Chapter 4

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

II. The p-Locus (Pink-Eyed Dilution)

The p-locus on chromosome 7 ¹³ is characterized by 13 alleles. The wild type allele, P, produces an intense pigmentation of both the hair and eyes, whereas the oldest and most common mutant at this locus, p (for pink-eyed dilution), a mutation carried in many varieties of the mouse fancy, reduces the pigmentation of both eyes and the coat. The eyes of p/p mice resemble those of albinos, possessing a beautiful pink tint. However, in contrast to albino eyes, p/p eyes are not completely free of pigment; small amounts of melanin are found in the iris and retina (Durham, 1908, 1911; Little, 1913) and a few melanocytes occur in the choroid as well ( Markert and Silvers, 1956).

A. Influence on Coat Color

Insofar as the pigmentation of the hair is concerned p drastically reduces black and brown pigments, but has only a slight influence on phaeomelanin synthesis. Thus, except for the color of the eyes, A^y/a;p/p and A^y/a;P/— genotypes look much alike. Agouti pink-eyed animals also superficially resemble orange or yellow animals although they possess dull or slate grey hair bases which are recognized most readily when the hair is blown. The genotype a/a;B/B;p/p ( Plate 2-F) is known as "blue lilac" and resembles "maltese blue" ( a/a;B/B;d/d), but is lighter ( Grüneberg, 1952). Similarly, a/a;b/b;p/p mice, which are known as "champagne" or "cafe au lait," are essentially a lighter edition of dilute chocolate ( a/a;b/b;d/d) ( Grüneberg, 1952).

B. Interaction with B, B^lt, and b

We have noted already that p/p is a dominance modifier of B (on some genetic backgrounds a/a;B/b;p/p mice are a little lighter than a/a;B/B;p/p animals) and that B^lt is a dominance modifier of P ( a/a;B^lt/—;P/P mice have a different phenotype from a/a;B^lt/—;P/p genotypes). There is, however, another dominance modifier of P, namely, b/b. Thus Wallace ( 1953) reported that brown nonagouti mice heterozygous for pink-eyed ( a/a;b/b;P/p) are distinguishable from brown animals homozygous for the wild type allele ( a/a;b/b;P/P). Indeed, as a result of the interaction of B and b with P and p, six different phenotypes may occur ( Wallace, 1953). These are presented below:

Although, depending upon the genetic background, there may be various degrees of overlap among the phenotypes expressed by some of these different genotypes, e.g., genotypes 2 and 3 may overlap as may 5 and 6, it is evident that on at least some genetic backgrounds the dominance of P over p (and B over b) is incomplete (see Chapter 12, Section I, C).

C. Other Alleles

Other reported alleles at the p-locus include Japanese ruby ( p^r); dark pink-eye ( p^d); p-sterile ( p^s); p-black-eyed sterile ( p^bs); p-cleft palate ( p^cp); pink-eyed mottled-1 ( p^m1); pink-eyed mottled-2 ( p^m2); pink-eyed unstable ( p^un); p-extra dark ( p^x); p-darkening ( p^dn); p^6H and p^25H.

1. Japanese Ruby (p^r)

Japanese ruby (p^r), which unfortunately may now be extinct ( M.C. Green, 1966a), was first discovered in a stock of Japanese waltzing mice ( So and Imai, 1926). It is also recessive to wild type and has no effect on yellow ( Searle, 1968a). The eyes of p^r/p^r mice are very variable in color, ranging from almost pink to almost black. Moreover, the two eyes of a single individual may differ (heterochromia iridis). Because of this variability the classification of p^r/p^r mice must usually be based on the color of the coat which (on a nonagouti background) is intermediate between that of blue lilac ( a/a;p/p) and black. This allele seems to be completely dominant over p ( Grüneberg, 1952).

2. Dark Pink-Eye ( p^d)

Dark pink-eye ( p^d), a mutation which was probably X-ray induced, when homozygous slightly dilutes the color of the coat. The eyes of these homozygotes are only slightly pigmented at birth but darken within the next few days. The coat of p^d/p mice is lighter than p^d/p^d but darker than p/p. Their eyes are colorless at birth but darken during the first 2 weeks ( Carter, 1958). ¹⁴

3. Pink-Eyed Sterile (p^s)

Pink-eyed sterile ( p^s) is evidently another X-ray induced mutant (Hollander et al., 1960a, 1960b). The interesting feature of this allele, which is recessive to p, is its pleiotropic effects. In addition to its influence on pigmentation, which is the same as p, this allele has a number of other effects some of which seem unrelated. p^s/p^s mice not only weigh less at birth than their normal littermates but grow poorly, partly because their incisors wear abnormally and they have difficulty in chewing, and are noticeably smaller as adults (Hollander et. al, 1960a, 1960b). The locomotion of these mice is also somewhat uncoordinated and at times almost waltzing. Moreover, males are almost completely sterile as a consequence of both a poor mating response and a very high incidence of abnormal sperm. The sperm defect has been traced to an abnormality of the acrosome cap which occurs at the spermatid stage resulting in sperm heads which are very variable in shape and structure, although sperm motility is not lost. The Golgi body seems implicated in the acrosome defect. Since p^s/p heterozygotes have normal sperm it is apparent that sperm morphology depends on the paternal genotype ( Searle, 1968a). p^s/p^s females, although more fertile than males, likewise often fail to breed and, when fertile, produce few litters and are poor mothers (Hollander et al., 1960a, 1960b). p^s/p^s mice also display extensive lesions of the pars nervosa and median eminence. The pituitary gland is reduced in size [in fact, D. Hunt and D. Johnson ( 1971) suggest that the syndrome is caused by a pituitary dysfunction] and changes are also apparent in their adrenals and pancreas ( D. Johnson and D. Hunt, 1972). Even though p^s has a similar linkage relationship to the albino locus ( c) as p ( Hollander et al., 1960a) the many, seemingly unrelated, abnormalities associated with this allele suggest that it could represent a small deletion rather than a point mutation ( Melvold, 1974).

4. p^6H and p^25H

p^6H and p^25H are two other alleged p-locus alleles which affect coat color the same as p. Like p^s, these alleles were radiation induced and cause abnormal reproduction and behavior. D. Hunt and D. Johnson ( 1971) concluded from electron microscope studies that the abnormal sperm head shape in males homozygous for either of these alleles results from abnormal proacrosome formation, and multiple tail number results from a failure in cytokinesis. Until recently, a good case could be made for arguing that p^6H and p^25H represented the same mutation. This is no longer tenable; while both, when homozygous, produce grossly abnormal sperm, the frequency of morphological types is different. Moreover, unlike the situation in p^6H or p^25H homozygotes, where copulation is never observed, somewhat fewer than 50% of p^6H/p^25H males copulate with normal females and form vaginal plugs ( Wolfe et al., 1977).

5. p-Black-Eyed Sterile ( p^bs)

p-black-eyed sterile ( p^bs), originally known as p^24H ( R.J.S. Phillips, 1965), was found in the progeny of an irradiated male. A/A;B/B;p^bs/p^bs animals have dark eyes at birth, are lighter than A/A;B/B;p^d/p^d animals, and bear a closer resemblance to A/A;B/B;p/p than to A/A;B/B;P/P mice ( R.J.S. Phillips, 1977). Like p^s/p^s animals, p^bs homozygotes have a behavioral disorder as well as a high incidence of sterility accompanied, at least in males, by reduced pituitary gonadotropin levels ( Melvold, 1974). The sterility of homozygous p^bs males is caused by a sperm abnormality but one distinguishable from the abnormality in p^25H/p^25H mice ( D. Johnson and D. Hunt, 1972; Wolfe et al., 1977). Both abnormalities are expressed in p^bs/p^25H males ( D. Johnson and D. Hunt, 1972). The coat color of p^bs/p heterozygotes is intermediate between p/p and p^bs/p^bs and, like the latter, they have dark eyes at birth but behave normally ( R.J.S. Phillips, 1977) and have normal spermiogenesis ( D. Johnson and D. Hunt, 1972).

6. p-Cleft Palate ( p^cp)

p-cleft palate ( p^cp), formerly p¹¹ ( R.J.S. Phillips, 1973), was found at Harwell in the progeny of a newborn irradiated male. p^cp/p mice are indistinguishable from p/p animals, but most p^cp homozygotes, while also of the same color as p/p, have cleft-palate and die soon after birth. Although occasionally p^cp homozygotes survive either because they are unaffected or only slightly affected, they nevertheless produce offspring with the same high incidence of cleft palate ( R.J.S. Phillips, 1973).

7. p-Extra Dark ( p^x) and p-Darkening ( p^dn)

p-extra dark ( p^x) and p-darkening ( p^dn) are two p-alleles which have only recently been reported ( R.J.S. Phillips, 1977) although they occurred some time ago. p^x occurred as a spontaneous mutation in C3H and p^dn was induced with ethyl methanesulfonate (EMS). p^x/p^x homozygotes have dark eyes and are slightly lighter than wild type. p^dn homozygotes have pink eyes at birth which turn dark by weaning. Their coats are slightly darker than p/p and they have light ears.

The mutations p^x, p^d, p^bs, p^dn, p^6H, p^25H, and p^cp are all maintained at the MRC Radiobiological Unit at Harwell, England by R.J.S. Phillips who has prepared a tabular summary of their phenotypic effects (see Table 4-1). Dr. Phillips also notes that when these alleles are heterozygous, "dark on the whole appears to be dominant to light but there are some intermediate effects — for instance eye color in p^d/p, ear color in p^d/p or p^d/p^bs, and coat color in p^bs/p. It may be that most or all of the effects are intermediate but that the differences are not observable by eye." Moreover she adds " p^x and p^d fall into a darker, less p-looking category than p^bs and p^dn and this effect is phenotypically dominant" ( R.J.S. Phillips, 1977).

8. Pink-Eyed Mottled-1 ( p^m1) and Pink-Eyed Mottled-2 ( p^m2)

Pink-eyed mottled-1 ( p^m1) and pink-eyed mottled-2 ( p^m2) are two p-locus alleles which arose at Oak Ridge in different radiation experiments, but which behave similarly (L. Russell, 1964, 1965). When heterozygous with p, each of these mutations produces a coat in which wild-type areas and areas of typical pink-eyed dilute phenotype are freely intermingled (L. Russell, 1964, 1965). While, on the average, about 50% of the coat expresses the wild-type coloration and 50% that of p/p mice, there may be a great deal of variation among animals. Inasmuch as at least 50% of the more than 1000 offspring produced from mating p^mp animals with p/p mice were mottled, the remainder being pink-eyed dilute (L. Russell, 1964, 1965), it appears that this condition cannot be attributed to a somatically, highly mutable "wild-type" allele. ¹⁵ It is therefore apparent that p^m occurs in all cells but for some unknown reason produces intense pigment in some and typical pink-eyed dilute pigment in others. Although there is no evidence that p^m is the result of a position effect, small rearrangements or other types of very closely linked variable suppressors of P have not been completely ruled out (L. Russell, 1964, 1965).

There is some similarity between the effect of the p^m alleles and mottled agouti ( a^m), an allele at the a-locus (see Chapter 2, Section I, A, 12) which often produces agouti and nonagouti patches of fur freely intermingled at their edges (L. Russell, 1964, 1965). This similarity is especially striking since a^m/a animals also produce about 50% mottled progeny when mated to a/a mice (L. Russell, 1964, 1965). Nonetheless, it should be borne in mind that although the mottling of p^m/p and a^m/a may seem similar they actually are quite different. This stems from the fact that whereas the a-locus acts via the hair bulb environment (see Chapter 2, Section I, E and Chapter 7, Section IV), the p-locus acts autonomously within the melanocyte ( Silvers, 1958b).

9. p-Unstable ( p^un)

p-unstable ( p^un) [formerly p' or p^m ( Wolfe, 1963)] is the final allele at the p-locus. It also produces a phenotype consisting of areas of dilute and intense pigmentation. However, unlike p^m1 and p^m2, this bicolored pattern seems to be due to a spontaneous somatic reversion to wild type at a relatively high frequency at all stages of development. Data from experimental matings have shown that the frequency of somatic reversion of p^un to + occurs more often in p^un/p^un progeny of heterozygous parents, p/p^un or +/p^un mated inter se, than from homozygous p^un/p^un parents mated inter se ( Melvold, 1971). It is also higher in p^un/p^un progeny from somatically mosaic parents, even when there is no evidence for germinal mosaicism ( Melvold, 1971). The proportion of the coat affected appears to follow a bimodal distribution, i.e., it is usually either very small or quite large with few intermediate cases (the majority of mosaic animals fall into the former category). ¹⁶ Melvold has calculated that the spontaneous somatic reversion rates (at day 10 of embryonic life) in p^un/p^un progeny from matings in which the parents are mosaic or heterozygous, are generally two to three times that of p^un/p^un progeny from solid colored (nonmottled) p^un/p^un inter se matings. ¹⁷ While the exact basis for the expression of this allele is not known, Melvold suggests "that certain heterozygous combinations allow the formation, by recombination, of new alleles of varying stability and that the choice of mosaic mice for experimental matings may, in fact, result in selection for unstable alleles." ¹⁸

When p^un/p^un mice are bicolored (see Figure 7-5), their phenotypes correspond to those of allophenic mice produced from merging embryos known to bear different alleles for coat-color determinants which act via the melanocyte (Mintz, 1967, 1969a). Indeed, the similarities between the pigment pattern of p^un/p^un and allophenic mice strongly supports Mintz's contention that all of the melanocytes of the adult coat are clonally derived from a finite number of primordial melanoblasts originating along the mid-dorsum of the embryo ( Mintz, 1967 (see Chapter 7, Section VI). ¹⁹

D. Influence of Pink-Eyed Dilution on Pigment Granules

The influence that substituting p/p for +/+ has on the attributes of the pigment granules, and their distribution in the hair of different genotypes, has been carefully examined by E.S. Russell ( 1946, 1948, 1949a, 1949b). The major effect of this substitution is to alter the size of the pigment granules of all nonagouti genotypes, changing them to a very characteristic irregular shred shape, with flocculent clumping ( Figure 4-6). Tiny flecks of pigment are also present, and the edges of the granules never appear as clear or as sharp as in +/+ mice ( E. Russell, 1949b). As granule size is one of the key pigmentation characteristics affecting shape and color intensity, it is possible that the unique shape of p/p granules results from their small size ( E. Russell, 1949b). The color quality of the pigment itself is not altered, but the degree of pigmentation in both nonagoutis and yellows is decreased as a consequence of fewer cortical granules and an increase in pigmentation lag. Although a/a;p/p granules are unique, their attributes may nevertheless be affected, albeit in a minor way, by certain other genic substitutions. Thus, pink-eyed dilute brown ( a/a;b/b;p/p) granules are, on the average, shorter and more rounded that those of pink-eyed dilute black ( a/a;B/B;p/p) mice ( E. Russell, 1949b).

E. Ultrastructural Observations and Possible Gene Action

The influence that pink-eye dilution has on the ultrastructure of the nonagouti melanin granule has received attention from Moyer( 1961, 1963, 1966), Rittenhouse ( 1968b), and Hearing et al. ( 1973). Unfortunately there is no unanimity in their observations. Moyer observed that in the very early development of the +/+ melanosome thin single fibers aggregate to form compound fibers which cross-link and become oriented parallel to each other, however, in p/p mice the fibers do not become arranged in the same orderly parallel fashion and there is much less cross-linking between them ( Figure 4-7). As a consequence the mature granule never achieves the size of a "normal" granule and its shape is altered. Although this lack of orientation of the fiber matrix in p/p mice makes it difficult to assess the granularity of the melanin deposited, Moyer did find occasional granules in which at least a portion of the matrix was fairly well oriented. A study of the melanin in these areas indicated that the p-locus did not affect the granularity of the melanin nor, as might be expected, was the second-order periodicity altered. Thus according to Moyer, the p-locus appears someone to be involved with the early structure of the melanosome matrix possibly by controlling the protein which provides cross-linkages. This cross-linkage is basic to the structure of the final granule and could influence tyrosinase activity in numerous ways ( Wolfe and Coleman, 1966).

In contrast to these observations, however, are those of Rittenhouse ( 1968b). She reported little difference between the structure of p/p and +/+ granules. Indeed, according to her observations the only striking characteristic of both pink-eyed black and pink-eyed brown melanocyte granules is that, in contrast to the wild-type (nonagouti), a high proportion of them are lightly melanized or unmelanized. p/p granules also are noticeably smaller than +/+ granules, apparently because they fail to enlarge after melanization is initiated. Rittenhouse's comparison of the granules of pink-eyed black and pink-eyed brown melanocytes with each other, and with intense black and brown granules, is particularly interesting because they indicate that p/p can shift the disorderly internal framework she found characteristic of a/a;b/b;P/P granules — an internal framework resembling a tangled ball of strands ( Rittenhouse, 1968a) — toward the much more organized pattern of longitudinal strands and membranes characteristic of B/B granules (see Chapter 3, Sections I, F and II, G).

While one might attribute the different observations of Moyer and Rittenhouse to the fact that the latter examined only the granules of the hair bulb melanocytes whereas the former paid particular attention to those of the retina, this does not seem to be completely responsible for the discord. Thus, more recently Hearing and his colleagues have examined pigment granules from retinal epithelial cells and from choroidal melanocytes of p/p mice and report no decisive differences. Indeed, their observations are similar to those of Rittenhouse since they found the only significant difference between p/p and +/+ granules was that the former were much less melanized ( Hearing et al., 1973).

The elegant study of Sidman and Pearlstein ( 1965) deserves special consideration not only because it too includes some ultrastructural observations on pink-eyed granules, but because it focuses attention on another possible mode of action of p-series alleles. These investigators cultured eye tissue from postnatal p/p, p^un/p, and p^un/p^un mice in vitro and observed that when tyrosine was present in the culture medium the amount of pigment was greatly increased over that found in vivo. This increase in melanin occurred both in melanocytes and pigment epithelial cells and was unrelated to the structure of the granule. Thus, in agreement with Moyer's observations, Sidman and Pearlstein also found numerous melanosomes in their pink-eyed cultures with internal membranes in disarray. Indeed, the fact that these abnormal melanosomes appeared to make pigment as effectively in vitro as normal melanosomes while maintaining their abnormal morphology indicates that, contrary to the suggestion of Moyer ( 1963), their structure is not responsible for the reduced amount of pigment they normally display.

It therefore appears that the effect of the p-series of alleles on pigment formation may result from a failure to use an existing melanin-synthesizing enzyme system at a rate adequate to achieve normal pigmentation ( Sidman and Pearlstein, 1965). This possibility is in accord with other biochemical studies (Foster, 1963, 1965; Coleman, 1962, 1963) and, if confirmed, might help explain some of the pleiotropic effects of p-locus alleles. If for some reason p/p melanocytes cannot use tyrosine, or a further metabolite in the melanin pathway, at a normal rate because of some intracellular competition which favors some alternate use of available substrate, this diversion of tyrosine could be involved in producing some of the other effects attributed to the locus ( Sidman and Pearlstein, 1965). On the other hand, at least some of the pleiotropic effects attributed to this locus could very well be a consequence of chromosomal deletions involving adjacent loci concerned with other functions ( Melvold, 1974). Indeed, this possibility is especially attractive since those " p alleles" with the most severe effects were all radiation induced.