Estimation for Raw Material Plants of a Henna Product Using LC-High Resolution MS and Multivariate Analysis

Abstract

Henna is a plant-based dye obtained from the powdered leaf of the pigmented plant Lawsonia inermis, and has often been used for grey hair dyeing, treatment, and body painting. As a henna product, the leaves of Indigofera tinctoria and Cassia auriculata can be blended to produce different colour variations. Although allergy from henna products attributed to p-phenylenediamine, which is added to enhance the dye, is reported occasionally, raw material plants of henna products could also contribute to the allergy. In this study, we reported that raw material plants of commercial henna products distributed in Japan can be estimated by LC-high resolution MS (LC-HRMS) and multivariate analysis. Principal Component Analysis (PCA) score plot clearly separated 17 samples into three groups [I; henna, II; blended henna primarily comprising Indigofera tinctoria, III; Cassia auriculata]. This grouping was consistent with the ingredient lists of products except that one sample listed as henna was classified as Group III, indicating that its ingredient label may differ from the actual formulation. The ingredients characteristic to Groups I, II, and III by PCA were lawsone (1), indirubin (2), and rutin (3), respectively, which were reported to be contained in each plant as ingredients. Therefore, henna products can be considered to have been manufactured from these plants. This study is the first to estimate raw material plants used in commercial plant-based dye by LC-HRMS and multivariate analysis.

Introduction

Lawsonia inermis, synonyms L. alba and L. spinosa, is a flowering plant that is the sole species in the genus Lawsonia in Lythraceae¹⁾ and is a shrub cultivated in India, Sri Lanka, and North Africa.²⁾ Powdered leaves of L. inermis are often used as a henna (natural henna) for grey hair dyeing, treatment, and body painting,³⁾ and the dyeing principle is lawsone (2-hydroxy-1,4-naphthoquinone) (1) which is commercially known as Yellow Natural Orange 6 (Colour index Number 754800).⁴⁾

Allergy caused by a henna product has been reported.^5–9) Most of the allergy caused by henna products is likely to be attributed to the chemical colouring additives such as the aromatic amine compound p-phenylenediamine (PPD),⁹⁾ which was added without approval and necessary notification, to create ‘black henna’ that rapidly stains the hair black.¹⁰⁾ PPD is reported to cause severe delayed hypersensitivity reactions following skin contact.¹¹⁾ However, some cases of contact dermatitis by natural henna itself have been reported.^8,9) Additionally, commercial henna products can contain not only the leaves of L. inermis but also those of other plants,¹²⁾ and the possibility of allergy caused by other ingredients of the plant added later has also been suggested.¹⁰⁾ Hence, the cause of allergy can be attributed to the type of raw material plants.

In this study, we reported that raw material plants of henna products being distributed in the domestic market could be estimated using LC-high resolution MS (LC-HRMS) and multivariate analysis.

Results and Discussion

First, we collected 16 henna products (Samples 1–16) and a raw material of L. inermis (Sample 17), as shown in Table 1. In the products used in this study, I. tinctoria and C. auriculata were mainly listed as ingredients as well as L. inermis as raw material plants. PPD as a possible allergen was analyzed using LC-HRMS, confirming the absence of PPD in the henna samples (Supplementary Fig. S1).

Table 1. Information on Henna Products and PCA Results

Sample No.	Category	Ingredient labelling	Group No.
1	Henna product	Henna leaves	I
2	Henna product	Cassia auriculata leaves	III
3	Henna product	Indigofera tinctoria leaves, Henna leaves	II
4	Henna product	Henna leaves	I
5	Henna product	Henna	III
6	Henna product	Henna leaves	I
7	Henna product	Indigofera tinctoria leaves, Henna leaves	II
8	Henna product	Indigofera tinctoria leaves, Henna, Ammarok fruit, Ziziphus spina-christi, Eucalyptus leaves, Aloe vera leaves, Turmeric root, Chamomilla officinalis	II
9	Henna product	Henna	I
10	Henna product	Indigofera tinctoria leaves, Henna leaves	II
11	Henna product	Henna (highest quality)	I
12	Henna product	Henna	I
13	Henna product	Henna, Indigofera tinctoria leaves, Ammarok fruit	II
14	Henna product	Henna	I
15	Henna product	Cassia (Organically grown)	III
16	Henna product	Cassia auriculata leaves	III
17	Raw material	Powdered leaves of Lawsonia inermis from India	I

Next, 17 samples were analyzed by LC-HRMS at three times, and the analytical data was submitted to a multivariate analysis software. Typical total ion chromatograms are shown in Fig. 1. The Principal Components Analysis (PCA) score plot, shown in Fig. 2, could be readily separated into three different groups that correspond to ingredients on the label: Group I [Samples 1, 4, 6, 9, 11, 12, 14, and 17], Group II [Samples 3, 7, 8, 10, and 13], and Group III [Samples 2, 5, 15, and 16], indicating that Group I was henna, Group II was blended henna primarily comprising I. tinctoria, and that Group III was C. Auriculata. The raw material plant of Sample 5, classified into Group III, labelled with henna, was extremely likely to be C. auriculata, indicating that the ingredient labels may differ from the actual formulation.

Fig. 1. Total Ion Chromatograms of Typical Henna Products by LC-HRMS

(A) Positive ion mode, (B) Negative ion mode.

Fig. 2. PCA Score Plot of 17 Samples by LC-HRMS

PC1 occupies 40.9% and PC2 25.2% of total variance.

The corresponding PCA loadings were used to estimate the differential contributors responsible for the separation among groups. In the loading plot of this study, a contributor characteristic to Group I was estimated to be lawsone (1), which is a pigment ingredient of henna,⁴⁾ Group II was to be indirubin (2), a pinky-red pigment,¹³⁾ and Group III was to be rutin (3), a yellow pigment¹⁴⁾ (Fig. 3). A comparative analysis for authentic standards of estimated contributors and extracts was conducted, revealing that the retention times of extracted ion chromatograms of each contributor were identical to those of the characteristic peaks (Fig. 4), although lawsone was also detected in Group II because henna is blended. In addition, the MS/MS spectra of authentic standards and contributors of each group coincided, respectively (Supplementary Fig. S2). Consequently, each contributor characteristic to Groups I, II, and III was identified as lawsone, indirubin, and rutin, respectively (Fig. 5). In sample 5 labelled with henna, lawsone was not detected, but rutin was detected, indicating that the ingredient label of sample 5 would probably differ from the actual formation.

Fig. 3. PCA Loading Plot of 17 Samples

Fig. 4. Extracted Ion Chromatograms for Authentic Standards and Typical Henna Products (Samples 1–3, 5)

Vertical axes are aligned. (A) Lawsone (1) (m/z 173.02–173.03 [M − H⁻]). t_R: 12.8 min., (B) indirubin (2) (m/z 263.08–263.09 [M + H⁺]). t_R: 28.1 min., (C) rutin (3) (m/z 609.14–609.16 [M − H⁻]). t_R: 10.5 min. Each 10 µg/mL adjusted by methanol.

Fig. 5. Structures of Lawsone (1), Indirubin (2) and Rutin (3)

As lawsone from L. inermis,¹⁵⁾ indirubin from I. tinctoria,¹⁶⁾ and rutin from C. auriculata¹⁷⁾ were reported as ingredients, henna products can be considered to have been manufactured from these plants. Moreover, these ingredients might serve as marker compounds for estimating raw material plants in henna products. However, further research is needed because indirubin is applicable only to blended henna primarily comprising I. tinctoria, and rutin is a relatively common ingredient found in various plants.

We indicated that LC-HRMS and PCA analyses can be useful for the estimation of raw material plants used in a henna product, for which allergy have been reported. This estimation method has already been applied to crude drugs.¹⁸⁾ However, applications to commercial plant-based dyes have not been reported. This is the first study to report the application of LC-HRMS and multivariate analysis for estimation of raw material plants used in commercial plant-based dyes.

Conclusion

In this study, the raw material plants in henna products could be estimated using LC-HRMS and multivariant analysis. The PCA score plot clearly classified the samples into three groups [I: henna, II: blended henna primarily comprising I. tinctoria, III: C. auriculata]. One of the products whose ingredient was listed as henna was classified as C. auriculata group in a multivariant analysis as it was found to differ from the actual formation. Further, ingredients 1–3 characteristic to each raw material plant were identified, respectively, and these ingredients were reported to be contained in each plant. Thus, henna products can be estimated to have been manufactured from these plants. This study is the first to estimate raw material plants used in commercial plant-based dyes by LC-HRMS and multivariate analysis.

Experimental

Reagents

Methanol for special-grade (Kanto Chemical Co., Inc., Tokyo, Japan), 0.1% formic acid in water for LC/MS grade (FUJIFILM Wako Pure Chemical Corporation, Osaka, Japan), and acetonitrile for LC/MS grade (FUJIFILM Wako Pure Chemical Corporation) were used. Dibutyl hydroxytoluene (Lot. QRU5G, Tokyo Chemical Industry Co., Ltd., Tokyo, Japan), lawsone (Lot. 00012091-199, ChromaDex Inc., Los Angeles, CA, U.S.A.), indirubin (Lot. E7Y6K-SJ, Tokyo Chemical Industry Co., Ltd.), rutin (Lot. 151674, TargetMol Chemicals Inc., Boston, MA, U.S.A.), and p-phenylene diamine (Lot. LKH3977, FUJIFILM Wako Pure Chemical Corporation) were used.

Henna Products

Nine henna products (Nos. 1–9) were provided by patients in Fujita Health University School of Medicine. Seven henna products (Nos. 10–16) were purchased from several domestic shops. The raw material (No. 17) was purchased from Kohshin Bussan K. K. (Tokyo, Japan) imported from India (No. NNIHEEX/2157/0723, Ref. No.: KISP-276). Detailed information was shown in Table 1.

LC-HRMS Analysis

Pre-treatment for Analysis

A dried henna product and antioxidant dibutyl hydroxytoluene were weighted (10 mg each) to prepare a concentration of 1 mg/mL each with 50% methanol in water, filtered using a 0.45-µm filter, and then subsequently used as the sample solutions.

Analysis Condition

Analysis was performed using an Ultra HPLC-Orbitrap Exploris 120 MS (Thermo Fischer Scientific Inc., Waltham, MA, U.S.A.). The separation was used Xbridge BEH C18 column (2.1 mm Id × 100 mm, 5 µm; Waters Corporation, Milford, MA, U.S.A.) with a mobile phase consisting of 0.1% formic acid in water (A)/acetonitrile (B) system at a gradient mode of 5% B (0 min)→95% B (60 min)→95% B (70 min)→5% B (75 min)→5% B (85 min) at a flow rate of 0.2 mL/min. Acquisition time was 5 to 60 min. Column oven temperature was set at 40 °C. Injection volume was 2 µL. The MS condition was measured by electrospray ionization (ESI) positive/negative mode at a source voltage of 3.5 kV/ − 2.5 kV. Ion transfer tube temperature was 320 °C. Sheath gas was 50. Auxiliary gas was 10. Sweep gas was 1. RF lens was 70%. Scan mode was full scan/data dependent scan (range: m/z 80–800). Mass resolution was 60000. A measure time was 5–60 min. Each sample was analyzed by three times.

Principal Component Analysis

PCA was performed for decomposition and visualization of analysis data by LC-HRMS on Compound Discoverer™ version 3.2 software (Thermo Fisher Scientific Inc.). Sequencing was performed using ‘Untargeted metabolomics with statistics detect unknown using online databases and mzLogic’, compound estimation was performed using mzCloud Search and Chemspider Search, and the mzLogic algorithm was applied to rank the results. Detail of parameter was as follows:

For “Detect Compounds,” a mass tolerance of 5 ppm, intensity tolerance of 30%, S/N threshold of 3, minimum peak intensity of 1000000 were set. “Group Compounds” utilized a Retention time (t_R) tolerance of 0.2 min. For “Assign compound annotation,” a mass tolerance of 5 ppm, Sfit threshold of 20, and Sfit Range of 20 were used. Lastly, for “Predict compositions,” a mass tolerance of 5 ppm, intensity tolerance of 30%, intensity threshold of 0.10%, and S/N threshold of 3 were used.

Identification of Contributors of Each Group

Lawsone (1); HR-ESI-MS m/z 173.0244 [M − H]⁻ (Calcd. for 173.0239, C₁₀H₅O₃). t_R 12.8 min.

Indirubin (2); HR-ESI-MS m/z 263.0812 [M + H]⁺ (Calcd. for 263.0821, C₁₆H₁₁O₂N₂). t_R 28.1 min.

Rutin (3); HR-ESI-MS m/z 609.1459 [M − H]⁻ (Calcd. for 609.1456, C₂₇H₂₉O₁₆). t_R 10.5 min.

Funding

This work was supported by AMED under Grant Number JP21mk0101201.

Conflict of Interest

The authors declare no conflict of interest.

Supplementary Materials

This article contains supplementary materials.

References

• 1) Kumar S., Singh Y., Singh M., Central Arid Zone Research Institute, Jodhpur, India, 11–12 (2005).
• 2) Kang I.-J., Lee M.-H., Contact Dermatitis, 55, 26–29 (2006).
• 3) Rahman Bhuiyan M. A., Islam A., Ali A., Islam M. N., J. Clean. Prod., 167, 14–22 (2017).
• 4) Rehman F., Adeel S., Qaiser S., Bhatti I. A., Muhammad S., Zuber M., Radiat. Phys. Chem., 81, 1752–1756 (2012).
• 5) Lyon M. J., Shaw J. C., Linder J. L., Arch. Dermatol., 136, 124–125 (2000).
• 6) Niazi M., Mehrabani M., Namazi M. R., Salmanpour M., Heydari M., Karami M. M., Parvizi M. M., Fatemi I., Mehrbani M., Complement. Ther. Med., 49, 102316 (2020).
• 7) Alinaghi F., Bennike N. H., Egeberg A., Thyssen J. P., Johansen J. D., Contact Dermatitis, 79, 1–6 (2018).
• 8) Pérez R. G., González R., González M., Soloeta R., Contact Dermatitis, 48, 238 (2003).
• 9) Belhadjali H., Ghannouchi N., Amri Ch., Youssef M., Amri M., Zili J., Contact Dermatitis, 58, 182 (2008).
• 10) de Groot A. C., Contact Dermatitis, 69, 1–25 (2013).
• 11) Chen W., Nkosi T. A. N., Combrinck S., Viljoen A. M., Cartwright-jones C., J. Pharm. Biomed. Anal., 128, 119–125 (2016).
• 12) Palaniappan V., Murthy A. B., Sadhasivamohan A., Karthikeyan K., Contact Dermatitis, 89, 111–117 (2023).
• 13) Chanayath N., Lhieochaiphant S., Phutrakul S., CMU Journal, 1, 1–10 (2002).
• 14) Luo Y., Zhang X., Herit. Sci., 9, 85 (2021).
• 15) Charoensup R., Duangyod T., Palanuvej C., Ruangrungsi N., Pharmacognosy Res., 9, 60–64 (2017).
• 16) Paitoon A., Supachok S., Suree P., Shui-Tein C., Warasan Khana Witthayasat Maha Witthayalai Chiang Mai, 34, 329–337 (2007).
• 17) Girme A. S., Saste G. B., Shengule S. A., Kunkulol R. R., Hingorani L. L., J. Pharmacogn. Phytochem., 8, 77–81 (2019).
• 18) Tada A., Ishizuki K., Sugimoto N., Yoshimatsu K., Kawahara N., Suematsu T., Arifuku K., Fukai T., Tamura Y., Ohtsuki T., Tahara M., Yamazaki T., Akiyama H., Shokuhin Eiseigaku Zasshi, 56, 217–227 (2015).