A UPLC/ESI–MS method for identifying wool,cashmere and yak fibres

Abstract

A new objective method has been developed for the identification of animal hair fibres, in particular wool, cashmere and yak. Enzymatic digestion of keratin extracted from these fibres and peptide analysis by ultra-performance liquid chromatography/electrospray–mass spectrometry (UPLC/ESI–MS) allows not only qualitative determination of the presence of fibres derived from these species but also a quantitative assessment of the relative percentages present in blends. Such an analysis will provide reliable objective data about the authenticity of commercial products. The effectiveness of the UPLC/ESI–MS method was assessed by analysing known samples of these fibres and confirmed using unknown wool/cashmere/yak blends, and the results were compared with those obtained by SEM method IWTO 58–00.

Keywords

UPLC/ESI–MS wool cashmere yak

Speciality animal hair fibres, in particular cashmere, yak, mohair, camel and alpaca, are obtained from domestic mammals of the genera Capra, Bos, Camelus and Llama. These fibres are valuable natural raw materials used by the fashion industry for manufacturing high quality luxury textiles, and as such are distinct from fibres derived from sheeps' wool.

Cashmere is one of the finest and softest animal fibres used by the textile industry. It is produced from the down of a variety of domesticated double-coated goat (Capra hircus laniger) indigenous to Asia. The principal producers of cashmere in the world are China, Mongolia, Afghanistan and Iran. Very little is today supplied by the Kashmir Province of India, from which the name is derived. The cashmere products of this area first attracted the attention of Europeans in the early 1800s. Cashmere is regarded as a speciality or luxury fiber due to its scarcity and its high economic value, and also to its softness and lustre.¹ Differentiation of cashmere from other animal fibers is essential to discourage its adulteration with cheaper fibers.

Yak is the fine undercoat fiber of Bos grunniens, bred primarily in the Himalayan region of south Central Asia and the Tibetan plateau as far north as Mongolia. There is also a small threatened wild yak population in China and Mongolia. The production of yak fibers in China is about 7000 tons per year, which is 1.5 times that of Chinese cashmere.² The qualities of fine yak fiber are similar to those of cashmere,³ for example the mean fiber diameter of yak ranges from 15 to 20 µm, but its price is only a quarter that of cashmere. The bleaching of naturally pigmented fine yak hair is therefore highly attractive economically. Most animal fiber blends of global textile relevance comprise sheep's wool on the one hand plus one of the rare animal fibers, such as cashmere, mohair, camel hair or yak, on the other.⁴

The traditional and most widely used methods for identifying animal fibers for the textile sector involve microscopy. For example, optical microscopy allows the internal structure of the fiber, such as pigmentation and medulla to be observed,^5–6 and scanning electron microscopy (SEM) shows the surface morphology and the fine structure and arrangement of the cuticle cells at high resolution.^7–9

These techniques are however expensive and cumbersome. Moreover, the identification can be subjective, depending to a large extent on the expertise of the operator. A number of methods have been developed in the past to improve the objectivity and accuracy of the identification of the animal hair fiber blends, including methods based on the extraction and analysis of protein fractions,^10–14 external and internal lipids,^13,15–16 DNA,^17–18 and methods based on antigen–antibody specific selectivity that apply monoclonal antibodies known as ‘anti-cashmere’.¹⁹ Nevertheless, the results obtained by these methods are often affected by the chemical treatments to which the fibers have been submitted during manufacture, such as bleaching, dyeing or depigmentation.

In a previous study a method for detection and quantification of yak, wool and cashmere fibers in mixed samples was developed and validated.²⁰ The results highlighted the fact that molecular markers could be specifically and unequivocally linked to individual species, and the qualitative identification of cashmere, wool and yak was excellent. The method was also validated for quantitative analysis, but has not yet been extensively tested, and is not directly comparable to SEM, which is the current benchmark method for fiber analysis.

The aim of the present study was to analyze a range of samples of different composition using different chemical treatments such as UPLC/ESI–MS, and to compare them with the SEM method, in order to establish an authenticated quantitative technique.

Experimental details

Material and methods

Authentic samples of speciality fibers were kindly supplied by the Cashmere and Camel Hair Manufacturers' Institute (CCMI), Boston, MA. The samples were cut into short snippets, following the IWTO 58–00 procedure, before analysis.²¹

SEM analysis

All samples were examined by SEM, method IWTO 58–00,²¹ using a Cambridge Stereoscan 240 at an acceleration voltage of 15 kV and a working distance of 12–15 mm. The fibers were mounted in aluminum specimen stubs using double-sided adhesive tape, and sputter-coated with a 20 nm gold layer in rarefied argon using an Emitech K 550 sputter-coater and a current of 20 mA for 180 s.

Protein extraction

Samples were scoured, dehaired, cleaned with petroleum-ether for 2 h in a Soxhlet extractor, rinsed for 1 h in water at room temperature and finally for 1 h in water at 50℃. The fibres were dried in an oven at 50℃ and then conditioned in standard atmosphere at 20℃, 65% RH, for 24 h. The fibers (20 mg) were placed in a test tube and 5 ml of extraction buffer (25 mM Tris–HCl, pH 8.5, 2.4 M thiourea, 5 M urea, 5% dithiothreitol (DTT)) added. The buffer was left in contact with the fibers for three days at 50ºC under slow stirring. The solution was then filtered through a Büchner funnel and the filtrate centrifuged at 5000 g for 20 min at room temperature. The supernatant liquor was collected and the protein concentration of the extract determined by the Bradford protein assay method (BioRad) using bovine serum albumin as a standard.

Trypsin digestion

The solution was briefly sonicated in order to dissolve any crystals of thiourea potentially present, then to 67 µl of the solution an equal volume of 100 mM NH₄HCO₃ was added, mixing by vortex. A 5 µl portion of a 200 mM DTT solution in 100 mM NH₄HCO₃ was added, followed again by vortex mixing. The solution was incubated for 1 h at room temperature and 4 µl of iodoacetamide 1 M in 100 mM NH₄HCO₃ was added, followed again by vortex treatment. The solution was further incubated for 1 h at room temperature and the excess iodoacetamide eliminated by adding 20 µl of a 200 mM DTT solution, followed again by treatment with vortex and incubation for 1 h at room temperature. Finally, 818 µl of deionized water and 20 µl of trypsin solution (100 ng/µl in 1% CH₃COOH) were added, followed again by vortex treatment. The solution was allowed to stand overnight at 37℃, and the volume was then reduced under nitrogen flux.

UPLC/ESI–MS analysis

The dried samples were dissolved in 100 µl of eluent A (see below), treated with vortex and transferred to vials for UPLC analysis.

The analysis was performed using an Acquity UPLC BEH300 C18 1.7 µm column (Waters). Gradient elution was conducted using two eluents: A (H₂O + 0.2% CH₃CN + 0.1% HCOOH) and B (CH₃CN + 0.1% HCOOH). Gradient: from 0 to 7 min, isocratic 100% A, from 7 to 50 min from 100% to 50% A, from 50 to 52.6 min isocratic 50% A, from 52.6 min to 53 min from 50% A to 0% A, from 53 min to 58.2 min isocratic 0% A, plus reconditioning. Flow 0.2 ml/min, injection volume 5 µl, detection by single quadrupole ESI/MS SQD (Waters), positive ion mode.

The acquisition parameters were carried out using capillary voltage 3.20 kV, cone voltage 30 V, extractor voltage 4 V, source temperature 150℃, desolvation temperature 200℃, cone gas flow (N₂) 100 l/h and desolvation gas flow (N₂) 650 l/h:

(a) Full scan mode from 100 to 2000 m/z, scan time 1 s, interscan delay 0.1 s. Solvent delay 7 min.

(b) SIR: 3.20 kV. cone voltage 30 V, extractor voltage 4 V, source temperature 150℃, desolvation temperature 200℃, cone gas flow (N₂) 100 l/h, desolvation gas flow (N₂) 650 l/h. The ions monitored are those characteristic of every marker peptide (see our previous study²⁰), span 1, solvent delay 7 min.

Results and discussion

Keratin extraction yield

The method recently reported for quantitative assessment of different species in fibers is related to the relative abundance of specific peptide markers derived from keratin. Since the ratio of the peptides is also related to the actual proportion of the fibers, it was important to confirm that the extraction yielded an equal quantity of keratin from wool, cashmere and yak fibers. It was found that using the thiourea/urea/DTT buffer the extraction yield was 67.7% for wool and 70.6% for cashmere,¹⁹ and applying the same method the extraction yield for yak fibers was about 70.3% – in other words, the results were comparable. The extraction yields were also calculated for mixed and blind samples, and it was found that the yield percentages were about 67% for these samples. The decrease in percentage yield was probably due to previous chemical treatments, and was not significant for the present quantification.

UPLC/ESI–MS analysis

All peptide markers of the different species identified in the tryptic digests of the extracted keratin were identified by full scan analysis. Among the markers detected (Table 1) the best for quantitative assessment were cashmere 1, wool 1, cashmere 2, wool 2, wool–cashmere 1 and yak 1. In order to increase the sensitivity of detection by quantitative analysis, all markers were determined in mixed samples by single ion recording (SIR) detection, specifically monitoring only the characteristic ions. Chromatographic areas associated to the markers were integrated in order to estimate the percentage in mixed samples.

Table 1.

Marker peptides of different species in tryptic digests of extracted keratin

Molecular mass (Da)	Retention time (min)	Characteristic ions (m/z)	ID^a
2634	41.6	879.0	Cashmere 1 (C1)
2691	41.0	898.1	Cashmere 2 (C2)
2607	42.0	870.1	Wool 1 (W1)
2664	41.3	889.0	Wool 2 (W2)
2234	39.0	745.7	Wool–cashmere 1 (WC1)
4738	43.4	948.5 + 1185.6	Wool–cashmere 2 (WC2)
2504	38.9	835.5	Yak 1 (Y1)
2520	39.1	1260.2 + 840.9	Yak 2 (Y2)

^a Markers are identified for the species for which they are specific, implying that for other species they are absent. Wool–cashmere markers, as the name suggests, can be found in both wool and cashmere, but they are not present in yak.

The quantitative indices were calculated as follows:

Yak percentage = ((area yak 1) / (area yak 1 + area wool–cashmere 1)) × 100

Wool percentage 1 = ((area wool 1) / (area wool 1 + area cashmere 1)) × 100

Wool percentage 2 = ((area wool 2) / (area wool 2 + area cashmere 2)) × 100

Wool percentage = ((wool percentage 1) + (wool percentage 2)) / 2) × (100 – yak percentage) / 100)

Cashmere percentage = 100 – yak percentage – wool percentage.

In order to verify the correspondence between the calculated and the actual percentages, three different calibration curves were produced by preparing two-fiber mixed samples (wool–cashmere, wool–yak and cashmere–yak) with different percentage composition. The calibration curves obtained are shown in Figures 1, 2 and 3. Each individual blend was prepared in triplicate and injected, and for the calibration curves one pair of marker peptides were therefore considered for yak–cashmere and yak–wool and two pairs of marker peptides were separately considered only for the calibration curve of wool–cashmere.

Figure 1.

Calibration curve wool–cashmere.

Figure 2.

Calibration curve cashmere–yak.

Figure 3.

Calibration curve wool–yak.

The lowest percentage of foreign fiber in the mixed samples was 5%, and at this concentration the limit of detection conditions (3 × standard deviation) were verified. Since this condition applied to all the samples, it was confirmed that the method was adequate for detecting the presence of a foreign fiber down to 5%, and probably lower.

Blind samples were prepared by the Cashmere and Camel Hair Manufacturers' Institute composed of two or three different animal fibers. These were assessed using the calibration curves (Tables 2 and 3) and the results were compared to the data obtained by the traditional SEM method (Table 2) in order to test the validity of the UPLC/ESI–MS method. Table 3 gives the declared composition of the fiber blend prepared by CCMI.

Table 2.

Results of blind samples analyzed by UPLC/ESI–MS compared to SEM analysis results

UPLC/ESI–MS			SEM
Sample	C%	W%	Y%	C%	W%	Y%
1	35 ± 6	0	65 ± 6	present	0	present
2	12 ± 1	88 ± 1	0	13 ± 2	87 ± 2	0
3	20 ± 2	80 ± 2	0	19 ± 3	81 ± 3	0
4	46 ± 4	54 ± 4	0	52 ± 4	48 ± 4	0
5	75 ± 3	25 ± 3	0	84 ± 3	16 ± 3	0
6	0	100	0	0	100	0
7	0	100	0	0	100	0
8	8 ± 1	92 ± 1	0	4 ± 1	96 ± 1	0
9	6 ± 1	94 ± 1	0	4 ± 1	96 ± 1	0

The results are the average of three independent replicates; C% = percentage cashmere; W% = percentage wool; Y% = percentage yak.

Table 3.

Declared results

Declared
Sample	C%	W%	Y%
1	50	0	50
2	10	90	0
3	20	80	0
4	50	50	0
5	80	20	0
6	0	100	0
7	0	100	0
8	8	92	0
9	8	92	0

These results confirm that UPLC/ESI–MS is an accurate and entirely acceptable method for distinguishing and quantifying the composition of wool, cashmere and yak fiber mixtures. On the other hand, the SEM method was unable to quantify yak fiber on the basis of external morphology due to its similarity to wool and cashmere fibers.

In order to confirm the validity of the new method to recognize and quantify yak fibers, Table 4 gives the results obtained for an additional four samples (samples 10, 11, 12 and 13) that had been incorrectly identified by SEM due to the speciality fibers showing the morphological characteristics of cashmere. Since the morphological characteristics of cashmere and yak fibers overlap, in this case the analytical technique based on the use of SEM was unable to exclude the presence of yak hairs.

Table 4.

Results of blind samples analyzed by UPLC/ESI–MS compared to SEM analysis

UPLC/ESI–MS		SEM		Declared
Sample	C%	Y%	C%	Y%	C%	Y%
10	74 ± 2	26 ± 2	100	Doubt	75	25
11	100	0	100	Doubt	100	0
12	100	0	100	Doubt	100	0
13	71 ± 2	29 ± 2	100	Doubt	75	25

The results are the average of three independent replicates.

In order to assess the limits of the UPLC/ESI–MS method, a selection of treated samples (depigmented, dyed or bleached) were analyzed in different physical forms (staple, yarn or fabric):

– Sample A was depigmented yak (staple);

– Sample B was bleached wool (fabric);

– Sample C was three mixed fibers (staple);

– Sample D was depigmented yak and bleached cashmere (staple);

– Sample E was depigmented yak and fully bleached wool (staple);

– Sample F was dyed yak and cashmere (yarn).

The results obtained by the UPLC/ESI–MS method are summarized in Table 5.

Table 5.

Results of treated samples analyzed by UPLC/ESI–MS

Samples	UPLC/ESI–MS			Declared
Samples	C%	W%	Y%	C%	W%	Y%
A	0	0	100	0	0	100
B	0	100	0	0	100	0
C	39 ± 1	16 ± 1	45 ± 1	50	10	40
D	41 ± 7	0	59 ± 7	50	0	50
E	0	70 ± 1	30 ± 1	0	70	30
F	78 ± 1	0	22 ± 1	90	0	10

The results are the average of three independent replicates. It is concluded that the chemical treatments did not significantly influence the identification or the quantification analysis.

Conclusions

In an earlier study we identified specific markers for wool, cashmere and yak fibers. A method has now been developed and validated for the individual detection and quantification of these fibers in mixed samples. The limit of detection is around 5% and the precision of the analysis is confirmed to be very good. Using a number of samples with differing percentage composition and with different chemical treatments, the UPLC/ESI–MS method was assessed and compared to the traditional SEM method.

The results obtained by the UPLC/ESI–MS method were very satisfactory, giving an accurate quantification of the composition of fibers in the blend. This method thus resolved the problem created by SEM analysis regarding the quantification of cashmere and yak fibers in mixtures. It is important to emphasize that the SEM method still retains validity for cases not covered by the UPLC/ESI–MS method, for instance in discriminating between synthetic and natural fibers. The UPLC/ESI–MS method should therefore be regarded not as a substitute, but as complementary to the SEM methodology.

Footnotes

Acknowledgment

This work was supported by the Cashmere and Camel Hair Manufacturers' Institute as part of the project, The UPLC/ESI–MS method for the discrimination of animal fibers.

References

McCarthy

. Specialty animal fibres. Textiles 1998; 1: 6–8.

. Mini-handbook of wool spinning, weaving, dyeing and finishing [in Chinese], Textile Industry Press, Beijing, 1993, pp. 15–16.

Song

Yan

. An investigation of properties of Chinese special animal hair fibers [in Chinese]. J. China Textile Uni 1987; 13(6): 11–18.

Wortmann

McCarthy

. Cashmere/yak blends: a vexing analytical problem. Wool Record 2007; 166(3750): 33–33.

Greaves

Saville

. Microscopy of textile fibres: Microscopy Handbook 32, Oxford: Royal Microscopical Society, BIOS Scientific Publications, 1995.

Wildman

. The microscopy of animal fibres, Leeds, UK: Wool Industrial Research Association, 1954.

Wortmann

Arns

. Quantitative fiber mixture analysis by scanning electron microscopy. Text Res J 1986; 56: 442–446.

Phan

. Electron microscopy and the characterization of keratin fibres, Portugal: EEC Comett Eurotex, Universidade do Minho–Guimaraes, 1991.

Langley

Kennedy

. The identification of speciality fibres. Text Res J 1981; 51: 703–709.

10.

Bauters

Mazingue

Ponchel

. Différentiation de la laine et du mohair en mélange par leur characteristiques chimiques. Bull Scientifique ITF 1981; 10: 67–77.

11.

Bauters

Ponchel

. Démarque analytique dans le dosage laine–mohair par voie chimique. IWTO Report 1982, pp. 3–3.

12.

Wortmann

. Chemical characterisation of fine animal hair. 1988. Proceedings of the 1st International Symposium on Speciality Animal Fibres. Aachen, Germany, 39–70.

13.

Tucker

Hudson

AHF

. The chemistry of speciality animal fibres. 1990. Proceedings of the 2nd International Symposium on Speciality Animal Fibres. Aachen, Germany, 1–19.

14.

Sagar

AJG

Calvert

McCarty

. Characterisation of keratin fibres by chemical analysis. 1990. Proceedings of the 2nd International Symposium on Speciality Animal Fibres. Aachen, Germany, 147–164.

15.

Korner

. Lipids in the analysis of fine animal fibres. 1988. Proceedings of the 1st International Symposium on Speciality Animal Fibres. Aachen, Germany, 104–127.

16.

Rivett D, Logan R, Tucker DJ and Hudson AHF. The lipid composition of animal fibres. In: Proceedings of the 1st International Symposium on Speciality Animal Fibres. Aachen, Germany, 1988; pp.128–136.

17.

Hamlyn PF, Bayliffe A, McCarty BJ and Nelson G. Molecular speciation of animal fibres. J Soc Dyers Colourists 1998; 114: 78–80.

18.

Hamlyn PF, Nelson G, Brown R and McCarty BJ. DNA analysis for the routine identification of speciality fibres. In: Proceedings of the Symposium of Biotechnology in the Textile Industry. Povoa de Varzim (Portugal), 2000.

19.

Tonetti C, Vineis C, Aluigi A, Tonin C. Immunological method for the identification of animal hair fibres. Tex. Res. J. 2012; 82(8): 766–772.

20.

Paolella S, Bencivenni M, Lambertini F, Prandi B, Faccini A, Vineis C, Tonetti C and Sforza S. Identification and quantification of different species in animal fibres by LC/ESI-MS analysis of keratin-derived proteolytic peptides. Journal of Mass spectrometry, accepted for publication, 2013.

21.

IWTOTM58–00, International Wool Textile Organisation. Scanning electron microscopic analysis of speciality fibres and sheep's wool and their blends, IWTO TM–58–00, The Woolmark Company, Ilkley, UK, 2001.