concentrations of the eucalyptus and turpentine oils were probably too low to allow for detecting any relevant activity when tested alone.

Overall, the results support the good anti-inflammatory activity of ilon classic ointment observed in clinical settings. Our results also suggest an additive, or possibly synergistic, effect of the active components, a feature already described for many herbal medicinal products, which are often a complex blend of different compounds that interact in an agonistic, synergistic, complementary, antagonistic, or toxic way ⁴¹ ,  ⁴² ,  ⁴³ .

Material and Methods

A detailed list of all chemicals and equipment used for the experiments is included in Table 1S , Supporting Information.

Materials

The APIs larch turpentine, pine turpentine oil, and eucalyptus oil are commercially available. Larch turpentine from L. decidua and pine turpentine oil from P. pinaster were provided by Gebrüder Unterweger GmbH. Eucalyptus oil from Eucalyptus globulus was purchased from Düllberg Konzentra GmbH & Co. KG. All API specifications comply with Ph. Eur. or Ph. Helv. ( Table 1 ).

Monocyte cell cultures

Human primary monocytes were prepared according to a previously published protocol ¹³ , using an endotoxin-free culture method, from buffy coats from three healthy human blood donors aged between 18 and 68 years, to account for potential interindividual variability among donors. Before blood donation, the donors provided written informed consent for use of their blood for research purposes at the blood donation center of the University Hospital of Freiburg, Germany. Twenty-five mL of lymphocyte separation medium (Amprotec) were loaded with 25 mL of buffy coats in 50 mL tubes. A gradient was established by centrifugation at 1800 rpm, 20 °C for 40 min, with slow acceleration and deceleration. Peripheral blood mononuclear cells in the interphase were carefully removed and resuspended in 50 mL prewarmed PBS (PAN-Biotech), followed by centrifugation. The supernatant was discarded, the pellet washed in 50 mL PBS, and centrifuged again for 10 min at 1600 rpm and 20 °C. The pellet resulting from a gradient centrifugation was resuspended in Roswell Park Memorial Institute (RPMI)-1640 low-endotoxin medium supplemented with 10% human serum (Hexcell). The cells were incubated at 37 °C under 5% CO ₂ . After 1 day, the medium and the nonadherent cells (mostly lymphocytes) were removed and fresh RPMI-1640 medium containing 1% human serum was added. Quality controls to confirm the identity of the monocytes were carried out according to published protocols ¹³ ,  ⁴⁴ .

Cytotoxicity and viability test

Viability testing was conducted on basal human monocyte cells obtained from one donor. Cells were exposed to a concentration gradient of the APIs or their mixture ( Fig. 1 ). Thus, cells were seeded in 96-well plates (well volume: 100 µL, 4 wells per concentration) at a density of 220 000 cells/well for viability measurements. Test items were dissolved in 0.1% dimethyl sulfoxide (DMSO) and 1 µL of the stock solutions or their dilutions were added to each well. NaF (100, 250, and 500 µg/mL) was used as a positive control, and the cell growth medium as a negative control.

After 24 h incubation, 10 µL alamarBlue (resazurin; Biosource) were added to each well. Fluorescence was measured after 2 h with a fluorescence spectrophotometer at 544EX nm/590EM nm to determine viability ¹³ ,  ⁴⁵ . The highest concentration of the constituents and of the mixture for the immunological assay was determined based on the cell toxicity of the mixture (10 µg/mL). The constituents were applied according to their share in the mixture.

Immunomodulation in human monocytes

The human monocyte model has been shown to reliably detect the anti-inflammatory mode of action of drugs and herbal extracts and has been described elsewhere ⁴⁶ . Briefly, human monocytes obtained from 3 donors were separately seeded in 24-well plates at a density of 2 200 000 cells/well. Thirty min after a treatment with 10 ng/mL LPS (from Salmonella enterica serotype typhimurium ; Sigma), the test items were added to the monocytes. The highest APIs and mixture concentrations to be tested in the immunomodulation model were chosen based on the viability tests. Five different concentrations of each test item ( Table 2 and Fig. 1 ) were dissolved in 0.1% DMSO. LPS-activated monocytes incubated with no test substances served as the control. All cultures contained 0.1% DMSO. Each concentration was tested on the 3 individual monocyte preparations in duplicate (n = 6). After 24 h incubation, the supernatants were removed and centrifuged. IL-1 β , IL-6, IL-8, MCP-1, PGE ₂ , and TNF- α concentrations in the wells were determined using commercially available ELISA assays (R&D, Caymen), according to the manufacturerʼs recommendations.

Statistical analysis

Data from the cytotoxicity assay are the results of the measurements of four individual wells and are expressed as % of cell viability of the respective controls. Results of the cytokine release assay are based on the measurements of six individual wells and are expressed as % of the corresponding controls (untreated LPS-activated monocytes for each donor), as the use of raw counts would introduce a large standard deviation caused by biological variation among buffy coats from different donors, thus not allowing for detecting whether any dose-dependent effects on cytokine release observed in one individual correlate with those seen in others ⁴⁷ . Inhibition of cytokine release > 20% was arbitrarily set as a cutoff value for biological significance. For the immunomodulation study, measurements from two wells were carried out for each of the three buffy coats, for a total of six measurements, and data are presented as means and standard deviations of six measurements, with the data normalized as percentages. An analysis of variance (ANOVA) with Bonferroni correction for multiple comparisons, but no interaction terms, was carried out for exploratory purposes only. Descriptive statistics and graphical displays were carried out using Stata SE (StataCorp LLC), version 18.

Contributorsʼ Statement

Nils Günnewich and Christian Zimmermann designed the study. Kurt Appel and Thorsten Rose performed all anti-inflammatory tests. Kurt Appel, Nils Günnewich and Christian Zimmerman jointly prepared the manuscript. All authors contributed to data analysis, drafting and critical review of the paper, gave final approval of the manuscript and are accountable for all aspects of the work.

Funding Statement

K. A. and T. R. gratefully acknowledge an unrestricted grant by Cesra Arzneimittel GmbH & Co. KG that has allowed for carrying out all experiments.