Chapter 4

Dilute and Leaden, the p-Locus, Ruby-Eye, and Ruby-eye-2

I. Dilute (d) and Leaden (ln)

Recessive mutations at two loci of the mouse produce coat-color deviants which are phenotypically very similar. One of these mutations, described initially by Murray ( 1931, 1933), is known as leaden ( ln; chromosome 1) because when homozygous it transforms the intensely pigmented nonagouti coat color to bluish-grey; ¹ the other is dilute ( d; chromosome 9), originally known, because of its influence on black, as the Maltese or blue dilution gene of the mouse fancy. These determinants produce a similar dilution of brown ( Plate 2-E) and yellow pigment, giving the coat an overall "washed-out look" ( Searle, 1968a).

A. d-Locus Alleles

Mutations at the d locus are particularly interesting because in addition to their diluting effect on hair pigment, they may also affect the nervous system. ² Thus the d^l (for dilute-lethal mutation ³ — a mutation recessive to both D and d — is phenotypically indistinguishable from d/d but, unlike d, produces a severe neuromuscular disorder characterized by convulsions and opisthotonus (arching upward of the head and tail). Myelin degeneration occurs in the CNS of d^l homozygotes ( Kelton and Rauch, 1962) and they usually die at about 3 weeks of age ( Searle, 1952). ⁴ Although it has been reported that phenylalanine metabolism is seriously disturbed in these mice ( Coleman, 1960; Rauch and Yost, 1963) this remains to be confirmed ( Zannoni et al., 1966; Mauer and Sideman, 1967). ⁵

Two other mutations, d^s (slight dilution) and d¹⁵ (dilute-15), have also been described ⁶ which affect the nervous system. d^s/d^s homozygotes have a coat which is darker than d/d ( Dickie, 1965a) and develop a peculiar behavior at about 4 months of age that is more pronounced in females than in males. The animals sway, lean, and lose their balance. the feet seem unable to support the body properly and slide out sideways, and the back arches in what seems to be an opisthotonic spasm. This characteristic, however, does not interfere with reproduction or significantly shorten life span ( Dickie, 1967b). The color of d^s/d heterozygotes is intermediate between d^s/d^s and d/d ( Dickie, 1965a). The d¹⁵ mutation when homozygous produces a slightly diluted pelage, i.e., darker than d/d but lighter than D/—, as well as behavior abnormalities similar to, but less severe than, d^l/d^l. ⁷ d/d¹⁵ heterozygotes are similar in color to d¹⁵ homozygotes but behave normally ( R.J.S. Phillips, 1962). ⁸

B. Influence on Hair Pigment

In spite of the fact that mutations at the d locus have a dilution effect when introduced into genotypes which otherwise provide for intensely pigmented eumelanotic and phaeomelanotic hairs, this effect is not due to a reduction in the amount of pigment in the hair. To the contrary, mutations at the d-locus produce phenotypes which probably have on the average more hair pigment than the corresponding nondilute animals. This is certainly the case for d/d animals irrespective of whether the mutation is on a lethal yellow or nonagouti background ( Brauch and W. Russell, 1946; E. Russell, 1948). The d/d phenotype is brought about by the fact that between one-third and two-thirds of the pigment is deposited into a few very large, conspicuous clumps with clear-cut edges ( Figure 4-1a, b, and c; see Figure 4-6) ( E. Russell, 1949b) and these clumps, like the contracted melanophores in amphibian skin, have little effect on light absorption ( E. Russell, 1948; Grüneberg, 1952). This clumping of granules in the septules of the hair is accompanied usually by a reduction of cortical granules in proportion to medullary number, by the irregular arrangement of nonclumped granules, and by some degree of pigmentation lag, i.e., the tips of the hairs often possess little pigment ( E. Russell, 1949a), and these factors also undoubtedly contribute to the phenotype. Indeed, Onslow ( 1915) has suggested that melanin in the hair cortex maintains color intensity by preventing light from reflecting off the air spaces in the medulla. As might be expected the most dramatic clumps in d/d hairs occur in the medulla where they sometimes extend over several septules.

C. Dilute and Leaden: Similarities and Differences

The situation in leaden mice is essentially the same as in dilute ( Figure 4-1d) except that some leaden genotypes, e.g., chocolate leaden animals ( a/a;b/b;D/D;ln/ln), are a little lighter in color than the corresponding dilute type ( a/a;b/b;d/d;Ln/Ln). This appears to be due to a more pronounced pigment lag in ln/ln hairs rather than to any noticeable differences in pigment clumping (Poole and Silvers, unpublished). When genes for dilute and leaden occur together ( d/d;ln/ln) their effect on pigment is no different than that observed when leaden occurs alone. Nevertheless, there are some important differences in the behavior of these mutants. These include the fact that (1) leaden, but not dilute (Poole and Silvers, unpublished) is epistatic to recessive yellow ( e/e;ln/ln animals are indistinguishable from E/E;ln/ln mice) ( Hauschka et al., 1968); (2) whereas A^y/a;D/d mice are lighter than A^y/a;D/D animals, A^y/a;Ln/Ln and A^y/a;Ln/ln mice are indistinguishable (Poole and Silvers, unpublished) ⁹; and (3) leaden black skin is more active than dilute black in tyrosinase and dopa oxidase activity, as well as in the ability to undergo in vitro darkening ( Foster and L. Thomson, 1958). It should also be noted that while choroidal pigment granules tend to clump in both ln/ln and d/d mice ( Hearing et al., 1973), according to Moyer ( 1966) retinal granules are clumped in d/d but not in ln/ln animals (Hearing et al. report no clumping in the retina of either genotype). Finally, there is no evidence that leaden mice display any reduction in phenylalanine hydroxylase activity ( Wolfe and Coleman, 1966). These differences are important because they indicate that while these mutations often have similar phenotypic effects, their primary action(s) may be quite different.

D. Influence on Melanocyte Morphology and Behavior

Although the primary effects of these genes remain to be determined, histological examination of both d/d and ln/ln skin reveals a difference in the morphology of their melanocytes as compared with the melanocytes of the wild type ( D/D;Ln/Ln) which can account for how these loci produce their phenotypic effect. Whereas the melanocytes of wild type animals are characterized by the possession of long, relatively thick dendritic processes which contain a substantial portion of the cell substance, in d/d and ln/ln mice the pigment cells have fewer and thinner dendritic processes ( Figure 4-2). Because of this altered morphology the melanin granules are largely clumped around the nucleus in the body of the cell ( Markert and Silvers, 1956; Moyer, 1966; Rittenhouse, 1968a) [such a melanocyte has been described as nucleopetal as contrasted with the "normal" nucleofugal type ( Markert and Silvers, 1956)] and this crowding, in conjunction with the inadequate development of dendrites, undoubtedly results in an uneven release of granules from the melanocyte to the epidermal cells of the hair bulb ( Rittenhouse, 1968a). This is especially the case since there is evidence that pigment granules may be released into hair cells from the cell body as well as from the tips of dendrites ( Straile, 1964). The fact that d/d and ln/ln melanocytes have few dendrites, which are often stubby, may also prevent them from establishing significant contact with those epithelial cells destined to form the hair complex and, if this is the case, it would account for the reduction in cortical pigment in dilute and leaden hairs ( Straile, 1964; McGrath and Quevedo, 1965).

The nucleopetal morphology of d/d and ln/ln melanocytes could also explain why some of these cells appear to be incorporated into the hair shaft, especially at the termination of the hair growth cycle. Melanocytes with short, stubby dendrites may not be anchored as well as nucleofugal cells in the hair bulb and, as a consequence, become dislodged and swept into the hair shaft in the face of flowing epithelial cells ( Quevedo and Chase, 1958; McGrath and Quevedo, 1965; Straile, 1964; S. Sweet and Quevedo, 1968). Indeed, this incorporation of nucleopetal melanocytes into the hair is undoubtedly responsible for the two or three enormous clumps which are sometimes found at the base of the hair causing local enlargement of the hair diameter ( Figure 4-1c and d) ( E. Russell, 1949b; McGrath and Quevedo, 1965).

Inasmuch as active dendritic melanocytes are not limited to the hair follicles of mice but occur in the dermis and epidermis of certain areas of the body (ears, soles of the feet, tail, scrotum, muzzle, and genital papilla), as well as throughout the connective tissue that encapsulates and subdivides the harderian gland, ¹⁰ it is scarcely surprising that in d/d and ln/ln genotypes the melanocytes in these regions are likewise predominantly nucleopetal ( Markert and Silvers, 1956) (Figures 4-3 and 4-4). Thus while Gerson and Szabó ( 1968) observed no difference in the number of melanocytes in either the dermis or epidermis of ear, tail, palm, or scrotal skin of d/d vs D/D mice, they found that the melanocytes in these regions of d/d skin could always be distinguished from those in D/D skin on the basis of differences in cell size, congestion of perikarya, or number and shape of dendrites. Moreover, these investigators also observed significantly less melanin in the malpighian cells of d/d mice than in D/D animals, a difference manifested both by a decrease in the number of pigmented malpighian cells as well as by a smaller number of pigment granules in those cells which were melanized. This decrease in pigmentation, which is likewise exemplified by the observation that dilute black ( d/d;B/—) mice tan less well than intense black ( D/D;B/—) animals when exposed to UV ( Quevedo, 1965), undoubtedly results from the fact that because of fewer dendrites, nucleopetal melanocytes come into contact with fewer epidermal cells ( Gerson and Szabo, 1968), a situation which could also interfere with their cytocrine activity. ¹¹

E. Effect of Cellular Environment on Melanocyte Morphology

To determine whether the genes at the leaden and dilute loci govern melanocyte morphology through the cells of the tissue environment or whether these genes act primarily within the melanoblasts themselves, Markert and Silvers ( 1959) transplanted embryonic tissue containing melanoblasts from normal, leaden, and dilute animals into the anterior chambers of the eyes of adult albino or pink eyed mice having the same or different Ln and D constitution as the graft. In all instances melanocytes of nucleofugal genotype ( D/D;Ln/Ln) displayed a nucleofugal morphology in the anterior chamber of the host, regardless of its genotype ( Figure 4-5a). On the other hand, donor melanocytes of nucleopetal genotype ( d/d or ln/ln) always assumed shapes which varied all the way from typical nucleopetal to typical nucleofugal (Figure 4-5b and c). These results are consistent with the hypothesis that although genes at both the dilute and leaden loci exert their activity from within the developing melanoblast, the number and size of dendritic extensions of a melanocyte is probably a function of the environment in which the cell resides. Melanocytes of d/d and ln/ln mice have an innately weak capacity for extending dendrites, as indicated by their altered morphology in the rather compact tissue environments in which they normally occur. However, in less restrictive environments, such as in the anterior chamber of the eye, these melanocytes do extend more and longer dendritic processes and in many instances are nucleofugal ( Markert and Silvers, 1959). ¹²

Further evidence that the d locus acts within the melanoblast is provided by the work of Reed ( 1938). He found that when skin from newborn albino (nondilute) mice was transplanted to both neonatal dilute and nondilute hosts, the phenotype of the pigmented hairs originating within the border of the graft was always determined by the d-locus genotype of the invading cell.