Perturbed Epidermal Pterin Metabolism in Hermansky–PudlakSyndrome

Karin U. Schallreuter, Wayne D. Beazley, Nigel A. Hibberts, Norma N. Swanson,* and Mark R. Pittelkow*Clinical and Experimental Dermatology, Department of Biomedical Sciences, University of Bradford, Bradford, West Yorkshire, U.K.; *Department ofDermatology, Mayo Clinic, Rochester, Minnesota, U.S.A.

In Hermansky–Pudlak Syndrome (HPS) a mutation in a79.3 kDa transmembrane protein has been shown. Thefunction of this protein has escaped definition so far.This study unveils a defective (6R)-L-erythro-5,6,7,8-tetrahydrobiopterin (6BH4) de novo synthesis/recyclingfor this cofactor in HPS, where activities of the keyenzyme GTP-cyclohydrolase I are in the normal range,but total biopterin levels are significantly decreased inhomozygotes (n J 5) compared with unaffected controls(n J 4) (p J 0.00001). Phenylalanine hydroxylase and4a-hydroxy-6BH4-dehydratase activities are significantlylower. mRNA of all enzymes involved in 6BH4 biosyn-thesis/recycling and GTP-cyclohydrolase I feedbackregulatory protein were expressed in keratinocytes fromhomozygotes, heterozygotes, and healthy controls.Thioredoxin/thioredoxin reductase can directly controlthe redox status of 6BH4. These activities are allostericallycontrolled by calcium. Therefore calcium would directly

Hermansky–Pudlak Syndrome (HPS) is an autosomalrecessive inherited disease characterized by the triadof a tyrosinase positive oculocutaneous albinism, amild bleeding diathesis due to a storage pool deficiencyand an accumulation of ceroid-like material in

reticuloendothelial cells and tissues (Hermansky and Pudlak, 1959;Witkop et al, 1974). The bleeding disorder is associated with decreasednumbers of platelet dense bodies concomitant with a decrease of 5-hydroxy-tryptamine, nonmetabolic adenosine diphosphate, and calciumin these organelles (Hardisty and Hutton, 1967; Logan et al, 1971;Hardisty et al, 1972; Maurer et al, 1972; Weiss et al, 1974; Nagayamaet al, 1987; Fuse et al, 1992). An impaired calcium homeostasis wasobserved in epidermal keratinocytes of HPS homozygotes (Schallreuterand Pittelkow, 1989). In addition, elevated dolichol excretion in urinehas been described in the affected Puerto Rican population (Witkopet al, 1987).

Recently, homozygous frameshift mutations in a novel transmem-

Manuscript received November 1, 1997; revised May 4, 1998; accepted forpublication May 13, 1998.

Reprint requests to: Professor K.U. Schallreuter, Clinical and ExperimentalDermatology, Department of Biomedical Sciences, University of Bradford,Bradford BD7 1DP, West Yorkshire, U.K.

Abbreviations: DH, 4a-hydroxy-6BH4-dehydratase; GFRP, GTP-cyclohy-drolase I feedback regulatory protein; GTP-CH-I, GTP-cyclohydrolase I; HPS,Hermansky–Pudlak Syndrome; PAH, phenylalanine hydroxylase; PTPS, 6-pyruvoyl-tetrahydropterin-synthase; 6BH4, (6R)-L-erythro-5,6,7,8-tetrahy-drobiopterin; SR, sepiapterin reductase.

0022-202X/98/$10.50 · Copyright © 1998 by The Society for Investigative Dermatology, Inc.

511

affect this redox status. In HPS these enzyme activitiesare low concomitant with a defective calcium uptake,suggesting an extracellular accumulation of this secondmessenger. In this context phenylalanine hydroxylaseis subject to phosphorylation/activation by calcium/calmodulin activated kinases. Therefore it was anticipatedthat calcium could directly affect the cellular L-phenylal-anine turnover to L-tyrosine. A significantly more rapidL-phenylalanine uptake and its turnover to L-tyrosinewas identified in normal human melanocytes (n J 5)and keratinocytes (n J 2), and was more enhanced inmelanocytes in the presence of 2 H 10–3 M calcium. Theturnover to L-tyrosine was significantly slower. Based onall evidence to date, we speculate that the mutated proteinin HPS could be primarily involved in maintainingcalcium homeostasis in this patient group. Key words:HPS/pigmentation/pterins. J Invest Dermatol 111:511–516,1998

brane protein were described in four HPS populations, i.e., PuertoRico, Switzerland, Japan, and Ireland (Oh et al, 1996). It has beensuggested that this protein may control the development of severalcytoplasmic organelles (Oh et al, 1996); however, at the present time,the true function of this protein remains to be elucidated. Thissingle gene defect must account for variable clinical and phenotypicalexpression such as oculocutaneous albinism, a storage pool deficiencywith decreased numbers of dense bodies in platelets, a ceroid/lipofuscin-like accumulation in various cells and tissues (Witkop et al, 1990),normal numbers of melanocytes with abnormal melanosomes con-taining tyrosinase (Kugelman and Scott, 1961; Frenk and Lattion,1982), a defective antioxidant status (Schallreuter and Witkop, 1988;Wolfe et al, 1981; Schallreuter and Wood, 1989a; Schallreuter et al,1993), an altered calcium homeostasis in platelets and keratinocytes(Nagayama et al, 1987; Schallreuter and Pittelkow, 1989), an impaired5-hydroxy-tryptamine transport together with a reduction of thenonmetabolic adenosine diphosphate-pool in platelets (Gerritson et al,1977; Fuse et al, 1992), and an increase of nonmelanoma skin cancer.1

The clinical skin and hair phenotypes of HPS homozygotes can varyfrom totally white skin/hair to normal pigmentation. Homozygoteshave epidermal melanocytes and express tyrosinase activities in theirhairbulb melanocytes (Kugelman and Scott, 1961; Frenk and Lattion,1982; King and Olds, 1985), but only rarely exhibit full melanogenesis.It was demonstrated that the L-tyrosine supply can be altered inepidermal cells of the depigmentation disorder vitiligo by a defective

1Schallreuter KU, Witkop Jr CJ: unpublished results.

512 SCHALLREUTER ET AL THE JOURNAL OF INVESTIGATIVE DERMATOLOGY

Figure 1. The key enzyme for de novo synthesis of 6BH4, GTP-CH-I,uses GTP to form dihydroneopterin triphosphate followed by theconversion to sepiapterin by PTPS. Sepiapterin is metabolised by SR to6BH4. 6BH4 serves as a cofactor for PAH to form L-tyrosine from the essentialamino acid L-phenylalanine in the presence of O2. The recycling of the cofactorproduces 4a-hydroxy-tetrahydrobiopterin, and nonenzymatically 7BH4, theisomer of 6BH4. 4a-hydroxy-tetrahydrobiopterin is metabolised by DH toquinonoid dihydropterin (qBH2) and then finally reduced back to 6BH4 bydihydropteridine reductase in the presence of NADH. GFRP controls the denovo synthesis, where L-phenylalanine upregulates GTP-CH-I activities, whereas6BH4 exerts the opposite effect.

de novo synthesis/recycling process of the cofactor (6R)-L-erythro-5,6,7,8-tetrahydrobiopterin (6BH4) (Schallreuter et al, 1994a, b). Thiscofactor drives the metabolism from the essential amino acid L-phenylalanine to L-tyrosine via phenylalanine hydroxylase (PAH) andfrom tryptophan to 5-hydroxytryptamine via tryptophan monooxygen-ase and tyrosine hydroxylase (TH), the key enzyme for catecholaminebiosynthesis (Nichol et al, 1985).

Epidermal melanocytes and keratinocytes hold the full capacity forde novo synthesis/recycling of 6BH4 (Schallreuter et al, 1994a, b), whereGTP-cyclohydrolase I (GTP-CH-I) is the rate limiting enzyme for denovo synthesis and 4a-hydroxy-6BH4-dehydratase (DH) controls therecycling process by coupling to PAH (Nichol et al, 1985; Davisand Kaufman, 1989). The de novo synthesis is controlled by GTP-cyclohydrolase I feedback regulatory protein (GFRP) (Harada et al,1993) (Fig 1).

The important role of 6BH4 in de novo melanogenesis of differentskin types has been realised, where the induction of GTP-CH-I bytumor necrosis factor alpha (TNFα), PAH, DH, and 6BH4 followedUVB radiation after application of only one individual minimalerythema dose (Schallreuter et al, 1997). One consequence of thisinduction is a rise in epidermal L-tyrosine concentration. This increaseof L-tyrosine levels has been identified in vivo using noninvasiveFourier-Transform Raman spectroscopy.2 It should be taken intoconsideration that HPS homozygotes have melanocytes, and theirtyrosinase activity can be demonstrated in hairbulbs in the presence ofsufficient concentrations of L-tyrosine (Witkop et al, 1990). Wetherefore aimed to follow epidermal 6BH4 de novo synthesis/recyclingand the cellular uptake of L-phenylalanine and its’ turnover to L-tyrosine in the presence and absence of calcium in the Puerto Ricanpatient group, where this albinism presents the most common singlegene disorder in 1–1800 persons (Oh et al, 1996).

MATERIALS AND METHODS

Patients and probands A written consent was obtained prior to this studyfrom each participant or from his/her parents. Nineteen members of PuertoRican HPS kindreds served for this project. The HPS homozygotes (n 5 8)were classified previously (Schallreuter and Witkop, 1988). Four individuals were

2Schallreuter KU, Zschiesche M, Moore J, et al: Evidence for compromisedphenylalanine hydroxylase activity in vitiligo. Pigment Cell Res 10:357, 1997(abstr.)

Table I. Epidemiologic characteristics of patients andprobands

Number Sex Age Probands Skin types(Fitzpatrick classification)

1 M 49 HPS homozygote White/freckles2 F 44 HPS homozygote White/freckles3 F 26 HPS homozygote White4 F 34 HPS homozygote White/freckles5 F 15 HPS homozygote White6 M 18 HPS homozygote White/freckles7 M 40 HPS homozygote White/freckles8 M 22 HPS homozygote White9 F 69 Obligate heterozygote IV10 M 40 Obligate heterozygote V11 F 37 Obligate heterozygote V12 F 36 Obligate heterozygote V13 F 45 Unaffected III14 F 18 Unaffected IV15 F 17 Unaffected IV16 M 38 Unaffected IV17 F 37 Unaffected III18 M 15 Unaffected V19 M 21 Unaffected V

obligate heterozygotes and seven probands were clinically and phenotypicallyunremarkable. None of the probands were on any medication at the time pointof examination. Table I summarizes the patients/probands characteristics.

Epidermal cell extracts Epidermal suction blister roofs were obtained andprepared as described previously (Schallreuter et al, 1994b). Protein content ofeach cell extract was determined using the method of Kalb and Bernlohr (1977).

Cell cultures of epidermal keratinocytes and melanocytes Keratinocytesand melanocytes were established from epidermal suction blister roofs fromHPS homozygotes (n 5 4), one obligate heterozygote, one uncharacterizedfamily member, and four normal controls. Keratinocytes were cultured by themethod of Wille et al (1984). Melanocytes were established following themethod of Pittelkow et al (1989). Pigmentation was followed upon the additionof L-tyrosine (10–6 M) to the culture medium over time (day 0 to day 8).

Determination of total epidermal biopterins by high performance liquidchromatography Biopterins were analyzed after acidic iodine oxidation andprepurification by cation exchange chromatography using DOWEX AG 50(Wx8) followed by reverse phase high performance liquid chromatography andfluorimetric detection as previously described (Kerler et al, 1990; Ziegler andHultner, 1992). Biopterin levels were determined as pM per mg protein.

Enzyme assays GTP-CH-I was determined by reverse phase high perform-ance liquid chromatography following the formation of neopterin with alkalinephosphatase using previously published methods (Blau and Niederwieser, 1983;Kerler et al, 1990). Specific activity of GTP-CH-I was defined as pM neopterinmg per protein per h.

PAH activity was evaluated following the formation of 14C-labeled tyrosinefrom [14C]L-phenylalanine, using the method of Schallreuter et al (1994a, b).Activities were determined in nM per mg protein per min.

DH activities were assessed as previously described (Schallreuter et al, 1994a,b). Enzyme activities were determined in nM per mg protein per min.

Expression of epidermal mRNA from undifferentiated keratinocytes inHPS and controls using reverse transcriptase-polymerase chain reaction(PCR) Total RNA was extracted from undifferentiated keratinocytes of HPSpatients (n 5 4), one normal heterozygote normal neonatal foreskin (n 5 2),and adult skin (n 5 2) using Ambion Totally RNA Isolation Kit (AMSBiotechnology, Oxon, U.K.), and subjected to reverse transcriptase-PCRanalysis as described previously (Schallreuter et al, 1998). Primers and conditionsfor the PCR reactions were as described below. PCR products were separatedon a 1.5% agarose gel. All primers were designed based upon the cDNAsequence from Genebank acquisition numbers:

GTP-CH-I (U19523) 59CCTGAACGAGCACAGAATGA39 and 59ATCAAA-GTGGCAGAGATGGG39. PCR conditions were 35 cycles at 92°C 19, 60°C19, and 72°C 19.GFRP (U71890) 59GACAGGCACAGAGAGAAAG39 and 59TGAGAGGAG-GAATGGAAG39. PCR conditions were 40 cycles at 95°C 19, 55°C 19, and72°C 19.

VOL. 111, NO. 3 SEPTEMBER 1998 PTERINS IN HERMANSKY–PUDLAK SYNDROME 513

PAH (K03020) 59TTGCTGACATTGCCTACAAC39 and 59ACCCAAG-AAATCCCGAGAG39. PCR conditions were 45 cycles at 95°C 19, 55°C 19,and 72°C 19.DH (L41559) 59GCCATCTTCAAGCAGTTTC39 and 59AGCTTCAGT-CACCCTTTCC39. PCR conditions were 35 cycles at 92°C 19, 55°C 19, and72°C 19.TR (G29720) 59ACATGGAAGAACATGGCA39 and 59CTCCTCAGAAAG-GCCACAAG39. PCR conditions were 35 cycles at 92°C 19, 60°C 19, and72°C 19.Thioredoxin (J04026) 59TTTCCATCGGTCCTTACAGC39 and 59TTGGCT-CCAGAAAATTCACC39. PCR conditions were 35 cycles at 35°C 19, 60°C19, and 72°C 19.

Uptake of [14C]L-phenylalanine and its turnover in human melanocytesand keratinocytes The uptake of [14C]L-phenylalanine and its metabolismto [14C]L-tyrosine was followed over 20 min in intact confluent monolayersof human melanocytes (n 5 5), undifferentiated keratinocytes (n 5 2), anddifferentiated keratinocytes (n 5 2). In order to study the effect of calcium onthe L-phenylalanine uptake and its turnover, melanocytes were exposed to2 3 10–3 M calcium in the medium for 24 h. The concentrations ofboth amino acids were determined after separation by cellulose thin layerchromatography and counting in the 14C channel of a Beckman Scintillationcounter (Schallreuter et al, 1994b)

Uptake of [3H]L-tyrosine in human melanocytes The uptake of [3H]L-tyrosine was followed over 20 min in intact monolayers of human melanocytes(n 5 5). The possible effect of 2 3 10–3 M calcium on the L-tyrosine uptakewas in studied melanocytes only. The final concentration of the amino acidwas measured in the 3H channel as above. For final calculation the activity wascorrelated to mg protein (Schallreuter and Pittelkow, 1989).

Statistical analysis The data were analyzed by the standard paired Studentt test from the mean values.

Chemicals 6BH4, 7BH4, 6-biopterin, and 7-biopterin were from Dr.B. Schircks, (Jona, Switzerland). L-[14CU]-phenylalanine (513 mCi per mM)and L-[3.5 3H]-tyrosine (47 Ci per mM) were obtained from Amersham LifeSciences (Bucks, U.K.). Phenylalanine hydroxylase, catalase, and all otherreagents were from Sigma (Poole, Dorset, U.K.).

RESULTS

Epidermal GFRP and GTP-CH-I mRNA expression, and nor-mal enzyme activites in HPS GTP-CH-I represents the keyenzyme for de novo synthesis of the cofactor 6BH4 (Nichol et al, 1985).Messenger RNA was expressed in undifferentiated keratinocytes fromhomozygotes (n 5 4), one obligate heterozygote and two controls.Determination of epidermal GTP-CH-I specific activities in patientswith HPS (n 5 5), obligate heterozygotes (n 5 3), and controls (n 53) showed no significant difference between these three groups rangingfrom 1.6 to 2.1 pM per mg protein per h. Analysis of the mRNA forGFRP showed a weak expression for homozygotes (n 5 4), theuncharacterized family member as well as in one obligate heterozygotecompared with controls.

Decreased 6-biopterin levels in the epidermis of HPSEvaluation of total biopterins from epidermal suction blister cell extractsdemonstrated significantly lower levels of total 6-biopterin (i.e., 6BH4plus 6-biopterin) with unaffected 7-biopterin levels in HPS homozy-gotes (n 5 5) compared with heterozygotes (n 5 3) and controls (n 54) (Fig 2). Because GTP-CH-I activities are normal and 6BH4 levels

are low in homozygotes, we examined mRNA expression of 6-pyruvoyl-tetrahydropterin-synthase (PTPS) and sepiapterin reductase(SR), the enzymes catalyzing the second and third intermediary stepsbetween GTP-CH-I and 6BH4 formation (Fig 1). Both enzymes areexpressed in HPS homozygotes, heterozygotes, and controls.

Low epidermal PAH and DH activities in HPS Determinationof PAH activities in epidermal cell extracts from HPS homozygotes(n 5 8) were significantly lower compared with heterozygotes (n 54) and controls (n 5 7), whereas mRNA expression appeared to beunaffected (Fig 3A, B). Significantly lower DH activities were alsofound in HPS homozygotes (n 5 8), whereas enzyme activitieswere normal in obligate heterozygotes and controls (Fig 4). mRNAexpression for DH showed no differences compared with controls.

Figure 2. Total 6- and 7-biopterin levels in HPS homozygotes (n J 5),obligate heterozygotes (n J 3), and controls (n J 3). The 6-biopterinlevels are significantly lower in homozygotes compared with heterozygotes andcontrols (p 5 0.00001), whereas 7-biopterin levels are in the physiologic range.Mean 6SD.

Because the redox status of the pterins can be controlled by thethioredoxin/thioredoxin reductase system and the activities of thisimportant antioxidant mechanism are low in HPS homozygotes, wefollowed the expression of thioredoxin/thioredoxin reductase mRNA(Schallreuter and Witkop, 1988; Schallreuter et al, 1993, 1994c). Therewere no differences between homozygotes, heterozygotes, and controls.

Successful pigmentation of HPS melanocytes in vitro Oneconsequence of decreased epidermal PAH activity is a decrease in L-tyrosine formation from the essential amino acid L-phenylalanine inthe presence of 6BH4. HPS is a tyrosinase positive albinism withdifferent phenotypical expression and tyrosinase activities in hairbulbsfrom homozygotes (Kugelman and Scott, 1961). Consequently, weestablished melanocytes from two HPS homozygotes (clinically whitewith freckles on the arms and dorsum of the hands) to followpigmentation under in vitro conditions in the presence of L-tyrosine(10–6 M) from day 0 to day 8. Under these conditions, melaninproduction started at day 2 (Fig 5).

The importance of L-phenylalanine uptake and intracellulartyrosine formation versus L-tyrosine uptake in melanocytescompared with keratinocytes It is noteworthy that pigmentationrequires 10–6 M L-tyrosine (Prota, 1992). Therefore the question arises;where does the intracellular tyrosine originate? Figure 6 demonstratesa rapid uptake of [14C]L-phenylalanine and its intracellular turnoverto [14C]L-tyrosine in human normal melanocytes (n 5 5) over 20 minin the presence of 0.2 and 2.0 3 10–3 M calcium. The results clearlyshow a significantly faster uptake of L-phenylalanine in the presenceof 2 3 10–3 M calcium, whereas the turnover to L-tyrosine is sloweddown. In order to compare this specific L-phenylalanine uptake andturnover with the direct L-tyrosine uptake, the experiment was repeatedin the presence of [3H]L-tyrosine. The [3H]L-tyrosine uptake is notaffected by calcium. Unfortunately we did not have sufficient numbers

514 SCHALLREUTER ET AL THE JOURNAL OF INVESTIGATIVE DERMATOLOGY

Figure 3. PAH activities and mRNA expression in HPS patientscompared with healthy controls. (A) PAH activities (nM per mg proteinper min) in HPS homozygotes (n 5 8), obligate heterozygotes (n 5 4), andcontrols (n 5 7). There is significantly lower activity in HPS homozygotes(p 5 0.0001) supporting a shortage in intracellular L-tyrosine supply. (B)Reverse transcriptase-PCR of PAH mRNA expression standardized to betaactin. The order of samples on the gel is from left to right: negative control,unrelated healthy control (n 5 2), uncharacterized family member (n 5 1),homozygote (n 5 4), obligate heterozygote (n 5 1).

of melanocytes to repeat this experiment with HPS cells. Figure 6illustrates a representative experiment of the L-tyrosine uptake, L-phenylalanine uptake, and its turnover to L-tyrosine. These kineticanalyses identify active transport for L-phenylalanine, whereas L-tyrosine shows facilitated diffusion.

DISCUSSION

A linkage analysis of the Puerto Rican HPS population mapped thissyndrome on chromosome 10q23 and the HPS gene was clonedfollowed by cDNA sequence analysis (Oh et al, 1996). The HPS genehas 20 exons coding for a 79.3 kDa protein. At the present time thisprotein has no reported sequence homology and its’ function is

Figures 4. DH activities (nM per mg protein per min) in HPShomozygotes (n J 8), obligate heterozygotes (n J 4), and controls(n J 7). The results show significantly lower DH activities in HPS homozygotes(p 5 0.0001). Mean 6 SD.

Figure 5. Epidermal melanocytes (passage 3) from one HPS homozygotein the presence of 10–6 M L-tyrosine show a clear formation of melaninin all cells under in vitro conditions (after 6 d in culture). Scale bar: 50 µm.

unknown (Oh et al, 1996). Accordingly, all of the Puerto Ricanpatients studied so far were homozygous for a 16 base pair duplicationin exon 15 indicative of a founder effect. Different mutations in theHPS gene were detected in Swiss, Irish, and Japanese probands withthis disorder (Oh et al, 1996). Puerto Rican HPS patients expressdifferent phenotypes with regard to skin and hair pigmentation(Schallreuter and Witkop, 1988). This result could point to a possibleregulation of pigmentation by the HPS protein rather than a directeffect on tyrosinase.

Our report shows that HPS melanocytes have the capacity in vitroto synthesize melanin in the presence of sufficient L-tyrosine concentra-tions. Because both melanocytes and keratinocytes have the full capacityto synthesize and recycle 6BH4 for their individual synthesis of L-

VOL. 111, NO. 3 SEPTEMBER 1998 PTERINS IN HERMANSKY–PUDLAK SYNDROME 515

Figure 6. L-phenylalanine uptake and metabolism in normal humanmelanocytes and keratinocytes. (A) A comparative study of [14C]L-phenylalanine uptake and metabolism to [14C]L-tyrosine over 20 min in normalhuman melanocytes (n 5 5) and keratinocytes (n 5 2) in the presence of0.2 3 10–3 M and 2 3 10–3 M calcium in the culture medium; (B) [14C]L-phenylalanine active transport kinetics in normal human melanocytes in thepresence of 0.2 3 10–3 M calcium and conversion to [14C]L-tyrosine comparedwith [3H]L-tyrosine facilitated diffusion under standardized experimentalconditions. For each experiment, 130 µM amino acid was added to the culturemedium and the time course for uptake was followed over 20 min.

tyrosine from the essential amino acid L-phenylalanine (Schallreuteret al, 1994a, b), it is noteworthy that these cells do not depend on anextracellular L-tyrosine supply, as suggested recently (Baillin et al, 1997;Feng et al, 1997). Our new data on L-phenylalanine uptake in intactmonolayers of healthy melanocytes clearly show a preferential transportof L-phenylalanine over L-tyrosine. These cells can acquire L-tyrosineas demonstrated in the induced pigmentation of HPS melanocytes,and in the [3H]L-tyrosine uptake studies, although this process is muchslower (Fig 6B). Our results also indicate that HPS homozygotes aredeficient in both the de novo synthesis and the recycling of 6BH4, aresult that could explain the low availability of L-tyrosine for melaninsynthesis in the majority of HPS homozygotes.

Total biopterin levels are significantly reduced even though GTP-CH-I is expressed and activities are normal in these patients; however,low 6BH4 levels could also be the result of impaired PTPS and/or SRactivities, enzymes catalysing the intermediate steps between GTP-CH-I and 6BH4 synthesis (Fig 1). Both PTPS and SR mRNA wereexpressed in all samples, but enzyme activities remain to be determined.In the recycling process of 6BH4, both PAH and DH mRNA areexpressed, although activities were reduced in HPS homozygotes.Analyses of the mRNA expression of thioredoxin/thioredoxin reduc-tase is unremarkable in HPS, despite the significantly reduced activitiesof membrane-associated TR in homozygotes in the Puerto Rican andSwiss populations (Schallreuter and Witkop, 1988; Schallreuter et al,1993). It is noteworthy that the decrease in the activities of thisimportant antioxidant enzyme was more pronounced in the PuertoRican compared with the Swiss population, which could account forthe clinically different expression of the disease and the differentmutations reported in these two locations (Schallreuter and Witkop,1988; Schallreuter et al, 1993; Oh et al, 1996).

How can the perturbations in 6BH4 synthesis/recycling, togetherwith TR-mediated decreased antioxidant status in membranes and animpaired calcium homeostasis, be reconciled in terms of the recentlyestablished mutations in the HPS protein? To date there are six majorclues in support of a perturbed calcium homeostasis:

1 Calcium is an important constituent of platelet dense bodies.Its uptake/efflux is impaired in keratinocytes from HPS individualscompared with controls as well as in their platelets (Schallreuter andPittelkow, 1989).2 Calcium influences the uptake of the essential amino acid L-

phenylalanine in normal human melanocytes and keratinocytes, whereasL-tyrosine uptake is unaffected.3 Calcium decreased the intracellular L-phenylalanine turnover to L-tyrosine in normal human melanocytes.4 PTPS, SR, and the coupled PAH/DH systems are activated byphosphorylation involving Ca21/calmodulin dependent kinases. SR aswell as the important 6BH4 recycling enzyme dihydropteridine reduc-tase, are inactivated by calcium activated neutral proteases (Katch et al,1994, 1997; Thony et al, 1997).5 TR is downregulated allosterically by calcium binding to a singleEF-hands regulatory domain. It could be possible that the reduced TRactivities in HPS homozygotes are caused by a direct calcium inhibitionof this enzyme (Schallreuter and Wood, 1989b).6 The formation of ceroid/lipofuscin like material in HPS homozygotesfrom Puerto Rico has been linked to lipid peroxidation (Wolfe et al,1981). The downregulation of TR activities, as a major antioxidantdefence mechanism in HPS, would support oxidative stress.

In summary, we would like to propose that the mutated gene inHPS on chromosome 10q23 expressing a protein with a single membraneassociated α-helix in its structure, could play an important role inorganelle-membrane function related to calcium channel integrity orcalcium ATPase activity. Further research in our laboratory is directedtoward elucidation of this interesting hypothesis.

The authors are deeply indebted to the patients and their family members who supportedthis research. We wish to dedicate this article to the late Carl Witkop, Jr., who was theoriginal generous mentor to KUS. The work has also been supported by a grant fromStiefel International to KUS.

REFERENCES

Baillin UT, Oh J, Feng GH, Fukai L, Spritz RA: Organisation and nucleotide sequenceof the human Hermansky Pudlak Syndrome (HPS) gene. J Invest Dermatol 108:923–927, 1997

Blau N, Niederwieser A: Guanosine triphosphate cyclohydrolase I assay in human and ratliver using performance liquid chromatography of neopterin phosphates andguanosine nucleotides. Anal Biochem 128:446–452, 1983

Davis MD, Kaufman S: Evidence for the formation of the 4a-carbinolamine during thetyrosine-dependent oxidation of tetrahydrobiopterin by rat liver phenylalaninehydroxylase. J Biol Chem 264:8585–8596, 1989

Feng GH, Baillin T, Oh J, Spritz RA: Mouse pale ear (ep) is homologous to humanHermansky Pudlak Syndrome and contains a rare ‘‘AT-AC’’ intron. Hum Mol Genet6:7922–7927, 1997

Frenk E, Lattion F: The melanin pigmentary disorder in a family with Hermansky PudlakSyndrome. J Invest Dermatol 78:141–143, 1982

Fuse I, Hattori A, Higuchi W: Analysis of platelet shape change, inositol metabolism andCa-mobilization in patients with platelet dysfunction. Rinsho-Byori 40:287–296, 1992

Gerritsen SM, Akkerman J-WN, Nijmejer B, Sixma JJ, Witkop CJ, White J: TheHermansky Pudlak syndrome. Evidence for a lowered 5-hydroxytryptamine contentin platelets of heterozygotes. Scand J Haematol 18:249–256, 1977

Harada T, Kagamiyama H, Hatakeyama K: Feedback regulatory mechanisms for the controlof GTP cyclohydrolase I activity. Science 260:1507–1510, 1993

Hardisty RM, Hutton RA: Bleeding tendency associated with ‘‘new’’ abnormality ofplatelet behaviour. Lancet 1:983–985, 1967

Hardisty RM, Mills DCB, Ketsa-Ard K: The platelet defect associated with albinism. BrJ Haematol 23:579–592, 1972

Hermansky F, Pudlak P: Albinism associated with hemorrhagic diathesis and unusualpigmented reticular cells in the bone marrow: Report of two cases with histochemicalstudies. Blood 14:162–169, 1959

Kalb VF Jr, Bernlohr RW: A rapid method for protein determination. Anal Biochem82:362–371, 1977

Katch S, Sueoka T, Yamamoto Y, Tahashi S: Phosphorylation by Ca21/calmodulin-dependent protein kinase II and protein kinase C of sepiapterin reductase, theterminal enzyme in the biosynthetic pathway of tetrahydrobiopterin. FEBS Lett341:227–232, 1994

Katoh S, Fujimoto K, Inaba A, et al: Ca21-dependent phosphorylation and proteolysis oftetrahydrobiopterin generating enzymes. In: Pfleiderer W, Rokos H (eds). TheChemistry and Biology of Pteridines and Folates. Oxford, Blackwell Science, 1997,pp. 635–638

Kerler F, Hultner L, Ziegler I, Katzenmeir G, Bacher A: Analysis of the tetrahydrobiopterinsynthesizing system during maturation of murine reticulocytes. J Cell Physiol 142:268–271, 1990

King RA, Olds DP: Hairbulb tyrosinase activity in oculocutaneous albinism: suggestionsfor pathway control and block location. Am J Med Gen 20:49–66, 1985

Kugelman TPV, Scott EJ: Tyrosinase activity in melanocytes of human albinos. J InvestDermatol 37:73–76, 1961

516 SCHALLREUTER ET AL THE JOURNAL OF INVESTIGATIVE DERMATOLOGY

Logan LJ, Rapaport SI, Maher BS: Albinism and abnormal platelet function. N Engl J Med284:1340–1345, 1971

Maurer HM, Wolff JA, Buckingam S, Spielvogel AR: ‘‘Impotent’’ platelets in albinos withprolonged bleeding times. Blood 39:490–499, 1972

Nagayama R, Hattori A, Fuse I, Takeshige T, Takizawa S, Shibata A: Platelet ionizedcalcium mobilisation (Aequorin method) in patients with primary platelet dysfunction.Throm Haem 58:528, 1987

Nichol CA, Smith GK, Duch DS: Biosynthesis and metabolism of tetrahydrobiopterinand molybdopterin. Ann Rev Biochem 54:729–764, 1985

Oh J, Baillin TU, Fukai K, et al: Positional cloning of a gene for Hermansky Pudlaksyndrome, a disorder of cytoplasmic organelles. Nature Genet 14:300–306, 1996

Pittelkow MR, Shipley JD: Serum-free culture of normal human melanocytes: growthkinetics and growth factor requirements. J Cell Physiol 140:565–576, 1989

Prota G: Melanins and Melanogenesis. New York, Academic Press, 1992Schallreuter KU, Pittelkow MR: The regulation of thioredoxin reductase by calcium in

Hermansky–Pudlak syndrome. Arch Dermatol Res 281:40–44, 1989Schallreuter KU, Witkop CJ Jr: Thioredoxin reductase activity in Hermansky Pudlak

syndrome: A method for identification of putative heterozygotes. J Invest Dermatol90:372–377, 1988

Schallreuter KU, Wood JM: Thioredoxin reductase in control of the pigmentary system.In: Mackie R (ed.). Clinics in Dermatology Disorders of Pigmentation, Vol. 7. Philadelphia,J.B. Lippincott, 1989a, pp. 92–105

Schallreuter KU, Wood JM: Calcium regulates thioredoxin reductase in human metastaticmelanoma. Biochim Biophys Acta 997:242–247, 1989b

Schallreuter KU, Frenk E, Wolfe LS, Witkop CJ, Wood JM: Hermansky Pudlak syndromein a Swiss population. Dermatology 187:248–256, 1993

Schallreuter KU, Wood JM, Pittelkow MR, et al: Regulation of melanin biosynthesis inthe human epidermis by tetrahydrobiopterin. Science 263:1444–1446, 1994a

Schallreuter KU, Wood JM, Ziegler I, Lemke KR, Pittelkow MR, Lindsey NJ, Gutlich M:Defective tetrahydrobiopterin and catecholamine biosynthesis in the depigmentationdisorder vitiligo. Biochim Biophys Acta 1226:181–192, 1994b

Schallreuter KU, Buttner G, Pittelkow MR, Wood JM, Swanson NN, Korner C: Thecytotoxicity of 6-biopterin to human melanocytes. Biochem Biophys Res Communs204:43–48, 1994c

Schallreuter KU, Schulz-Douglas V, Bunz A, Beazley W, Korner C: Pteridines in thecontrol of pigmentation. J Invest Dermatol 109:31–35, 1997

Schallreuter KU, Zschiesche M, Moore J, et al: In vivo evidence for compromisedphenylalanine metabolism in vitiligo. Biochem Biophys Res Commun 243:395–399,1988.

Thony B, Leimbacher W, Scherer-Oppliger T, Heizmann CW, Blau N: Molecular analysisof tetrahydrobiopterin deficiency due to 6-pyruvoyl tetrahydropterin synthasemutations: 2-D gel analysis of the in vivo phosphorylation state. In: Pfleiderer W,Rokos H (eds). The Chemistry and Biology of Pteridines and Folates. Oxford, BlackwellScience, 1997, pp. 713–718

Weiss HJ, Tschopp TB, Rogers J, Brand H: Studies of platelet 5-hydroxytryptamine(serotonin) in storage pool disease and albinism. J Clin Invest 54:421–432, 1974

Wille JJ, Pittelkow MR, Shipley GD, Scott RE: Integrated control of growth anddifferentiation of normal human prokeratinocytes cultured in serum-free medium:Clonal analyses, growth kinetics and cell cycle studies. J Cell Physiol 121:31–44, 1984

Witkop Jr CJ, White RG, King RA: Oculocutaneous albinism. In: Nyhan WL (ed.).Heritable Disorders of Amino Acid Metabolism. New York, John Wiley, 1974, p. 177

Witkop CJ, Wolfe LS, Stanley X, White JG, Townsend D, Keenan KM: Elevated urinarydolichol excretion in the Hermansky Pudlak syndrome: Indicator of lysosomaldysfunction. Am J Med 82:463–470, 1987

Witkop CJ, Nunez Babcock M, Rao GH, et al: Albinism and Hermansky-Pudlak syndromein Puerto Rico. Bol Assoc Med PR 82:333–339, 1990

Wolfe LS, Ng Ying Kin NMK, Baker RR: Batten disease and related disorders: Newfindings on the chemistry of the storage material. In: Callahan JW, Lowden JA (eds).Lysosomes and Lysosomal Storage Disease. New York, Raven Press, 1981, pp. 315–330

Ziegler I, Hultner L: Tetrahydro-6-biopterin is associated with tetrahydro-7-biopterin inprimary murine mast cells. FEBS Lett 307:147–150, 1992