Defective Pigment Granule Biogenesis and Aberrant Behavior Caused by Mutations in the Drosophila AP-3ß Adaptin Gene ruby

Corresponding author: Gert O. Pflugfelder, Lehrstuhl für Genetik, Theodor-Boveri-Institut, Biozentrum, Universität Würzburg, Am Hubland, D 97074 Würzburg, Germany., pflugfel{at}biozentrum.uni-wuerzburg.de (E-mail)

Communicating editor: T. SCHÜPBACH

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Lysosomal protein trafficking is a fundamental process conserved from yeast to humans. This conservation extends to lysosome-like organelles such as mammalian melanosomes and insect eye pigment granules. Recently, eye and coat color mutations in mouse (mocha and pearl) and Drosophila (garnet and carmine) were shown to affect subunits of the heterotetrameric adaptor protein complex AP-3 involved in vesicle trafficking. Here we demonstrate that the Drosophila eye color mutant ruby is defective in the AP-3ß subunit gene. ruby expression was found in retinal pigment and photoreceptor cells and in the developing central nervous system. ruby mutations lead to a decreased number and altered size of pigment granules in various cell types in and adjacent to the retina. Humans with lesions in the related AP-3ßA gene suffer from Hermansky-Pudlak syndrome, which is caused by defects in a number of lysosome-related organelles. Hermansky-Pudlak patients have a reduced skin pigmentation and suffer from internal bleeding, pulmonary fibrosis, and visual system malfunction. The Drosophila AP-3ß adaptin also appears to be involved in processes other than eye pigment granule biogenesis because all ruby allele combinations tested exhibited defective behavior in a visual fixation paradigm.

TRAFFICKING of integral membrane proteins within the eukaryotic cell occurs through transport vesicles that form by the assembly of coat proteins onto the cytoplasmic face of precursor membranes. The coat is required for budding of the vesicle and for cargo selection (WATERS et al. 1991 ; SCHEKMAN and ORCI 1996 ; ROBINSON 1997 ). Clathrin, the first coat to be identified, assembles with adaptor protein complexes that provide the link between the membrane and the clathrin scaffold. The adaptor protein complexes provide membrane specificity for clathrin assembly by recognizing sorting motifs in the cytoplasmic domains of membrane proteins and by interacting with putative docking molecules (reviewed in HEILKER et al. 1999 ). Originally, two homologous adaptor protein complexes were identified: AP-1 for sorting at the trans-Golgi network (TGN) during biosynthetic transport to the endosomal/lysosomal system and AP-2 for endocytosis at the plasma membrane. More recently, two further complexes, AP-3 (ODORIZZI et al. 1998 ) and AP-4 (DELL'ANGELICA et al. 1999B ) were characterized. For AP-3, the question of clathrin coat association is controversial (DELL'ANGELICA et al. 1998 ). AP-4 was not found in association with clathrin-coated vesicles (DELL'ANGELICA et al. 1999B ). All AP complexes are heterotetrameric proteins built from homologous subunits. They consist of two large chains (, , , and , and ß1-ß4, respectively, of 90–130 kD), one medium chain (µ1–µ4 of ~50 kD), and one small chain (1–4 of ~20 kD). The interaction of AP-1 and AP-2 with clathrin is mediated by the ß subunits (SCHRODER and UNGEWICKELL 1991 ; SHIH et al. 1995 ).

Evidence for the function of the "classical" complexes AP-1 and AP-2 was predominantly gained from biochemical experiments. In the case of AP-3, genetics is providing information on its role in an apparently conserved transport pathway to lysosome-like organelles (ODORIZZI et al. 1998 ). In Saccharomyces cerevisiae, AP-3 subunits were identified as factors involved in the transport of alkaline phosphatase from the TGN to the vacuole. This pathway is independent of the function of AP-1 and AP-2 or clathrin (COWLES et al. 1997 ). AP-3 in mouse and humans consists of neuronal and non-neuronal isoforms. The latter complex consists of ß3A and µ3A and the ubiquitous , 3A, or 3B subunits (DELL'ANGELICA et al. 1997A , DELL'ANGELICA et al. 1997B ; SIMPSON et al. 1997 ). In neuronal cells, AP-3 contains ß3B, µ3B, and the ubiquitous subunits (NEWMAN et al. 1995 ; PEVSNER et al. 1995 ; SIMPSON et al. 1996 ). Mutations in the two large mouse AP-3 subunit genes, AP-3 in mocha (KANTHETI et al. 1998 ) and AP-3ßA in pearl (FENG et al. 1999 ), were recently identified as the cause of a related set of syndromes.

Conditions affecting human AP-3 subunits are of clinical relevance. A patient with autoimmunity against AP-3ßB suffered from degeneration of the cerebellum, where the protein is prominently expressed (NEWMAN et al. 1995 ). Mutations in AP-3ßA were recognized in a subset of patients suffering from the autosomal recessive disease Hermansky-Pudlak syndrome (HPS, Online Mendelian Inheritance in Man no. 203300) (DELL'ANGELICA et al. 1999A ). HPS is characterized by defects in a number of lysosome-related organelles. Patients exhibit subnormal eye and skin pigmentation due to aberrant melanosome development. Associated with this oculocutaneous albinism, patients suffer from impairment in visual acuity, misrouting of optic fibers, nystagmus, and underdeveloped fovea (SUMMERS et al. 1988 ). Clinically more serious is a deficiency of platelet dense granules and a lysosomal malfunction in reticuloendothelial cells, which generally leads to early death. HPS is genetically heterogeneous (HAZELWOOD et al. 1997 ). In addition to patients affected in the human pearl homolog, patients with mutations in the homolog of the mouse pale ear gene have been diagnosed with HPS (OH et al. 1996 ; GARDNER et al. 1997 ). A phenotype similar to HPS was described in 14 mouse mutants, all originally isolated by their eye/skin pigmentation defect (ODORIZZI et al. 1998 ; SWANK et al. 1998 ; SPRITZ 1999 ).

Pigmentation mutants of Drosophila offer another approach to the study of lysosome and AP-3 function (LLOYD et al. 1998 ). The eye color genes garnet and carmine were recently identified as the Drosophila homologs of AP-3 and AP-3µ, respectively (OOI et al. 1997 ; SIMPSON et al. 1997 ; MULLINS et al. 1999 ). We report here that the Drosophila AP-3ß homolog is encoded by the eye color gene ruby (rb). ruby flies, first described by Bridges in 1914 (MORGAN et al. 1925 ), show a reduced number of eye pigment granules. The expression of rb throughout ontogenesis and behavioral aberrations of adult flies indicate that the gene plays a role beyond eye pigmentation. ruby therefore can provide a model for studying the role of AP-3-mediated vesicular trafficking in development and in adult central nervous system function.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Fly stocks and genetic screens:
rb^P2 was generated in this study by -irradiation. Df(1)ovo^D1rG7 was obtained from Dr. A. P. Mahowald (Chicago), rb¹⁰, Df(1)rb¹, Df(1)rb⁴⁶, In(1)rb^D1, and Dp(1;2)A1125 from Dr. W. L. Pak (Purdue University), and rb^S1 and bi rb cx from the Mid-American Drosophila Stock Center (Bowling Green, OH). The cytological breakpoints of the deficiencies are listed in PFLUGFELDER et al. 1990 . Dp(1;2)A1125, also known as Tp(1;2)rb^+72g (LINDSLEY and ZIMM 1992 ), covers the chromosomal region from 3F3 to 5E8. rb¹⁰ contains a second site lethality not covered by Dp(1;2)A1125 (POECK 1992 ).

Molecular methods:
Southern and Northern blot analysis, cloning, and sequencing were performed as described (PFLUGFELDER et al. 1990 ). cDNAs were isolated from a head cDNA gt11 library (ZINSMAIER et al. 1990 ). More recent sequence analysis employed an ABI Prism Dye terminator cycle sequencing kit (Perkin-Elmer, Norwalk, CT) and an automated ABI 373 sequencer.

In situ hybridization:
Frozen sections of pupal and adult tissue were fixed and hybridized as described in POECK et al. 1993 . For whole mounts, adult brains were dissected on ice in phosphate-buffered saline (PBS, pH 7.5) and fixed in 4% paraformaldehyde/PBS overnight at 4°. Hybridization was carried out as above in a volume of 50 µl. Whole mounts were stained for 8–16 hr, dehydrated in an ethanol series, and embedded in Epon for 1-µm serial sections.

Tissue sections for light and electron microscopy:
Adult heads were prepared for light and electron microscopy as described in RENFRANZ and BENZER 1989 . For light microscopy, 1-µm serial sections were cut and stained with 1% toluidine blue in 1% Borax. Ultrathin Epon plastic sections were poststained with 2% uranyl acetate, followed by Reynolds' lead citrate, and stabilized for transmission electron microscopy by carbon coating. Examination was done with a Philips 201C electron microscope.

Eye pigment spectroscopy:
Eye pigments were extracted using a scaled-down variant of the method of REAL et al. 1985 . For the determination of red pigments (drosopterins), 15 longitudinally split heads of male flies were used. For the extraction of brown pigments (xanthommatins), 100 male heads were homogenized. Spectra were measured in a Beckman (Fullerton, CA) DU40 spectrophotometer.

Behavioral analysis:
Optomotor response of fixed flies walking on a styrofoam ball, which was floating on an air jet, was determined as in BUCHNER 1976 (24° pattern wave length, 1 Hz pattern frequency, light intensity 50 lux). The landing response was tested in suspended flight with a spiral stimulus (BRAITENBERG and TADDEI-FERRETTI 1966 ). Fixation in the walking mode was measured by releasing wing-clipped flies into the middle of an illuminated arena (29 cm diameter) that contained an upright and an inverted "T" at opposite sides of the vertical wall (line width 10°, height from bottom 19°, light intensity 3000 lux). The angle of initial approach to these visual stimuli was measured (WEHNER 1972 ). In Buridan's paradigm (GOTZ 1980 ), individual flies walked spontaneously between two opposing inaccessible landmarks (vertical black stripes on an otherwise uniformly white surround) on an elevated circular disk of 85 mm diameter surrounded by a water-filled moat. Wings were shortened to about one-third of their normal length under cold anesthesia (4°). At least 4 hr of postoperative recovery time with access to water were given prior to testing. Fly positions were recorded at 5 Hz sampling rate with a video scanning device (STRAUSS and PICHLER 1998 ).

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Genetic and molecular characterization of the ruby gene:
ruby was mapped to cytological position 4C6 proximal to the gene optomotor-blind (omb) (BANGA et al. 1986 ; PFLUGFELDER et al. 1990 ; LINDSLEY and ZIMM 1992 ). Essential parts of the ruby gene were revealed by the 8.5-kb overlap of the deficiencies Df(1)rb⁴⁶ and Df(1)ovo^D1rG7 (Fig 1). Flies transheterozygous for these deficiencies were viable and had ruby-colored eyes (POECK 1992 ). Probes from the 8.5-kb interval detected two transcripts of 5.4 and 5.2 kb in length in Northern blots (Fig 2). Both transcripts were confined to a 6-kb interval within the deficiency overlap, indicating the lack of large introns. The smaller transcript was only expressed in adult females and in early embryonic stages and is considered a maternal variant. Both transcripts were eliminated in rb mutant flies trans-heterozygous for Df(1)rb¹ and rb¹⁰ [Df(1)rb¹ eliminates the entire region and rb¹⁰ carries a 7-kb insertion in the rb transcription unit]. This allele combination is, therefore, null for rb function. Transcript size was slightly reduced in rb^P2 flies (data not shown). A 4.2-kb cDNA (pcD4.2) and several smaller cDNAs were isolated from a head cDNA library (ROTH 1991 ). Use of pcD4.2 as a Northern blot probe confirmed the identity of this transcription unit as the rb gene (Fig 2). ruby transcript abundance was relatively uniform throughout development. The apparent modulation of signal intensity in Fig 2A was mainly caused by unequal RNA load (compare Fig 2B). The direction of transcription was determined as proximal to distal by comparative restriction mapping of genomic and cDNA clones.

View larger version (9K): In this window In a new window Download PPT slide   Figure 1. Genetic map of the ruby locus. The breakpoints of Df(1)rb46 and Df(1)ovoD1rG7 that uncover rb are shown. The drawn lines correspond to the remaining DNA. Distal is left. The zero origin of the molecular scale is the distal breakpoint of In(1)ombH31 (PFLUGFELDER et al. 1990 ). Restriction sites for BamHI, EcoRI, HindIII, and SalI are symbolized by B, E, H, and S. Position and size of the 7-kb insertion rb10 and of the small deletion rbP2 are as indicated. For In(1)rbD1, the width of the symbol depicts the interval in which the breakpoint could be localized by Southern analysis. The position of the rb transcription unit is shown below.

View larger version (85K): In this window In a new window Download PPT slide   Figure 2. Northern blot analysis of rb expression. A total of 10 µg of poly(A)+ RNA was loaded in each lane. Developmental stages are indicated above. The genotype is wild-type Berlin except in lane labeled [Df(1)rb1/rb10]. Hybridization probe in A was pcD4.2, in B as loading control a cDNA clone encoding ribosomal protein 49 (O'CONNELL and ROSBASH 1984 ). Arrows indicate the two transcripts of 5.4 and 5.2 kb. For the embryonic RNA preparations (E) the time interval in hours after egg laying is noted. L3 is third larval instar; P1, P3, and P5 are days of pupal development (all at 25°).

Sequencing of pcD4.2 revealed a 1071-amino-acid (aa) open reading frame (ORF) open at the 5' end. A search of the computer database showed strong sequence homology over the entire ORF to the human ß3A and ß3B adaptins and indicated that ~90 aa might still be lacking from the N terminus. Sequencing of genomic DNA upstream of pcD4.2 showed that the ORF extended for another 89 codons, including 6 ATG codons. A 5' expressed sequence tag clone (Berkeley Drosophila Genome Project GH25915) extended our cDNA by 544 bp and confirmed the continuation of the ORF. On the basis of the sequence homology between the ORFs of ruby, ß3A, and ß3B that extended all the way to the first start codon of the ruby ORF and the good fit of the first initiation codon environment to the Cavener consensus (CAVENER 1987 ), we assume that ruby is translated from the first AUG codon to yield a 1160-aa protein (Fig 3). A genomic sequence encompassing the ruby ORF submitted by Zhang and Broadie under accession no. AJ011778 differs in only three (silent) positions from pcD4.2. Our cDNA extends 629 bp beyond the 3' end of the AJ01178 sequence and includes a polyA tail preceded by several potential polyadenylation signals, indicating the presence of a 1029-bp 3' untranslated region. Sequence identity between RUBY and human ß3A and ß3B is 51 and 54%, respectively, and extends over the entire length of the proteins (Fig 3). In the N-terminal two-thirds (trunk domain) sequence identity is 66 and 70%, respectively. At the beginning of the C-terminal domain, the sequence LLDLDD (aa 875–880 in RUBY), which has been identified as a clathrin-binding motif (DELL'ANGELICA et al. 1998 ), is fully conserved. Sequence similarity to the mammalian ß1, ß2, and ß4 proteins is much lower and is obvious only in parts of the N-terminal trunk domain. RUBY is similarly related to the rat ß2 and to the Drosophila ß1/2-type adaptin BAD-1 (CAMIDGE and PEARSE 1994 ) (36% in 572 aa of the N-terminal domain).

View larger version (180K): In this window In a new window Download PPT slide   Figure 3. Predicted amino acid sequence of RUBY (accession no. AF-247194) and the human ß3A and ß3B proteins (accession nos. U81504 and AF022152). Sequence alignment was performed with the program MegAlign from DNAStar; identical amino acids are boxed.

ruby expression in the developing retina and central nervous system:
For the spatial analysis of the rb expression pattern, whole mount and cryostat head sections of stage 13 pupae (BAINBRIDGE and BOWNES 1981 ) and adults were analyzed by in situ hybridization. In the pupal eye/lamina complex, rb was strongly expressed in a distal zone of the retina extending up to the cornea corresponding to the cell body position of the retinal pigment and outer photoreceptor cells (Fig 4A and Fig C). Expression in the photoreceptor cells was also visualized by the staining pattern of tangential eye sections, which in each ommatidium outlines the outer rhabdomeres (Fig 4E). In addition, we found expression in the cortex of the entire pupal brain (Fig 4A). This cortical expression declined with pupal maturation and was no longer found in the adult (Fig 4G). This decreasing cortical expression is not an artifact of low probe accessibilty to the adult brain cortex because probes for a less abundantly transcribed gene yield detectable signal under the same conditions (POECK et al. 1993 ). In the adult visual system, rb expression was found in a layer of cells covering the retina basolaterally and within the retina (Fig 4G). The distal staining ensheathed the pseudocone and showed tapering extensions toward the basement membrane at ommatidial periodicity, indicative of expression in pigment cells (Fig 4H). Plastic sections of whole-mount stained head preparations identified the proximal expression in cells lying directly underneath the basement membrane, known as the fenestrated-layer glia (SAINT MARIE and CARLSON 1983 ) or subretinal glial cells (CAGAN and READY 1989 ) (Fig 4I).

View larger version (122K): In this window In a new window Download PPT slide   Figure 4. ruby expression. Hybridization with sense and antisense RNA probes derived from pcD4.2 was performed on frontal cryostat sections (A–H) and on whole mounts (I). The latter were subsequently Epon-embedded and sectioned. Hybridization in A, C, E, G, H, and I was performed with antisense; in B, D, and F, with sense RNA. Both probes caused a strong background signal over the cornea tissue (compare A and B). (A–F) rb expression in the visual system of stage 13 pupae. A general rb expression was found in the entire brain cortex. The strongest signal was detectable in a layer of cells between the optic lobes and the retina (arrow in A) and in the distal retina (arrowheads in A and C). In the tangential section in E and F, subsequent ommatidial rows are cut at an increasingly proximal level (also evidenced by their decreasing diameter), revealing rb transcripts in the direct vicinity of the rhabdomeres. In the adult visual system, expression is most prominent in the pigment cells ensheathing the ommatidia (arrowhead in H) and around the basement membrane that separates the retina from the underlying brain tissue (G–I). The higher magnification in I shows hybridization predominantly brain-proximal to the basement membrane. Bar in A, 100 µm in A and B; in C–F and H, 50 µm; in G, 90 µm; in I, 30 µm. cb, central brain; ol, optic lobes; re, retina; la, lamina; mc, monopolar cells; fl, fenestrated layer.

We investigated the cell autonomy of rb function by inducing rb mutant eye spots by X-ray irradiating ruby/white larvae (STERN 1936 ). Ruby-colored and white twin spots were generated at comparable frequency and appeared not to differ in size (data not shown). While we could not determine rb genotype independent of ruby phenotype, the experiment, nonetheless, suggests a cell-autonomous requirement of rb function for normal eye pigmentation.

Reduced number of pigment granules and altered retina-brain interface in ruby:
In the adult fly visual system, pigment granules were described in the retinula cells, the cone cells, the primary, secondary, and tertiary pigment cells, and in the subretinal glial cells (CAGAN and READY 1989 ). We have investigated the influence of rb^S1 on pigment granule number at two levels of the retinal/subretinal complex. The eye color of rb^S1 or rb¹ did not differ from that of the presumptive null allele combination Df(1)rb¹/rb¹⁰. rb^S1 therefore can be considered as null or strongly hypomorphic for pigmentation function. Tangential sections through the eye below the level of the primary pigment cells and above the cone cell endfeet revealed an about twofold reduction in the number of pigment granules in secondary and tertiary pigment cells (Fig 5, A–D). At light microscopic resolution, only the large granules of the pigment cells can be recognized (STARK and SAPP 1988 ). Below the basement membrane, the reduction in granule number was more than threefold. The mutant granules appeared to have a larger diameter than wild type. The thickness of the pigment containing the fenestrated layer was reduced by ~40% (Fig 5E and Fig F).

View larger version (118K): In this window In a new window Download PPT slide   Figure 5. ruby influence on pigment granule levels in the retina (wild type in A, C, and E; rbS1 in B, D, and F). (A and B) Tangential sections of Epon-embedded eyes stained with toluidine blue for light microscopic visualization. At the level of sectioning, the pigment granules of the secondary and tertiary pigment cells can be seen to encircle each ommatidium. (C and D) Electron micrographs of ommatidial cross sections. The number of pigment granules is reduced about twofold in rbS1; their size appears to be slightly enlarged. (E and F) Pigment granules in subretinal glial cells. Ultrathin sections from plastic-embedded heads were chosen in which the plain of section lay approximately orthogonal to the plain of the basement membrane (arrows). Similar to the findings in ommatidia, there are fewer but larger pigment granules in the fenestrated layer. The width of the fenestrated layer between the basement membrane and the cellular cortex of the lamina monopolar cells is reduced. In the electron microscopic preparations, fixation and sectioning often led to loss of the black pigment granule content that therefore generally appears as white holes. Bar (in A), 5 µm in A and B; 2.5 µm in C and D; and 1 µm in E and F. re, retina; bm, basement membrane; fl, fenestrated layer; mc, monopolar cells.

Both red (drosopterins) and brown pigments (xanthommatins) were reduced in ruby. In rb¹ we found a reduction in drosopterin content to 25% of the wild-type level. The drosopterin content was similarly reduced in the g¹ allele of garnet (which encodes the subunit of Drosophila AP-3). In the rb¹ g¹ double mutant drosopterin content was reduced to 18%. Xanthommatin content was less strongly affected. The single mutants rb¹ and g¹ showed a uniform reduction to 58%, the double mutant to 41% of the wild-type value (Fig 6).

View larger version (34K): In this window In a new window Download PPT slide   Figure 6. Spectral analyis of pigment content. Drosopterin (A) and xanthommatin (B) were extracted from heads of flies reared under identical conditions. The genotypes are indicated next to the curves. The small deflection in the curves below 380 nm is caused by the ultraviolet to visible wavelengths transition of the spectrophotometer. In B, the curves for rb1 and g1 were nearly superimposable. Only the rb curve is therefore shown. WT, wild type; rb, ruby; g, garnet.

Behavioral defects of ruby mutants:
In rb we observed a reduction in thickness of the fenestrated glia layer. It has been shown in other mutants that glial function is necessary for correct development and maintenance of the fly's visual system (SAINT MARIE and CARLSON 1983 ; XIONG and MONTELL 1995 ). Because the brain structure of ruby mutants at the resolution of 1-µm serial plastic sections appeared normal otherwise, we conducted an analysis of behavior, which is a more sensitive way to detect brain malfunction. Four allelic combinations of rb were tested [In(1)rb^D1/rb^S1, rb¹⁰/rb^S1, Df(1)rb¹/rb^S1, and rb¹⁰/Df(1)rb¹] and responded normally to moving (landing and optomotor response) and stationary visual stimuli (pattern orientation; data not shown). The same mutants showed, however, abnormal behavior in Buridan's paradigm, a test for alternating pattern fixation. Fixation of visual landmarks was significantly impaired in rb mutants. Unlike that in blind flies, this behavior was not random in rb. ruby mutations also caused a highly significant increase in walking speed between the landmarks [17.6 +/- 0.9 (rb) vs. 13.4 +/- 0.5 (control) cm/sec; P < 0.001]. Importantly, introduction of the duplication Dp(1;2)A1125 into the rb mutant background rescued both the eye color and the behavioral defects (Fig 7). The one tested g allele g¹ was similarly disturbed in both orientation behavior and walking speed (data not shown). As will be argued below, the altered eye pigmentation is unlikely to account for the highly specific behavioral phenotype.

View larger version (29K): In this window In a new window Download PPT slide   Figure 7. Behavior analysis in Buridan's paradigm. (A–C) Traces of single flies walking for 15 min between inacessible visual landmarks (symbolized by solid bars flanking the circles). The genotypes are as indicated. In B, the rb phenotype is rescued by the duplication Dp(1;2)A1125. The wild-type trace in C is taken from STRAUSS and PICHLER 1998 . (D) Quantitative evaluation of 12 flies for each genotype. The angle of orientation between fly and approached target was measured every 0.2 sec (4500 recordings per fly). Angles between 0° and 5°, 5° and 10°, etc., were pooled and plotted against the normalized frequency with which they were observed. The error bars are smaller than the symbols. The difference in orientation behavior between the two genotypes is highly significant (t-test, two-tailed, P < 10-4). The curve for random orientation was calculated as in STRAUSS and PICHLER 1998 .

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

We have determined by genetic and molecular means the identity of the Drosophila ruby gene. High sequence homology throughout the protein coding region to the mammalian ß3 adaptins and phenotypic similarity to mammalian ß3A mutants suggests that ruby encodes a functional ß3 adaptin homolog. In humans, two ß3 variants with 61% aa sequence identity were described. The ß1 and ß2 adaptins constitute another pair of related proteins (83% aa sequence identity) (DELL'ANGELICA et al. 1997B ; SIMPSON et al. 1997 ). Sequence conservation between the three human gene groups ß1/2, ß3A/B, and ß4 lies between 21 and 28% aa sequence identity (DELL'ANGELICA et al. 1999B ). RUBY is about equidistant to human ß3A and ß3B (51 and 54%), but is clearly more closely related to these proteins than to ß1/2 or ß4 where sequence similarity is confined to N-terminal regions. Another Drosophila AP-ß protein has previously been described (BAD-1, ß-adaptin-Drosophila; CAMIDGE and PEARSE 1994 ). Its sequence is equidistant to human ß1 and ß2 (73% homology). We found no evidence (from low stringency Southern hybridizations or from the sequence database) for the existence of more ß3-related genes in Drosophila. It appears, therefore, that the evolutionary split into ß1 and ß2 and into ß3A and ß3B took place after the separation of protostomes and deuterostomes and only in the deuterostome lineage. RUBY contains a putative clathrin-binding motif. In Drosophila, several mutations in clathrin heavy chain gene were described (BAZINET et al. 1993 ). The significance of the clathrin-binding motif can now be assessed in transgenic animals expressing mutant versions of the rb gene and by testing for genetic interactions.

ruby is transcribed throughout development. Two transcripts differing only slightly in size were found in young embryos and in adult females. The smaller transcript appears to be a purely maternal form, not found in later embryonic stages or in adult males. Surprisingly, the larger transcript also was found during the first hour of embryonic development before the onset of zygotic transcription. In contrast to garnet where female sterility and to carmine where lethality was reported for some alleles, all rb mutants and allele combinations were fertile. garnet also differs from rb in that strong alleles have an orange eye color (LINDSLEY and ZIMM 1992 ).

If all subunits of a multiprotein complex were equally necessary for its function, the phenotype of mutants in all subunit genes should be the same. This appears to be the case for yeast AP-3 (COWLES et al. 1997 ; PANEK and STEPP 1997 ). The difference in the severity of strong rb, g and cm phenotypes could mean that ß3, , and µ3 differ in the degree with which their function can be substituted by related proteins. There are observations that support this hypothesis. In mammalian AP-3 subunits, a higher constraint on vs. ß3 function is evident from the sequence conservation between mouse and humans: 98% for and 88% for ß3A (KANTHETI et al. 1998 ; FENG et al. 1999 ). Mutation of AP-3 in the mocha mouse leads to a nearly complete loss of the three other subunits of the complex; mutation of AP-3ßA in the pearl mouse has a more subtle effect on the complementary subunits (KANTHETI et al. 1998 ). In transfected Cos cells, the ß1/2-type Drosophila BAD-1 protein can be accommodated in AP-1 and AP-2 complexes without disturbing their targeting to the TGN or plasma membrane (CAMIDGE and PEARSE 1994 ). Similarly, in the mammalian AP-3 complex, one subunit can interact with different isoforms of ß, µ, and (ODORIZZI et al. 1998 ).

Eye pigmentation in Drosophila proceeds through many steps. Low-molecular-weight pigment precursors are synthesized in the fat body and Malpighian tubules and are taken up by the future pigmented cells during pupal development. Precursor transport across the pigment cell membrane is accomplished by heteromeric ABC transporters encoded by white, brown, and scarlet (DREESEN et al. 1988 ; TEARLE et al. 1989 ). In contrast to mutations in genes encoding precursor-synthesizing enzymes, mutations affecting uptake functions or pigment granule biogenesis are autonomous. Autonomy of the ruby eye color phenotype is suggested by our mosaic experiments, by the existence of the variegating allele rb^m48, and by the pigment autonomy of rb eye disks transplanted into wild-type hosts (BEADLE and EPHRUSSI 1936 ; LINDSLEY and ZIMM 1992 ). ruby reduces the level of both types of eye pigments, the drosopterins and ommochromes (NOLTE 1950 , NOLTE 1952 ; our data). This property is shared by many of the more than 85 eye pigmentation mutants known in Drosophila, most of which have not yet been cloned. In addition to rb, the sequence and mutant phenotype of four pigmentation genes currently suggests their involvement in vesicle transport: garnet, deep orange, light, and carnation (LLOYD et al. 1998 ; SEVRIOUKOV et al. 1999 ). carnation was recently shown to be the constituent of a protein complex unrelated to AP but also involved in protein transport to lysosomes and pigment granules (SEVRIOUKOV et al. 1999 ). carnation, carmine, g, and rb were collectively termed "ruby group" based on the similarity of their pigmentation phenotype (NOLTE 1952 ). Carmine, g, and rb have by now been identified as genes encoding AP-3 components. Since carnation does not encode the 3 subunit, mutations in the 3 gene will probably differ in eye phenotype from the members of the "ruby group."

Allele combinations null for rb function still contain pigment granules and have ruby-colored eyes. This can be explained by various hypotheses. As discussed above, Drosophila ß-type adaptin function may be partially redundant. Another possibility is that alternative pathways for pigment granule biogenesis can be activated in rb or are constitutively active at a low level. This is in agreement with observations in yeast, where inactivation of AP-3 only partially eliminates transport of alkaline phosphatase to the vacuole (COWLES et al. 1997 ; STEPP et al. 1997 ). Lysosomes and melanosomes can receive proteins through both the biosynthetic route from the TGN and via endocytosis from the plasma membrane (LLOYD et al. 1998 ). Inactivation of AP-3 by mutations in ß3A or with antisense oligonucleotides leads to an increased expression of lysosomal proteins on the surface of mammalian cells (LEBORGNE and HOFLACK 1998 ; DELL'ANGELICA et al. 1999A ). A similar effect was observed in AP-3 mutants in yeast (COWLES et al. 1997 ; STEPP et al. 1997 ). The coexistence of alternative pathways appears to be a general characteristic of membrane trafficking (MOSTOV et al. 1999 ). Our observation of an increased granule size in rb (Fig 5) may be an indication of an alternative biogenic pathway. Similar observations have previously been made in several Drosophila pigmentation mutants, including rb (NOLTE 1950 ; SHOUP 1966 ; STARK and SAPP 1988 ). Redundancy at the ß-adaptin or pathway level also appears to exist in the mouse where the pearl and mocha mutants have dilute, but not albino, pigmentation (KANTHETI et al. 1998 ). As in Drosophila, altered melanosome size or shape was observed in several mouse pigmentation mutants, including the two HPS models pearl and pale ear (MANGINI et al. 1985 ; GARDNER et al. 1997 ).

In rb, the thickness of the pigmented layer underneath the retinal basement membrane was found to be reduced (Fig 5). Similar findings were described by NOLTE 1950 at the resolution of light microscopy for rb and several other pigment mutants. Since pigment granules appear to take up a considerable fraction of the cellular volume, the volume reduction could be a consequence of the reduction in granule number. There are similarities in the structural organization of the interfaces between the vertebrate photoreceptor cells and the retinal pigment epithelium and between the dipteran fly retina and the optic lobes. In both cases pigmented glia cells extend apical processes in close contact with the photoreceptor cells (SAINT MARIE and CARLSON 1983 ; BOK 1985 ; WILLIAMS et al. 1985 ). The pigment granules/melanosomes in these glial cells appear to be similarly affected in the pearl mouse and the rb fly. They are reduced in number and increased in size (MANGINI et al. 1985 ; WILLIAMS et al. 1985 ).

pearl mice show a defective optokinetic nystagmus, a defect pearl shares (with variation) with several mouse hypopigmentation mutants. This defect appears to correlate with a reduced numerical density of melanosomes in the retinal pigment epithelium. pearl, like many other hypopigmented mammalian mutants, displays a reduced ipsilateral retinal projection, the developmental origins of which are not yet clear (MARCUS et al. 1996 ). We did not observe pathfinding defects in the visual system of rb flies. Indeed, ruby flies responded normally to moving visual stimuli (optomotor and landing response). They were, however, abnormal in alternating pattern fixation in Buridan's paradigm. In a different pattern fixation paradigm, in which only the first target approach was evaluated, rb flies performed normally. In agreement with previous investigations of various eye color mutants (HENGSTENBERG and GOTZ 1967 ), the normal performance of rb mutants in several visual tests indicates that the altered eye pigmentation has little effect on the processing of visual signals in the distal periphery of the visual system. The pattern fixation test in which rb flies performed normally requires a slightly higher visual acuity than Buridan's paradigm, indicating that the behavioral defect in the latter is not caused by an influence of eye pigmentation on visual acuity. This suggests that the development or function of a higher coordination center is disturbed in rb. The two behavioral defects, affecting orientation and walking speed, may map to different foci. Intriguingly, in a mocha mouse mutant an altered neocortical activity was observed (NOEBELS and SIDMAN 1989 ). Our in situ hybridization data showed rb expression in the central nervous system (CNS) cortex during pupal development but not in the adult stage. This indicates a developmental role of rb in the CNS, rather than a function in neuronal activity. An adult neuronal function is fulfilled by an AP-2-related Drosophila adaptor complex that is involved in synaptic vesicle recycling from the presynaptic cell membrane (GONZALEZ-GAITAN and JACKLE 1997 ), possibly in conjunction with another clathrin adaptor protein (ZHANG et al. 1998 , ZHANG et al. 1999 ). In vitro experiments have implied mammalian AP-3 in synaptic vesicle recycling from endosomes (FAUNDEZ et al. 1998 ). Whether AP-3 has a synaptic function in Drosophila can now be studied electrophysiologically in mutants of the identified AP-3 subunit genes.

There is indirect evidence for the developmental function of two other Drosophila eye pigmentation genes in the brain. The combination of the two viable mutations carnation and light leads to synthetic lethality. By mosaic analysis, the focus of this lethal interaction was localized to the brain, which is abnormal in double-mutant larvae (MCCARTHY 1980 ). The high penetrance of the aberrant behavior revealed in Buridan's paradigm will allow an investigation of AP-3's neurobiological function. The spatial and temporal requirements of rb function for normal behavior can be determined using Gal4 (BRAND and PERRIMON 1993 ) and flippase (GOLIC and LINDQUIST 1989 ) technology.

	FOOTNOTES

¹ These authors contributed equally to this work.
² Present address: Institut für Medizinische Genetik, Klinikum Osnabrück, D 49090 Osnabrück, Germany.

	ACKNOWLEDGMENTS

Special thanks are due Rosalind Young, Tina Wanke, and Susanne Meyer for technical assistance and to Seymour Benzer who kindly provided the opportunity to perform some of the experiments in his laboratory. We are grateful to M. Heisenberg and A. Hofbauer for inspecting rb head sections and for constructive criticism. E. Buchner and T. Zars provided valuable suggestions on the manuscript. The work was supported by grants from the Deutsche Forschungsgemeinschaft to G.O.P.

Manuscript received November 12, 1999; Accepted for publication February 8, 2000.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

BAINBRIDGE, S. P. and M. BOWNES, 1981  Staging the metamorphosis of Drosophila melanogaster.. J. Embryol. Exp. Morphol. 66:57-80[Medline].

BANGA, S. S., B. T. BLOOMQUIST, R. K. BRODBERG, Q. N. PYE, and D. C. LARRIVEE et al., 1986  Cytogenetic characterization of the 4BC region on the X chromosome of Drosophila melanogaster: localization of the mei-9, norpA, and omb genes. Chromosoma 93:341-346[Medline].

BAZINET, C., A. L. KATZEN, M. MORGAN, A. P. MAHOWALD, and S. K. LEMMON, 1993  The Drosophila clathrin heavy chain gene: clathrin function is essential in a multicellular organism. Genetics 134:1119-1134[Abstract].

BEADLE, G. W. and B. EPHRUSSI, 1936  The differentiation of eye pigments in Drosophila as studied by transplantation. Genetics 21:225-247[Medline].

BOK, D., 1985  Retinal photoreceptor-pigment epithelium interactions. Investig. Opthalmol. Vis. Sci. 26:1659-1694[Free Full Text].

BRAITENBERG, V. and C. TADDEI-FERRETTI, 1966  Landing reaction of Musca domestica induced by visual stimuli. Naturwissenshaften 53:155[Medline].

BRAND, A. H. and N. PERRIMON, 1993  Targeted gene expression as a means of altering cell fates and generating dominant phenotypes. Development 118:401-415[Abstract].

BUCHNER, E., 1976  Elementary movement detectors in an insect visual system. Biol. Cybernetics 24:85-101.

CAGAN, R. L. and D. F. READY, 1989  The emergence of order in the Drosophila pupal retina. Dev. Biol. 136:346-362[Medline].

CAMIDGE, D. R. and B. M. F. PEARSE, 1994  Cloning of Drosophila ß-adaptin and its localization on expression in mammalian cells. J. Cell Sci. 107:709-718[Abstract].

CAVENER, D. R., 1987  Comparison of the consensus sequence flanking translational start sites in Drosophila and vertebrates. Nucleic Acids Res. 15:1353-1361[Abstract/Free Full Text].

COWLES, C. R., G. ODORIZZI, G. S. PAYNE, and S. D. EMR, 1997  The AP-3 adaptor complex is essential for cargo-selective transport to the yeast vacuole. Cell 91:109-118[Medline].

DELL'ANGELICA, E. C., H. OHNO, C. E. OOI, E. RABINOVITCH, and J. S. BONIFACINO, 1997a  AP-3: an adaptor-like protein complex with ubiquitous expression. EMBO J. 16:917-928[Medline].

DELL'ANGELICA, E. C., C. E. OOI, and J. S. BONIFACINO, 1997b  ß3A-adaptin, a subunit of the adaptor-like complex AP-3. J. Biol. Chem. 272:15078-15084[Abstract/Free Full Text].

DELL'ANGELICA, E. C., J. KLUMPERMAN, W. STOORVOGEL, and J. S. BONIFACINO, 1998  Association of the AP-3 adaptor complex with clathrin. Science 280:431-434[Abstract/Free Full Text].

DELL'ANGELICA, E. C., V. SHOTELERSUK, R. C. AGUILAR, W. A. GAHL, and J. S. BONIFACINO, 1999a  Altered trafficking of lysosomal proteins in Hermansky-Pudlak syndrome due to mutations in the ß3A subunit of the AP-3 adaptor. Mol. Cell 3:11-21[Medline].

DELL'ANGELICA, E. C., C. MULLINS, and J. S. BONIFACINO, 1999b  AP-4, a novel protein complex related to clathrin adaptors. J. Biol. Chem. 12:7278-7285.

DREESEN, T. D., D. H. JOHNSON, and S. HENIKOFF, 1988  The brown protein of Drosophila melanogaster is similar to the white protein and to components of active transport complexes. Mol. Cell. Biol. 8:5206-5215[Abstract/Free Full Text].

FAÚNDEZ, V., J.-T. HORNG, and R. B. KELLY, 1998  A function for the AP3 coat complex in synaptic vesicle formation from endosomes. Cell 93:423-432[Medline].

FENG, L., A. B. SEYMOUR, S. JIANG, A. TO, and A. A. PEDEN et al., 1999  The ß3A subunit gene (Ap3ß1) of the AP-3 adaptor complex is altered in the mouse hypopigmentation mutant pearl, a model for Hermansky-Pudlak syndrome and night blindness. Hum. Mol. Genet. 8:323-330[Abstract/Free Full Text].

GARDNER, J. M., S. C. WILDENBERG, N. M. KEIPER, E. K. NOVAK, and M. E. RUSINIAK et al., 1997  The mouse pale ear (ep) mutation is the homologue of human Hermansky-Pudlak syndrome. Proc. Natl. Acad. Sci. USA 94:9238-9243[Abstract/Free Full Text].

GOLIC, K. G. and S. LINDQUIST, 1989  The FLP recombinase of yeast catalyzes site-specific recombination in the Drosophila genome. Cell 59:499-509[Medline].

GONZÁLEZ-GAITAN, M. and H. JÄCKLE, 1997  Role of Drosophila alpha-adaptin in presynaptic vesicle recycling. Cell 88:767-776[Medline].

GÖTZ, K. G., 1980 Visual guidance in Drosophila, pp. 391–407 in Development and Neurobiology of Drosophila, edited by O. SIDDIQI, P. BABU, L. M. HALL and J. C. HALL. Plenum Press, New York.

HAZELWOOD, S., V. SHOTELERSUK, S. C. WILDENBERG, D. CHEN, and F. IWATA et al., 1997  Evidence for locus heterogeneity in Puerto Ricans with Hermansky-Pudlak syndrome. Am. J. Hum. Genet. 61:1088-1094[Medline].

HEILKER, R., M. SPIESS, and P. CROTTET, 1999  Recognition of sorting signals by clathrin adaptors. Bioessays 21:558-567[Medline].

HENGSTENBERG, R. and K. G. GÖTZ, 1967  Der Einfluss des Schirmpigmentgehalts auf die Helligkeits- und Kontrastwahrnehmung bei Drosophila-Augenmutanten. Kybernetik 3:276-285[Medline].

KANTHETI, P., X. QIAO, M. E. DIAZ, A. A. PEDEN, and G. A. MEYER et al., 1998  Mutation in AP-3 in the mocha mouse links endosomal transport to storage deficiency in platelets, melanosomes, and synaptic vesicles. Neuron 21:111-122[Medline].

LEBORGNE, R. and B. HOFLACK, 1998  Mechanisms of protein sorting and coat assembly: insight from the clathrin-coated vesicle pathway. Curr. Opin. Cell Biol. 10:499-503[Medline].

LINDSLEY, D. L., and G. G. ZIMM, 1992 The Genome of Drosophila melanogaster. Academic Press, Inc., San Diego.

LLOYD, V., M. RAMASWAMI, and H. KRÄMER, 1998  Not just pretty eyes: Drosophila eye-colour and lysosomal delivery. Trends Cell Biol. 8:257-259[Medline].

MANGINI, N. J., J. W. VANABLE, JR., L. A. WILLIAMS, and L. H. PINTO, 1985  The optokinetic nystagmus and ocular pigmentation of hypopigmented mouse mutants. J. Comp. Neurol. 8:191-209.

MARCUS, R. C., L.-C. WANG, and C. A. MASON, 1996  Retinal axon divergence in the optic chiasm: midline cells are unaffected by the albino mutation. Development 122:859-868[Abstract].

MCCARTHY, A., 1980  Morphology of the carnation-light synthetic lethal focus in Drosophila melanogaster.. Experientia 36:1361-1362[Medline].

MORGAN, T. H., S. B. BRIDGES, and A. H. STURTEVANT, 1925  The genetics of Drosophila. Bibliogr. Genet. 2:1-262.

MOSTOV, K., M. B. A. TER BEEST, and S. J. CHAPIN, 1999  Catch the µ1B train to the basolateral surface. Cell 99:121-122[Medline].

MULLINS, C., L. M. HARTNELL, D. A. WASSARMAN, and J. S. BONIFACINO, 1999  Defective expression of the µ3 subunit of the AP-3 adaptor complex in the Drosophila pigmentation mutant carmine.. Mol. Gen. Genet. 262:401-412[Medline].

NEWMAN, L. S., M. O. MCKEEVER, H. J. OKANO, and R. B. DARNELL, 1995  ß-NAP, a cerebellar degeneration antigen, is neuron-specific vesicle coat protein. Cell 82:773-783[Medline].

NOEBELS, J. L. and R. L. SIDMAN, 1989  Persistent hypersynchronization of neocortical neurons in the mocha mutant of mouse. J. Neurogenet. 6:53-56[Medline].

NOLTE, D. J., 1950  The eye-pigmentary system of Drosophila: the pigment cells. J. Genet. 50:79-99.

NOLTE, D. J., 1952  The eye-pigmentary system of Drosophila III. The action of eye-colour genes. J. Genet. 51:142-186.

O'CONNELL, P. and M. ROSBASH, 1984  Sequence, structure, and codon preference of the Drosophila ribosomal protein 49 gene. Nucleic Acids Res. 12:5495-5513[Abstract/Free Full Text].

ODORIZZI, G., C. R. COWLES, and S. D. EMR, 1998  The AP-3 complex: a coat of many colours. Trends Cell Biol. 8:282-288[Medline].

OH, J., T. BAILIN, K. FUJAI, G. H. FENG, and L. HO et al., 1996  Positional cloning of a gene for Hermansky-Pudlak syndrome, a disorder of cytoplasmic organelles. Nature Genetics 14:300-306[Medline].

OOI, C. E., J. E. MOREIRA, E. C. DELL'ANGELICA, G. POY, and D. A. WASSARMAN et al., 1997  Altered expression of a novel adaptin leads to defective pigment granule biogenesis in the Drosophila eye color mutant garnet.. EMBO J. 16:4508-4518[Medline].

PANEK, H. R. and J. D. STEPP, 1997  Suppressors of YCK-encoded yeast casein kinase 1 deficiency define the four subunits of a novel clathrin AP-like complex. EMBO J. 16:4194-4204[Medline].

PEVSNER, J., W. VOLKNANDT, B. R. WONG, and R. H. SCHELLER, 1995  Two rat homologs of clathrin-associated adaptor proteins. Gene 146:279-283.

PFLUGFELDER, G. O., H. SCHWARZ, H. ROTH, B. POECK, and A. SIGL et al., 1990  Genetic and molecular characterization of the optomotor-blind gene locus in Drosophila melanogaster.. Genetics 126:91-104[Abstract].

POECK, B., 1992 Untersuchungen zur Genetik und Expression im optomotor-blind (omb) Genlokus von Drosophila melanogaster. Ph.D. Thesis. Universität Würzburg, Würzburg, Germany.

POECK, B., A. HOFBAUER, and G. O. PFLUGFELDER, 1993  Expression of the Drosophila optomotor-blind gene transcript in neuronal and glial cells of the developing nervous system. Development 117:1017-1029[Abstract].

REAL, M. D., J. FERRÉ, and J. L. MÉNSUA, 1985  Methods for the quantitative estimation of the red and brown pigments of Drosophila melanogaster.. Dros. Inf. Serv. 61:198-199.

RENFRANZ, P. J. and S. BENZER, 1989  Monoclonal antibody probes discriminate early and late mutant defects in development of the Drosophila retina. Dev. Biol. 136:411-429[Medline].

ROBINSON, M. S., 1997  Coats and vesicle budding. Trends Cell Biol. 7:99-102[Medline].

ROTH, H., 1991 Molekulare Analyse des omb Gen Locus von Drosophila melanogaster. Ph.D. Thesis, Universität Würzburg, Würzburg, Germany.

SAINT MARIE, S. L. and S. D. CARLSON, 1983  The fine structure of neuroglia in the lamina ganglionaris of the housefly, Musca domestica L. J. Neurocytol. 12:213-241[Medline].

SCHEKMAN, R. and L. ORCI, 1996  Coat proteins and vesicle budding. Science 271:1526-1533[Abstract].

SCHRÖDER, S. and E. UNGEWICKELL, 1991  Subunit interaction and function of clathrin-coated vesicle adaptors from the Golgi and the plasma membrane. J. Biol. Chem. 266:7910-7918[Abstract/Free Full Text].

SEVRIOUKOV, E. A., J.-P. HE, N. MOGHRABI, A. SUNIO, and H. KRÄMER, 1999  A role for deep orange and carnation eye color genes in lysosomal delivery. Mol. Cell 4:479-486[Medline].

SHIH, W., A. GALLUSSER, and T. KIRCHHAUSEN, 1995  A clathrin-binding site in the hinge of the ß-2 chain of mammalian AP-2 complexes. J. Biol. Chem. 270:31083-31090[Abstract/Free Full Text].

SHOUP, J. R., 1966  The development of pigment granules in the eyes of wild type and mutant Drosophila melanogaster.. J. Cell Biol. 29:223-249[Abstract/Free Full Text].

SIMPSON, F., N. A. BRIGHT, M. A. WEST, L. S. NEWMAN, and R. B. DARNELL et al., 1996  A novel adaptor-related protein complex. J. Cell Biol. 133:749-760[Abstract/Free Full Text].

SIMPSON, F., A. A. PEDEN, L. CHRISTOPOLOU, and M. S. ROBINSON, 1997  Characterization of the adaptor-related protein complex, AP-3. J. Cell Biol. 137:835-845[Abstract/Free Full Text].

SPRITZ, R. A., 1999  Multi-organellar disorders of pigmentation. Intracellular traffic jams in mammals, flies and yeast. Trends Genet. 15:337-340[Medline].

STARK, W. S. and R. SAPP, 1988  Eye color pigment granules in wild-type and mutant Drosophila melanogaster.. Can. J. Zool. 66:1301-1308.

STEPP, J. D., K. HUNAG, and S. K. LEMMON, 1997  The yeast adaptor protein complex, AP-3, is essential for the efficient delivery of alkaline phosphatase by the alternate pathway to the vacuole. J. Cell Biol. 139:1761-1774[Abstract/Free Full Text].

STERN, C., 1936  Somatic crossing over and segregation in Drosophila melanogaster. Genetics 21:625-730[Free Full Text].

STRAUSS, R. and J. PICHLER, 1998  Persistence of orientation toward a temporarily invisible landmark in Drosophila melanogaster.. J. Comp. Physiol. A 182:411-423[Medline].

SUMMERS, S. G., W. H. KNOBLOCH, C. J. WITKOP, JR., and R. A. KING, 1988  Hermansky-Pudlak syndrome: ophthalmic findings. Ophthalmology 95:545-554[Medline].

SWANK, R. T., E. K. NOVAK, M. P. MCGARRY, M. E. RUSINIAK, and L. FENG, 1998  Mouse models of Hermansky Pudlak syndrome: a review. Pigment Cell Res. 11:60-80[Medline].

TEARLE, R. G., J. M. BELOTE, M. MCKEOWN, B. S. BAKER, and A. J. HOWELLS, 1989  Cloning and characterization of the scarlet gene of Drosophila melanogaster.. Genetics 122:595-606[Abstract/Free Full Text].

WATERS, M. G., T. SERAFINI, and J. E. ROTHMAN, 1991  "Coatomer": a cytosolic protein complex containing subunits of non-clathrin-coated Golgi transport vesicles. Nature 349:248-251[Medline].

WEHNER, R., 1972  Spontaneous pattern preferences of Drosophila melanogaster to black areas in various parts of the visual field. J. Insect Physiol. 18:1250-1268.

WILLIAMS, L. A., L. H. PINTO, and J. GHERSON, 1985  The retinal pigment epithelium of wild type (C57BL/6J +/+) and pearl mutant (C57BL/6J pe/pe) mice. Investig. Ophthalmol. Vis. Sci. 26:657-669[Abstract/Free Full Text].

XIONG, W.-C. and C. MONTELL, 1995  Defective glia induce neural apoptosis in the repo visual system of Drosophila.. Neuron 14:581-590[Medline].

ZHANG, B., Y. H. KOH, R. B. BECKSTEAD, V. BUDNIK, and B. GANETZKY et al., 1998  Synaptic vesicle size and number are regulated by a clathrin adaptor protein required for endocytosis. Neuron 21:1465-1475[Medline].

Z