26 JOURNAL OF COSMETIC SCIENCE DNA. No staining was evident for the cortex of the hair fiber, but staining was visible throughout all the cuticle layers. The same result was seen whether the hair was stained before or after sectioning. DISCUSSION This paper describes the extraction and identification of human DNA from the hair shaft, which substantiates previous findings (1-4), along with showing loss of DNA with washing and coloring, and lower levels of DNA in tip-end hair compared to root-end hair. DNA was found to be localized in the cuticle region of the hair fiber. The level of DNA extracted from hair shafts was low but confirms that DNA is therefore not completely degraded during the keratinization process. As PCR was successful for the HLA DQA1 genotype, DNA in some hairs must be relatively intact. The concen- tration of DNA was too low to be visualized on an agarose gel, thus rendering size estimation of the DNA difficult. Previous literature has, however, described the detec- tion of fully intact genomic DNA in the fiber (5). The inhibition of PCR by melanin, as has been previously described (7), was also confirmed, and hair extracts that had not amplified did amplify after being put through a purification column. When the levels of DNA were compared before and after coloring, coloring caused a significant decrease in levels of DNA. The peroxide present in colorant treatments may attack bonds within DNA, as occurs for protein disulphide bonds (8), fragmenting the DNA and allowing it to leach from the fibers on washing, as occurs for proteins (9). With washing, DNA was detectable in the wash solution after 20 washes, but not after one, five, or ten washes, suggesting the more the hair is washed, the more DNA is lost. The increase after 20 washes may be due in part to the levels of DNA from five and ten washes giving fluorescence signals below the absolute values that are recommended to be confirmed from the DNA calibration curve. The hair may also be becoming more damaged from washing, allowing DNA to leach out more easily. The highest level of DNA was lost after the hair had been soaked in surfactant for 2 h. This level of DNA loss may represent the maximum level of DNA that can be lost from the fiber, as large molecules of DNA may be trapped in the hair fiber while smaller molecules can leach out more easily. This theory is backed up by the finding that tip-end hair still had detectable levels of DNA, despite possibly having been washed several hundred times by the person from whom it was donated. Levels of DNA in hair that had been surfactant- washed many times have been shown in a previous study to be substantially lower compared to levels in unwashed hair, although the SDS used is a significantly harsher surfactant (5). In preliminary washing experiments, small switches were used, but the level of DNA washed out was too low to be accurately detected due to the larger amount of wash solution required. While cutting up the hair may have introduced a larger surface area and led to an increase in the amount of extractable material compared to an in vivo situation, the results described should be interpreted in relation to each other. DNA was significantly reduced in tip-end hair compared to root-end hair from real consumers' heads washed in the conventional manner, which backs up the results seen in vitro. It is hypothesized that the endocuticle region of the hair fiber contains remnants of the

DNA ANALYSIS OF HAIR FIBERS 27 cell nucleus (8) and that this would be a logical portion of the fiber to contain nucleic acid material. Fluorescence microscopy demonstrated for the first time that DNA does appear to be present in the cuticle of the hair fiber, although the exact location within the cuticle could not be determined at this level of magnification. Histochemical studies have shown nuclear remnants to be present in both the cuticle and cortex (8). Microscope and epi-illumination observations carried out in-house show disk-like remains of the cell nucleus in the endocuticle, pushed to one side as the A-layer, exocuticle, and inner layer were formed in the hair follicle. The endocuticle is the least organized of all the hair structures, and so more likely to have retained original cell organelles. The cortex is much more crystalline, and so any nuclear remnants are likely to be eliminated as the dense microfibrillar structures become set. The cuticle also keratinizes quite early, and so there may be less time for endonucleases to destroy DNA. These findings are further evidence that DNA is present in mature hair fibers and appears to reside in the cuticle. DNA is also affected by cosmetic procedures, including shampoo washing and permanent coloration. These findings add to our knowledge of the biological composition of hair and how it is affected by potentially damaging procedures. ACKNOWLEDGMENTS We thank Mr. M. Ashton for carrying out the fluorescence microscopy and Ms. N. Noel for technical assistance with the extraction and measurement of DNA. REFERENCES (1) M. R. Wilson, D. Polanskey, J. Butler, J. A. DiZinno, J. Replogle, and B. Budowle, Extraction, PCR amplification and sequencing of mitochondrial DNA from human hair shafts, BioTechniques, 18, 662-669 (1995). (2) R. Uchihi, K. Tamaki, T. Kojima, T. Yamamoto, and Y. Katsumata, Deoxyribonucleic acid (DNA) typing of human leukocyte antigen (HLA)-DQA 1 from single hairs in Japanese, J. Forensic Sci., 37, 853-859 (1992). (3) J. Kalbe, R. Kuropka, L. S. Meyer-Stork, S. L. Sauter, P. Loss, K. Henco, D. Riesner, H. Hocker, and H. Berndt, Isolation and characterization of high-molecular-mass DNA from hair shafts, Biol. Chem., 369, 413-416 (1988). (4) R. Higuchi, C. H. von Beroldingen, G. F. Sensabaugh, and H. A. Erlich, DNA typing from single hairs, Nature, 332, 543-545 (1988). (5) A. Schreiber, E. Amtmann, V. Stoerch, and G. Sauer, The extraction of high-molecular-mass DNA from hair shafts, FEBS Lett., 230, 209-211 (1988). (6) M. W. A. C. Hukkelhoven, E. Vromans, A.M. G. Markslag, A. J. M. Vermorken, A simple fluoff- metric microassay for DNA in hair follicles or fractions of hair follicles, Anticancer Res., 1, 341-344 (1981). (7) T. Yoshii, T. K. Akiyama, K. Tamura, I. Ishiyama, PCR inhibitor: Water-soluble melanin, which inhibits DNA polymerases and DNases, Adv. Forensic Haemogenetics, 5, 393-396 (1994). (8) C. R. Robbins, Chemical and Physical Behavior of Human Hair, 3rd ed. (Springer-Verlag, New York, 1994). (9) K. Kizawa, T. Takizawa, T. Inoue, I. Sasaki, M. Yamaguchi, and H. Uchiwa, New model for hair damage: The key role of S100A3 expressed in cuticles, Proceedings of the 20th IFSCC Congress, P014, 1-8 (1999).