JOURNAL OF COSMETIC SCIENCE 454 because we know that isopeptide bonding is critical to the chemical resistance (attack by oxidizing and reducing agents) displayed by these membranes in human hair and wool fi ber in the Allworden reaction and other reactions. Thus, the stability and resistance to chemical attack by this membranous material is provided through two types of cross- links: cystine crosslinks (vulnerable to oxidizing and reducing agents) and isopeptide crosslinks that are resistant to oxidizing and reducing agents. PROTEINS IN THE CORTICAL CELL MEMBRANES Proteinaceous material called “resistant membranes” have been isolated from both the oxidation of wool and/or hair with performic or peracetic acid followed by treatment with either ammonia or alkaline urea (71). The authors of this paper state that this material from the performic acid reaction is similar to that of their own analysis of the Allworden membrane nevertheless, it is clearly different from the Allworden membrane analysis by Allen et al. (30) as summarized in Table IV. Treatment of keratin fi bers with either peracetic or performic acids and separation into three fractions according to solubilities has been called the “keratose” method by Alexander and Earland (72). After oxidation, adjustment of the pH to the alkaline side provides an insoluble fraction called beta-keratose, about 10% of the weight of the hair. Acidifi ca- tion to pH 4 provides a fraction greater than 50%, called alpha-keratose, containing crystalline material, by X-ray diffraction, and the third fraction is called gamma-keratose. The beta-keratose fraction is believed to be proteins derived primarily from cell mem- brane material however, other proteins are likely to be present. According to a different workup procedure by Bradbury et al. (71), only 1–1.55% residue is provided. Other workup procedures have been applied to the keratose method (73). Table IV Allen’s (30) Analysis of the Allworden Membrane vs Resistant Membranes from Performic Acid Reaction with Wool Fiber (71) Amino acids Allworden membrane (30) Resistant membranes (71) Asp 3 5.4 Glu 8.6 10.3 Thr 2.1 5.7 Ser 14.3 10 Tyr 0 0 Pro 4.2 7.1 Gly 23.8 14.2 Ala 3.2 6.5 Val 5.6 4.9 Iso 1.2 2.6 Leu 2.9 4.9 Trp — 0 Phe 0.4 1.5 His 0.2 1.3 Lys 4.5 8.4 Arg 2.5 4.2 Met 0 0 Cys 21.1 13 Totals 97.6 100

CELL MEMBRANE COMPLEX 455 Since the cell membrane lipids of cortical cells are not bound by thioester linkages as in cuticle cells, but are bound by polar bonding and ionic bonds, a high level of cysteine links via an UHSP is not necessary for the cell membrane proteins of cortical cells. The resistant membrane material from the reaction of performic acid on wool fi ber by Bradbury et al. (71) provides only about 62% of the amount of cystine as the composition of the Allworden membrane by Allen et al. (30) and is consistent with the expectation of a lower cystine content in cortical cell membranes versus cuticle cell membranes. In ad- dition, one would expect the proteins of the cortical cell membranes to contain a larger number of basic amino acids (for bonding to cholesterol sulfate and to fatty acids), and more polar sites such as carboxyl groups would be preferred for polar bonding to fatty acid groups of lipids we fi nd nearly twice the basic amino acid content for salt link formation to cholesterol sulfate and fatty acids and more carboxyl groups of acidic amino acids in the membranes by Bradbury et al. (71). Furthermore, one would expect this performic acid-derived membrane material by Bradbury et al. to be richer in cortical cell membranes than cuticle cell membranes, since the cortical cell membranes should be a higher percentage of the total membrane matter in total wool fi ber, and the data is cer- tainly consistent with this expectation. A cleaner experimental scheme to isolate pure cortical cell membranes would be to start with pure cortex to exclude cuticle cell membranes and A-layer proteins. Pure cortex from human hair could be provided by the glass fi ber method of Wortmann et al. (74) and then perhaps by the performic acid reaction or another scheme to provide cortical cell membranes in the absence of cuticle contamination for further workup and analysis. PROTEINS EXTRACTED FROM HAIR/WOOL BELIEVED TO BE FROM THE DELTA LAYER OF THE CMC Leeder and Marshall (51) extracted Merino wool with formic acid and also with n-propanol/ water (50/50). Proteinaceous matter was removed from the hair fi bers with each of these solvent systems. With formic acid, these scientists concluded that the proteins were at least partially derived from the CMC, most likely from the delta layer because the extract contained virtually no cystine. If this proteinaceous material is from the delta layer, it most likely is not from the central proteins sometimes called the contact zone (or it con- tains small amounts of these proteins) because Naito et al. (75) have provided evidence that the contact zone contains hydrophilic protein with disulfi de bonding. Leeder and Marshall (51) concluded that the proteins derived by their own propanol/ water extraction of wool is not entirely from the cell membrane complex, but that they also contain high glycine–tyrosine proteins possibly derived from the cortex. The amino acid composition of proteins extracted by formic acid, by n-propanol/water, and by chlo- roform/methanol are compared with that of the Allworden membrane in Table V. Logan et al. (56) have shown that a chloroform–methanol azeotropic mixture provides a very different mixture of proteins than the high temperature propanol/water extraction (see Table V). Could this chloroform–methanol extraction be partially derived from corti- cal cell membranes or part of the outer lamella (outermost layer) of the delta layer pro- teins of the CMC? Since Mansour and Jones (37) have shown that chloroform/methanol extraction provides large changes to the cortex–cortex CMC in wool, it is possible that these proteins removed by chloroform/methanol extraction are at least partially attached to beta layers and are at least in part delta layer proteins of the cortex–cortex CMC.