Diagnosis of oculocutaneous albinism with molecular analysis

PURPOSE

To use molecular analysis to diagnose oculocutaneous albinism in a patient with an atypical clinical presentation.

METHODS

A 34-year-old woman with a history of strabismus and absent cutaneous pigment underwent comprehensive ophthalmic examination, visual-evoked potentials to detect altered optic decussation, and molecular analysis.

RESULTS

Examination showed fine nystagmus, iris transillumination, foveal hypoplasia, and corrected visual acuity of 20/25 in each eye. Misrouting of the retinostriate fibers was demonstrated with visual-evoked potentials. Mutations in the tyrosinase gene established the diagnosis of oculocutaneous albinism 1 even though the patient had atypical clinical features.

CONCLUSIONS

Molecular analysis can establish the diagnosis of oculocutaneous albinism 1 in the patient with atypical ocular features.

Linkage analysis with multiplexed short tandem repeat polymorphisms using infrared fluorescence and M13 tailed primers

The use of short tandem repeat polymorphisms (STRPs) as marker loci for linkage analysis is becoming increasingly important due to their large numbers in the human genome and their high degree of polymorphism. Fluorescence-based detection of the STRP pattern with an automated DNA sequencer has improved the efficiency of this technique by eliminating the need for radioactivity and producing a digitized autoradiogram-like image that can be used for computer analysis. In an effort to simplify the procedure and to reduce the cost of fluorescence STRP analysis, we have developed a technique known as multiplexing STRPs with tailed primers (MSTP) using primers that have a 19-bp extension, identical to the sequence of an M13 sequencing primer, on the 5' end of the forward primer in conjunction with multiplexing several primer pairs in a single polymerase chain reaction (PCR) amplification. The banding pattern is detected with the addition of the M13 primer-dye conjugate as the sole primer conjugated to the fluorescent dye, eliminating the need for direct conjugation of the infrared fluorescent dye to the STRP primers. The use of MSTP for linkage analysis greatly reduces the number of PCR reactions. Up to five primer pairs can be multiplexed together in the same reaction. At present, a set of 148 STRP markers spaced at an average genetic distance of 28 cM throughout the autosomal genome can be analyzed in 37 sets of multiplexed amplification reactions. We have automated the analysis of these patterns for linkage using software that both detects the STRP banding pattern and determines their sizes. This information can then be exported in a user-defined format from a database manager for linkage analysis.

A gene causing Hermansky-Pudlak syndrome in a Puerto Rican population maps to chromosome 10q2

Hermansky-Pudlak syndrome (HPS) is an autosomal recessive disorder that affects pigment production and platelet function and causes the deposition of a ceroid-like material in various tissues. Variability in the phenotype and the presence of several potential mouse models suggest that HPS may be a heterogeneous disorder. In order to identify a gene responsible for HPS, we collected blood samples from a relatively homogeneous population in Puerto Rico where the HPS carrier frequency is estimated to be 1 in 21. Analysis of pooled DNA samples allowed us to rapidly screen the genome for candidate loci, and significant evidence for linkage was detected for a marker on chromosome 10q. This region of the human genome is conserved syntenically with the region on mouse chromosome 19 where two possible mouse models for HPS, pale ear and ruby eye, are located. This linkage result was verified with additional markers, and a maximum LOD score of 5.07 at theta = .001 was calculated for marker D10S198. Haplotype analysis places the HPS gene in a region of approximately 14 cM that contains the markers D10S198 and D10S1239.

Detection of a Tsp509I polymorphism in the 3' UTR of the human tyrosinase related protein-1 (TYRP) gene

We have identified a Tsp509I polymorphism in the 3' UTR of the human tyrosinase related protein-1 gene (TYRP). TYRP is one of several genes involved in melanin pigment production.

Molecular basis of oculocutaneous albinism

Oculocutaneous albinism (OCA) is a complex group of genetic disorders that have historically been defined by clinical and biochemical methods. Recent advances in the molecular biology of pigmentation have greatly increased our understanding of the complexity of this group of disorders. To date, two different types of OCA (OCA1 and OCA2) have been mapped to specific chromosomal regions. Mutations have been found in the tyrosinase locus associated with OCA1 and the human homologue to the murine pink-eyed dilution locus associated with OCA2. Analysis of these genes and their mutations will allow us to better define and categorize the different types of albinism. Further, the analysis of these genes and their mutations will provide information on the role of these gene products in melanin biosynthesis and the effect specific mutations have on the pathogenesis of albinism.

Analysis of tyrosinase mutations associated with tyrosinase-related oculocutaneous albinism (OCA1)

Mutations of the tyrosinase gene associated with a partial or complete loss of enzymatic activity are responsible for tyrosinase related oculocutaneous albinism (OCA1). A large number of mutations have been identified and their analysis has provided insight into the biology of tyrosinase and the pathogenesis of these different mutations. Missense mutations produce their effect on the activity of an enzyme by altering an amino acid at a specific site. The location of these mutations in the peptide can be used to indicate potential domains important for enzymatic activity. Missense mutations of the tyrosinase polypeptide cluster in four regions, suggesting that these are important functional domains. Two of the potential domains involve the copper binding sites while the others are likely involved in substrate binding. More critical analysis of the copper binding domain of tyrosinase can be gained by analyzing the structure of hemocyanin, a copper-binding protein with a high degree of homology to tyrosinase in the copper binding region. This analysis indicates a single catalytic site in tyrosinase for all enzymatic activities.

Analysis of tyrosinase gene mutations using direct automated infrared fluorescence DNA sequencing of amplified exons

The ability to correctly diagnose the molecular cause of genetic diseases is becoming increasingly important in medicine. This requires an efficient method for the analysis of the DNA sequence of specific genes and the detection of mutations in affected individuals. We report a method to determine the mutations responsible for tyrosinase related albinism (OCA1) using a combination of polymerase chain reaction-single stranded conformational polymorphism (PCR-SSCP) analysis and direct DNA cycle sequencing using fluorescently labeled oligonucleotides and an automated DNA sequencer based on infrared fluorescence technology. This method allows DNA from several individuals to be sequenced quickly and simultaneously so that the specific location of each mutation and the carrier status of family members can be determined.

Evolution of the tyrosinase related gene (TYRL) in primates

Tyrosinase is the major enzyme responsible for the formation of melanin pigment and is found throughout the animal kingdom. In humans, the tyrosinase gene (TYR) maps to the long arm of chromosome 11 at band q14-->q21, while a tyrosinase related gene (TYRL) maps to the short arm of chromosome 11 at p11.2-->cen. We and others have found that the TYRL locus contains sequences that are similar to exons IV and V of the authentic tyrosinase gene but lacks sequences of exons I, II, and III. In an attempt to understand the evolution of the human tyrosinase gene, we have analyzed TYR and TYRL in primates and have found that exons IV and V of the chimpanzee and gorilla TYR are very similar to the human, with the gorilla sequence being more similar than the chimpanzee. We have also found that the gorilla but not the chimpanzee contains a TYRL locus similar to the human TYRL locus.

Molecular basis of type I (tyrosinase-related) oculocutaneous albinism: mutations and polymorphisms of the human tyrosinase gene

Type I (tyrosinase related) oculocutaneous albinism (OCA) results from mutations of the tyrosinase gene on chromosome 11q that lead to reduced or absent melanin pigment synthesis. The phenotype of Type I OCA is broad, ranging from a total lack to only a moderate reduction of melanin, and the phenotypic variation is associated with different mutant alleles at the tyrosinase locus. A total of 36 mutations have been identified in Type I OCA including 24 missense, 4 nonsense, and 8 frameshift mutations. The majority of affected individuals have been compound heterozygotes with different maternal and paternal alleles. Six polymorphic sites for haplotype analysis have been identified in the tyrosinase gene including 2 in the promoter region, 2 in the coding region associated with alternative amino acids in the protein, and 2 RFLPs in the first intron.

A frequent tyrosinase gene mutation associated with type I-A (tyrosinase-negative) oculocutaneous albinism in Puerto Rico

We have determined the mutations in the tyrosinase gene from 12 unrelated Puerto Rican individuals who have type I-A (tyrosinase-negative) oculocutaneous albinism (OCA). All but one individual are of Hispanic descent. Nine individuals were homozygous for a missense mutation (G47D) in exon I at codon 47. Two individuals were heterozygous for the G47D mutation, with one having a missense mutation at codon 373 (T373K) in the homologous allele and the other having an undetermined mutation in the homologous allele. One individual with negroid features was homozygous for a nonsense mutation (W236X). The population migration between Puerto Rico and the Canary Islands is well recognized. Analysis of three individuals with OCA from the Canary Islands showed that one was a compound heterozygote for the G47D mutation and for a novel missense mutation (L216M), one was homozygous for a missense mutation (P81L), and one was heterozygous for the missense mutation P81L. The G47D and P81L missense mutations have been previously described in extended families in the United States. Haplotypes were determined using four polymorphisms linked to the tyrosinase locus. Haplotype analysis showed that the G47D mutation occurred on a single haplotype, consistent with a common founder for all individuals having this mutation. Two different haplotypes were found associated with the P81L mutation, suggesting that this may be either a recurring mutation for the tyrosinase gene or a recombination between haplotypes.

Analysis of mutations in the copper B binding region associated with type I (tyrosinase-related) oculocutaneous albinism

Mutations of the tyrosinase gene are responsible for type I (tyrosinase-related) oculocutaneous albinism (OCA), an autosomal recessive genetic syndrome with a broad phenotypic spectrum. Mutant tyrosinase alleles can be associated with no melanin synthesis (I-A, tyrosinase-negative OCA), small to moderate amounts of melanin (I-B, yellow OCA) or unusual pigment patterns (I-TS, temperature-sensitive OCA). A total of 26 mutations of this gene have been described in type I OCA. Analysis of all known mis-sense mutations (n = 17) shows that most cluster in three areas of the coding region. Two clusters involve the copper A or copper B binding sites and may disrupt the metal ion-protein interaction necessary for enzyme function and the third cluster is located in exon I. Computer modeling of the secondary structure of the copper binding regions based on homology with the known crystal structure of hemocyanin show that they both consist of two alpha helices containing three histidine ligands that complex to a single copper atom. Mutations in the copper B binding region lie in the region between the two alpha helices that consists of a loop structure. These mutations may affect tyrosinase activity by either altering the position of the alpha helical domains and thus preventing proper copper binding to the histidine ligands, or affecting a catalytic or substrate binding site located between the two alpha helical domains.

Molecular analysis of type I-A (tyrosinase negative) oculocutaneous albinism

Type I oculocutaneous albinism (OCA) is caused by the reduction in or absence of activity of tyrosinase in melanocytes in skin, hair, and the eyes, the result of mutations of the tyrosinase gene. To date, a total of 22 unique mutations in the coding region of tyrosinase have been described in the literature. In this report we present 5 additional mutations of the tyrosinase gene associated with type I-A OCA in four individuals, including 2 missense, 1 frameshift and 2 nonsense mutations, and review the relevant literature on all published mutations. Analysis of the distribution of all identified missense mutations (n = 17) shows that most cluster in three areas of the gene and involve amino acids conserved between humans and the mouse. Two clusters involve the copper A and copper B binding sites and may disrupt the metal ion-protein interaction necessary for enzyme function. The third cluster in exon I could represent a functional domain important in enzyme function such as the tyrosine or the dihydroxyphenylalanine (DOPA) binding site of the enzyme. Small deletions or insertions resulting in frameshift mutations and nonsense mutations are distributed throughout the coding region and do not appear to cluster.

Three different frameshift mutations of the tyrosinase gene in type IA oculocutaneous albinism

Mutations in the gene for the pigment-producing enzyme tyrosinase are responsible for type IA (tyrosinase-negative) oculocutaneous albinism (OCA). Most reported mutations have been single base substitutions. We now report three different frameshift mutations in three unrelated individuals with type IA OCA. The first individual has a single base deletion within a series of five guanidines, resulting in a premature stop codon in exon I on one allele and a missense mutation at codon 382 in exon III on the homologous allele. The second individual is a genetic compound of two separate frameshift mutations, including both the same exon I single base deletion found in the first individual and a deletion of a thymidine-guanidine pair, within the sequence GTGTG, forming a termination codon (TAG) in exon I on the homologous allele. The third individual has a single base insertion in exon I on one allele and a missense mutation at codon 373 in exon III on the homologous allele. The two missense mutations occur within the copper Bbinding region and may interfere with either copper binding to the enzyme or oxygen binding to the copper. These five different mutations disrupt tyrosinase function and are associated with a total lack of melanin biosynthesis.

Molecular analysis of an extended family with type IA (tyrosinase-negative) oculocutaneous albinism

We have analyzed the tyrosinase coding region of three individuals having Type IA OCA within an extended family using genomic DNA amplification and dideoxy sequencing. Two of the affected individuals are dizygotic twins. All three have a common missense mutation at codon 81 (Pro----Leu) within exon I. The twins have a second missense mutation at codon 371 (Asn----Thr) within exon III and the third individual has a second missense mutation at codon 47 (Gly----Asp) within exon I. For each of these three individuals, the loss of enzyme function is the result of two different mutations, showing that they are compound heterozygotes of two mutant tyrosinase alleles.

Non-random distribution of missense mutations within the human tyrosinase gene in type I (tyrosinase-related) oculocutaneous albinism

Type I oculocutaneous albinism (OCA) is produced by mutations of the tyrosinase gene. We report four new missense mutations in the tyrosinase gene in patients with type IA OCA. Three of these mutations occur within exon I and the fourth mutation within exon IV. Analysis of the distribution of these four missense mutations and 12 previously reported missense mutations shows that most cluster in four areas of the gene. Two clusters involve the copper A and copper B binding sites and could disrupt the metal ion-protein interaction necessary for enzyme function. The other two clusters are in exon I and exon IV and could represent important functional domains of the enzyme. We conclude that analysis of the tyrosinase missense mutations will provide insight into the structure-function relationship of this enzyme.

Production and secretion of the 21-23.5 kDa prolactin-like molecules

We recently described the presence of a series of prolactin (PRL)-like molecules (PLMs) in the rat pituitary gland and showed that their formation was not due to artifactual proteolysis of 24 kDa PRL during extraction or to degradation of PRL in lysosomes. In this study we have found (1) in vitro translation of pituitary cell RNA to result in the production of only 24 kDa monomer isoform 2 and no PLMs, (2) that secretion of newly synthesized PLMs is differently regulated than at least a proportion of newly synthesized monomers, (3) that secretion of newly synthesized PLMs occurs after at least a 6 h delay, (4) that cysteamine (a) inhibits the release of the PLMs, (b) causes an increase in their amount versus isoform 2, and (c) causes an intracellular accumulation of pleiomorphic, immature secretory granules, and (5) that cells grown under degranulating culture conditions do not contain PLMs. These results, using normal anterior pituitary cells in primary culture, demonstrate the potential for differential release of the PLMs versus monomer PRL in vivo and are consistent with the production of the PLMs from 24 kDa monomer isoform 2 during secretory granule condensation.

Differential isoform distribution between stored and secreted prolactin

PRL exists within the mammotroph population in a number of different molecular forms. Three of these forms are best described as isoforms, as they have the same mol wt (24K) but differ in their net molecular charges. In this study we have examined the relative proportions of newly synthesized isoforms found in stored (intracellular) vs. secreted (extracellular) PRL. Dissociated cells from female rat anterior pituitaries were cultured for 48 h and then incubated in [35S]methionine (6 h; 37 C). Intracellular and medium proteins were then extracted and resolved by one- and two-dimensional polyacrylamide gel electrophoresis, followed by silver staining or autoradiography. Control experiments, in which biosynthetically labeled PRL was re-extracted, ensured that the isolation conditions did not in themselves promote isoform interconversion. The relative proportions of the PRL isoforms were determined by densitometric scanning of developed autoradiograms. In the cell extracts, the relative proportions were 13.6 +/- 2.1% isoform 1 (least negatively charged), 71.5 +/- 3.26% isoform 2, and 14.7 +/- 1.9% isoform 3 (most negatively charged). In the medium, the relative proportions were 60 +/- 2.89% isoform 1, 20 +/- 1.75% isoform 2, and 11 +/- 1.14% isoform 3. When the labeling was performed in the presence of 0.5 mM cysteamine (an agent we show to distinguish between newly synthesized and older stored hormone and, hence, between the previously described functional subpopulations of mammotrophs), the same ratios of newly synthesized isoforms were secreted from the cells. We conclude that secretion of the different isoforms is more complex than simple proportional release of each form, and based on the cysteamine results, this nonproportional release cannot be attributed to release of one isoform per functional subpopulation of mammotrophs.

Phosphorylation of prolactin

Rat prolactin exhibits microheterogeneity when examined in electrophoretic systems, running as three isoforms having the same molecular weight but different net charges (prolactins 1, 2, and 3 with isoform 3 being the most acidic). As there is precedent for the phosphorylation of a pituitary hormone and phosphorylation is a common cause of microheterogeneity, we examined the possibility that rat prolactin existed in differentially phosphorylated forms. The investigation included examinations of rat prolactin phosphorylation both in vitro and in vivo. For the in vitro studies, purified rat prolactin was incubated with [gamma-32P]ATP and low levels of each of five purified protein kinases. Phosphorylated rat prolactin was identified by autoradiography of silver-stained one- and two-dimensional gels. For the in vivo studies, rat anterior pituitary cells in primary culture were incubated in the presence of H3 32PO4 for 2 or 12 h, after which time the proteins were extracted from the cells, cold acetone-precipitated, or immunoprecipitated and run on two-dimensional gels. We report the in vitro phosphorylation of rat prolactin by cAMP-dependent protein kinase, casein kinase I, protease-activated kinase I, and the calcium/phospholipid-dependent kinase, that phosphorylation with these kinases results in phosphate incorporation only into isoforms 2 and 3, and the phosphorylation of prolactin in rat pituitary cells in primary culture.

Intracellular processing of prolactin

PRL has been reported to exist as a number of mol wt and charge variants. We have attempted to determine the relationships among the charge and lower mol wt variants by two-dimensional gel analysis of pituitary and pituitary cell extracts. Silver-stained gels of the extracts showed three major charge isoforms of 24,000 mol wt PRL (PRLs 1, 2, and 3, with PRL 3 being the most anodic) and what appeared to be an arc of products originating in the region of PRL 3. Each spot in the arc was distinct and represented a small decrease in size (two to eight residues), with a corresponding increase in net negative charge. When primary cell cultures were labeled with [35S] methionine, the three PRL isoforms and arc products were detectable by autoradiography. Western blots of the two-dimensional gels showed the arc products to be immunologically related to PRL. Treatment of the cell cultures with hydroxynorvaline (5 mM), which inhibits processing of pre-PRL to PRL, resulted in doublet spots in the arc. Elimination of protease inhibitors or an increase in temperature during protein isolation had no effect on the relative concentrations of PRL and arc products. Inclusion of standard PRL or [125I]iodo-PRL in the extraction solution did not increase the size of the spots in the arc or produce labeled arc products, respectively. Treatment of the cell cultures with chloroquine (10(-5) M) before and during radiolabeling had no effect on the production of radiolabeled arc products. Analysis of cell culture medium showed at least some of the arc products to be secreted. We conclude that PRL or a PRL-like molecule is processed intracellularly into a number of smaller derivatives. As the arc products accumulate within the cell and are secreted, we suggest that they may be biologically important relatives of PRL.

C pigment locus mutants of the fowl produce enzymatically inactive tyrosinase-like molecules

Three albino mutants of the fowl were tested for tyrosinase activity. Two of these mutants (c and ca) are alleles at the autosomal C locus, while the third mutant (sal) is sex-linked. Both the standard type, E, and sal are tyrosinase positive whereas the two C mutants are tyrosinase negative. Anti-chicken tyrosinase mouse serum was produced and all four genotypes were found to have cross-reacting material to this antiserum. Tyrosinase from the standard type was isolated and its location on denaturing two-dimensional gels determined. A co-migrating series of spots was found within the protein pattern of both the standard type and the tyrosinase positive albino, sal. The same pattern of spots was also observed for c and ca with no apparent change in either the pI or the molecular weight. Transmembrane blots also showed spots that reacted with anti-tyrosinase serum in all four genotypes and that migrated to the same location as that of standard tyrosinase. It is proposed that both c and ca are CRM+ mutants which produce tyrosinase-like molecules that are inactive due to a change that is electrophoretically and antigenically "silent".

Mammotroph autoregulation: uptake of secreted prolactin and inhibition of secretion

A dissociated preparation of normal adult rat pituitary cells has been used to study PRL autoregulation at the level of the mammotroph . Female rat pituitary cells previously cultured for 48 h on polylysine-coated petri dishes were washed to remove serum and accumulated PRL and then incubated in fresh medium in the absence or presence of increasing concentrations of rat PRL. Accurate balance sheets, allowing for degradation and nonspecific adsorption of PRL, showed exogenous PRL to regulate the amount of PRL released by the cells. That this regulation was partly produced by uptake of secreted PRL from the medium was demonstrated by supplementing the medium with [125I]iodo-rat PRL. Inhibition of secretion also played a role and was implied by experiments showing that ease of reversal of the inhibition was inversely proportional to the density of cell culture, which was itself proportional to the amount of PRL in the medium and the duration of autoregulation. These results indicate that normal adult rat pituitary cells in primary culture are capable of regulating the amount of PRL in their external milieu and that uptake of already secreted PRL is an important component of the regulatory mechanism.