Genetic Analysis of hook, a Gene Required for Endocytic Trafficking in Drosophila

Corresponding author: Helmut Krämer, Center for Basic Neuroscience, University of Texas Southwestern Medical Center, Dallas, TX 75235-9111., kramer{at}utsw.swmed.edu (E-mail)

Communicating editor: T. SCHÜPBACH

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The Drosophila hook gene encodes a novel component of the endocytic compartment. Previously identified hook alleles, which still expressed truncated Hook proteins, affected the accumulation of internalized transmembrane ligands into multivesicular bodies (MVBs). To determine the hook null phenotype, we isolated nine new hook alleles on the basis of their characteristic hooked-bristle phenotype. At least one of these alleles, hk¹¹, is a complete loss-of-function allele. Flies carrying the hk¹¹ allele are viable and fertile but neither transmembrane ligands nor soluble ligands accumulate in MVBs. This effect on endocytosed ligands can be mimicked by the expression of Hook proteins truncated for the N- and C-terminal domains flanking the central coiled-coil region. The importance of all three domains for Hook function was confirmed by their conservation between two Drosophila and two human Hook proteins.

ENDOCYTOSIS is a ubiquitous process by which cells take up nutrients and signaling molecules. Ultrastructural markers in cultured cells have facilitated the detailed description of the pathway through the endocytic compartment to lysosomes (GRUENBERG and MAXFIELD 1995 ). The classic endocytosis pathway of transmembrane proteins and their ligands begins with their accumulation in coated pits, whose invagination is mediated by a clathrin coat. After invagination, coated pits pinch off the plasma membrane through a process that requires the dynamin GTPase (KOSAKA and IKEDA 1983 ; ROBINSON 1994 ). The resulting internalized vesicles uncoat and fuse to the early endosomal compartment. In early endosomes, internalized proteins can take two routes: they may be sequestered in the recycling compartment of the early endosome and transported back to the plasma membrane, or they may be concentrated in the vacuolar compartment of sorting endosomes (MUKHERJEE et al. 1997 ). After accumulation of additional small vesicles, these vacuolar structures bud from the early endosome and form mature multivesicular bodies (MVBs), which contain many internal vesicles. Finally, these mature MVBs fuse directly to multilamellar late endosomes or lysosomes (FUTTER et al. 1996 ).

An unusual mode of endocytosis has been described for the Boss transmembrane ligand. The Boss protein is expressed on the surface of R8 photoreceptor cells in the developing compound eye of Drosophila (KRAMER et al. 1991 ; ZIPURSKY and RUBIN 1994 ). Upon binding to the Sevenless receptor, the entire Boss transmembrane protein is internalized across cell boundaries from the surface of R8 cells into R7 cells. In R7 cells, Boss accumulates in MVBs (KRAMER et al. 1991 ; CAGAN et al. 1992 ). The specific expression of the Boss ligand in R8 cells and its internalization into the neighboring R7 cells provide an immunocytochemical assay for mutations that interfere with endocytosis in Drosophila (KRAMER et al. 1991 ). Using this Boss endocytosis assay, we previously identified hook as a gene affecting endocytic trafficking in Drosophila (KRAMER and PHISTRY 1996 ).

The hook gene encodes a 679-amino-acid (aa) cytoplasmic protein whose only predicted structural feature is an extended central coiled-coil domain. Two findings indicated that the Hook protein is a novel component of the endocytotic compartment: (i) its requirement for proper accumulation of Boss in MVBs and (ii) its localization to vesicular structures that were identified as components of the endocytotic compartment by internalized markers such as dextran particles or ligands of the Sevenless receptor (KRAMER and PHISTRY 1996 ). All previously identified alleles of hook still expressed truncated Hook proteins, which included the central coiled-coil dimerization domain. In this article, we genetically analyze the function of the hook gene by isolating a complete loss-of-function allele and by expressing truncated Hook proteins.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Fly stocks and genetic screens:
The hook alleles hk¹, hk^C1, hk⁴⁹², and Df(2L)TW130 have previously been described (MOHR 1927 ; STATHAKIS et al. 1995 ; KRAMER and PHISTRY 1996 ). Other fly stocks were obtained from the Bloomington stock center and have been described (LINDSLEY and ZIMM 1992 ). To obtain new alleles of the hook gene, cn bw males were mutagenized with -rays (4000 rad), EMS (25 mM), or ENU (3 mM) as previously described (ASHBURNER 1989 ). Mutagenized males were mated to hk¹ virgins and progeny were raised at 25°. Male progeny that exhibited the characteristic hooked appearance of bristles (MOHR 1927 ; KRAMER and PHISTRY 1996 ) were crossed to a CyO balancer. Viability of hook alleles was tested in trans to Df(2L)TW130, a deficiency that completely deletes the hook genomic region (STATHAKIS et al. 1995 ). A total of 20,000 progeny from EMS-mutagenized males yielded five new hook alleles, 15,000 progeny from -ray-mutagenized males yielded two new hook alleles, and 10,000 progeny from ENU-mutagenized males yielded three new hook alleles (Table 1). Altered sequences of mutant hook alleles were determined by direct sequencing of PCR products generated from flies that carried the indicated allele over the Df(2L)TW130.

Transgenic flies:
We previously described constructs that encode an HA-epitope-tagged full-length Hook protein, HA2-679Hook, an HA-epitope-tagged N-terminally truncated Hook protein, HA248-679Hook, and a Myc-epitope-tagged C-terminally truncated Hook protein, Hook1-551Myc (KRAMER and PHISTRY 1996 ). To express the truncated proteins in R7 cells at the time of Boss endocytosis, these constructs were cloned in the transformation vector pBD365 in which their expression is controlled by a duplicated sevenless enhancer (BASLER et al. 1991 ), as described for the Hook full-length protein (KRAMER and PHISTRY 1996 ). To test the effect of these transgenes on Boss endocytosis, homozygous or balanced lines were tested by antibody staining as described below. To test the effect of truncated Hook proteins in the adult eye, we cloned the Hook1-551Myc construct into the pGMR transformation vector (HAY et al. 1994 ). Transgenic flies were established using standard techniques (RUBIN and SPRADLING 1982 ).

Cloning the hook gene from Drosophila virilis:
DNA manipulations were performed using standard procedures (AUSUBEL et al. 1994 ). Among several pairs of degenerate oligonucleotides attempted, only the pair degRD2 (CATYTTIGGYTTNGGYTCYTCYTCNAC) and degFD2 (GAYGAYGCIAAYAARMGITGYG) resulted in the amplification of a fragment of the hook gene from D. virilis genomic DNA. This 300-bp fragment was used to isolate a clone containing the hook gene from a D. virilis genomic library (kindly provided by John W. Tamkun, University of California at Santa Cruz). The sequence of the D. virilis protein was deduced from the genomic sequence by comparison to the D. melanogaster protein; this was aided by the strong conservation of the protein sequences and the exon/intron boundaries. Sequences were deposited at GenBank under accession number AF044926.

Human Hook cDNAs:
Sequence-homology searches of expressed sequence tags (EST) databases using the BLAST program (ALTSCHUL et al. 1990 ) identified two distinct groups of human ESTs with extensive similarity to the Drosophila hook gene. The first group consisted of six I.M.A.G.E. Consortium (LLNL) cDNA clones: 592403, 112311, 113403, 214189, 296115, and AA317590 (LENNON et al. 1996 ). The second group consisted of cDNAs 590906, 289823, 346067, 511010, 547297, 594759, 773511, and AA19491. The longest cDNA of the first group (592403) and several cDNAs of the second group (590906, 773511, 346067, 511010) were fully sequenced. Based on extended identities within each group, the cDNAs were derived from two human genes designated h-hook1 and h-hook2. The longest cDNA available for h-hook1 (592403) contained the entire open reading frame. The longest available cDNA for h-hook2 (590906) extended from bp 123 (aa 17) to the 3' end; at its 5' end this cDNA was fused to an unrelated cDNA. The 5' end of the h-hook2 sequence was determined by 5' RACE from placenta cDNAs using the Marathon 5' RACE kit according to the manufacturer's instructions (Clontech, Palo Alto, CA). Alternative splicing appears to be responsible for a second class of shorter h-hook2 cDNAs found in the EST databases (e.g., I.M.A.G.E. Consortium cDNA clone 773511). This cDNA was identical in sequence to the h-hook2 cDNA, but lacked sequences encoding amino acids 173 to 522 of the full-length h-Hook2 protein and 46 bp of 3' untranslated bases. Sequences of the human h-hook1 and h-hook2 cDNAs were deposited at GenBank under the accession numbers AF044923 and AF044924.

Western analysis:
To assess the presence of Hook proteins in various alleles, all hook alleles were placed in trans to Df(2L)TW130 and processed for Western analysis. Three flies were homogenized in 300 µl preheated (100°) 1x Laemmli loading buffer and immediately heated to 100° for 2 min. After removal of insoluble material by a 15-min spin at 20,000 x g, samples were stored at -80° until loading 20-µl aliquots. To assess overexpression of Hook1-551Myc under control of the pGMR promoter/enhancer cassette, 10 fly heads were homogenized in 200 µl of preheated loading buffer and processed as described above. After separation of extracts by SDS-PAGE, Hook proteins were detected with affinity-purified anti-Hook antibodies at a dilution of 1:3000 as described (KRAMER and PHISTRY 1996 ). To control for loading, the blots were stripped by washing twice in 30 mM triethanolamine (pH 11) and 100 mM glycine/HCl (pH 2.5) for 10 min each. The stripped blots were blocked with 3% nonfat dry milk for 1 hr and probed with monoclonal antibodies against -tubulin (mAbDM1A; Sigma, St. Louis) at a dilution of 1:1000. Primary antibodies were visualized using horseradish peroxidase (HRP)-coupled secondary antibodies and enhanced chemiluminescence (Super-Signal; Pierce, Rockford, IL).

Histology:
Boss protein in eye imaginal discs was visualized using affinity-purified anti-BossNN1 antibodies as described (SEVRIOUKOV et al. 1998 ). Quantitative analysis of endocytosed Boss protein in different genetic backgrounds was performed blind. One of us (M.P.) encoded the slides with mounted, Boss-stained eye discs, while a second person (H.K.) counted any detectable Boss staining in the position of R7 cells from rows 6 to 12 in at least five eye discs. In this manner, at least 500 R7 cells were assessed for Boss staining for each genetic background listed in Figure 3.

View larger version (126K): In this window In a new window Download PPT slide   Figure 1. The Hook protein is conserved between flies and humans. (A) Comparison between the two Drosophila and the two human Hook proteins reveals conservation over their entire length, including the determinants of the coiled-coil secondary structure (indicated by gray shading). Numbers under the different domains of each protein indicate the percentage of identical amino acids compared with the D. melanogaster protein. The numbers in parentheses above h-Hook1 indicate identity between the two human proteins. Sequences of the human h-hook1 and h-hook2 cDNAs and the D. virilis hook gene and the deduced proteins were deposited at GenBank under the accession numbers AF044923, AF044924, and AF044926, respectively. (B) Alignment of the conserved 125-aa N-terminal domain of Hook proteins using the program MACAW (SCHULER et al. 1991 ). Black boxes indicate identity between all four proteins. Dark or light-gray boxes indicate identity between three or two proteins, respectively.

View larger version (80K): In this window In a new window Download PPT slide   Figure 2. Expression of truncated Hook proteins mimics the hook phenotype. The effect of full-length or truncated Hook proteins on endocytosed Boss was assayed using anti-Boss antibody staining of eye imaginal discs: (A) wild type; (B) hk1; (C) w1118; P[w+, sev-HA248-679Hook]; (D) w1118; P [w+, sev-Hook1-551Myc]. The expression of N- and C-terminally truncated Hook proteins (C and D) dramatically reduces the level of detectable Boss in R7 cells. In A to D, posterior is to the left, the right edge of each panel corresponds to row 3 posterior to the furrow, the onset of Boss expression. Arrowheads indicate staining for Boss in R7 cells. The inset in A indicates the relative position of R cells in developing ommatidia. The scale bar in D corresponds to 4 µm in all panels. (E) Boss staining in R7 cells from row 6 to row 12 was quantified by scoring the percentage of R7 cells with detectable Boss staining. Schematic drawings indicate the expressed wild-type and mutant Hook proteins at the hook locus (endogenous) and those expressed under control of the sev enhancer (transgene); coiled-coil domains of Hook are indicated as gray blocks. A total of at least 500 R7 cells were scored in five eye discs for each genetic background. Three different transgenic lines expressing the C-terminally truncated Hook1-551Myc protein exhibited the same drastic reduction in Boss staining in R7 cells. Note that by examining different focal planes on the microscope we detected low amounts of Boss immunoreactivity in R7 cells of hk1 eye discs not detectable in the micrograph (B). Error bars indicate standard errors of the means.

View larger version (53K): In this window In a new window Download PPT slide   Figure 3. Molecular analysis of hook alleles. The position of mutations in the Hook proteins of sequenced hook alleles is summarized in the schematic drawing (A). In the hk11 allele only the first 38 aa are correctly translated due to a mutation in a splice acceptor site. The conserved 125-aa N-terminal domain and the indicated coiled-coil domains of the Hook protein are therefore deleted in hk11. (B) The predicted truncations of Hook proteins in the indicated hook alleles were confirmed by Western analysis. Reduced levels of truncated proteins can be detected in the hook alleles hk1, hkC1, hk492, hk16, and hk20. The internal deletion of 100 aa in hk14 results in a truncated, stable Hook protein expressed at near wild-type levels. Note that in the hk11 allele no Hook protein is detectable that may be derived from the use of an alternative splice acceptor site. Anti-Hook antibodies used in these blots were raised against a fusion protein encompassing the 414 C-terminal amino acids of the Hook protein (KRAMER and PHISTRY 1996 ). Thus it was not possible to detect C-terminally truncated proteins shorter than ~300 aa; the shortest detected was the 331-aa hk16 gene product. Blots were stripped and reprobed with anti-tubulin antibodies to confirm uniformity of loading and absence of protein degradation. Size markers were prestained proteins (Bio-Rad).

Double-immunofluorescence detection of Boss [anti-bossNN1 (1:3000); KRAMER et al. 1991 ] and Delta [mAb202 (1:10); PARKS et al. 1995 ] or Scabrous [mAbSca1 (1:100); LEE et al. 1996 ] was performed as previously described (KRAMER and PHISTRY 1996 ). Staining was visualized using a Bio-Rad (Richmond, CA) MRC1000 confocal microscope. Z-sections were collected with a 100x objective at a stepsize of 0.8 µm starting at the apical surface, which was defined by the Boss-labeling of the R8 cell surface. Five to six sections were projected into one image and processed as described (KRAMER and PHISTRY 1996 ). Sections of plastic-embedded adult eyes for light microscopy were obtained using standard procedures (VAN VACTOR et al. 1991 ).

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The Hook protein is conserved:
As a first step toward defining functionally important domains in the Hook protein we isolated homologs from different species. A Hook homolog isolated from D. virilis was 83% identical to the D. melanogaster protein over its entire length (Figure 1). Such a high level of identity indicates functional conservation, but is not helpful in delineating functional domains. Additional homologs were identified from the human EST database. Two distinct classes of cDNAs encoded proteins that we named h-Hook1 and h-Hook2, based on their similarity to the Drosophila Hook protein. Both human proteins were similar in size to the D. melanogaster Hook protein (719 aa and 728 aa, respectively) and sequence identities to the D. melanogaster protein over the entire lengths of the proteins were 33 and 30%, respectively. Conservation included the determinants of the centrally located coiled-coil domain (Figure 1A). A 125-aa domain, N terminal to the coiled-coil domain, exhibited a notably higher conservation with 49 and 47% identity to the D. melanogaster protein, respectively (Figure 1B).

Expression of truncated Hook proteins mimics the Hook phenotype:
To test the importance of this conserved domain we expressed full-length and truncated Hook proteins in eye imaginal discs under control of a duplicated sevenless enhancer (BASLER et al. 1991 ) and measured their effects on endocytosed Boss protein. In wild-type eye discs, Boss accumulation in MVBs could be detected by immunohistochemistry in >90% of R7 cells (see arrowheads in Figure 2A). In hk¹ eye discs, Boss accumulation in R7 cells was reduced. In <30% of R7 cells could we detect any indication of internalized Boss (Figure 2B and Figure E). This mutant phenotype could be rescued by expression of the full-length Hook protein in R7 cells (Figure 2E and KRAMER and PHISTRY 1996 ).

We generated transgenic flies that express an N-terminally truncated Hook protein (HA248-679Hook) in R7 cells under control of the sevenless enhancer at the time of Boss endocytosis. Overexpression of this truncated Hook protein in a wild-type background caused a significant reduction (50%) in Boss-positive MVBs of R7 cells (Figure 2C and Figure E). An even stronger effect resulted from the expression of a C-terminally truncated Hook protein (Hook1-551-Myc). In eye discs expressing Hook1-551-Myc in R7 cells, Boss staining could be detected in <7% of R7 cells (Figure 2D and Figure E). Identical results were obtained from three independent transgenic lines expressing Hook1-551-Myc, confirming that this change in Boss accumulation in R7's MVBs was due to the expression of the C-terminally truncated Hook protein. Neither of these transgenes could rescue the hook mutant phenotype when expressed in a hk¹ background (data not shown). Attempts to revert the phenotypes of these truncated proteins by increasing the level of full-length Hook protein were inconsistent between different transgenic lines. Therefore, the phenotypes caused by these transgenes are consistent with dominant-negative activities of truncated Hook proteins, but further tests will be necessary to establish whether they act as classic antimorphs by titrating out the wild-type Hook proteins.

Identification of a hook null allele:
The phenotypes caused by expression of the two truncated Hook proteins raised an important issue. All previously described hook alleles expressed truncated Hook proteins of 50 to 70 kD in size, although at lower levels than wild type (Figure 3 and KRAMER and PHISTRY 1996 ). Sequencing revealed that these three proteins were C-terminal truncations due to mutations in a splice acceptor site for hk¹, a nonsense mutation at aa 546 for hk^C1, and a large insertion for hk⁴⁹² (Table 1). Together with our observations of the effects of overexpressed truncated Hook proteins, these findings raised the possibility that previously identified phenotypes of these hook alleles could be partially due to inhibitory effects of the truncated Hook proteins rather than representing the loss-of-function phenotype of hook.

To address this issue, we screened for new hook alleles by noncomplementation of hk¹. When hk¹ was placed in trans to Df(2L)TW130, a deficiency of the hook genomic region, the resulting flies were viable, fertile, and exhibited the characteristic hooked-bristle phenotype, confirming that it was possible to isolate hook null alleles in this screen. A cytologically visible deficiency (hk²¹) and nine new hook alleles were recovered from 45,000 males screened (Table 1). All of the nine new hook alleles were viable and fertile when placed over Df(2L)TW130 and exhibited the hooked-bristle phenotype. In addition, all alleles reduced Boss accumulation in MVBs of R7 cells as described for the previously existing alleles (data not shown).

Sequence analysis of the different hook alleles revealed mutations distributed throughout the protein (Figure 3A and Table 1). Most mutations resulted in truncated Hook proteins with C-terminal deletions of various lengths. An exception was hk¹⁴; in this allele an in-frame deletion resulted in the loss of amino acids 219 to 318. These findings were confirmed by Western analysis. The truncated hk¹⁴ protein was expressed at levels close to wild type, indicating that an important function of Hook is associated with the part deleted in hk¹⁴. In alleles with C-terminally located mutations, truncated proteins of the predicted sizes could be detected (Figure 3B).

Animals bearing the hk¹¹ null allele are viable:
Sequence analysis indicated that the hk¹¹ allele is truncated after only 38 amino acids due to a mutation in a splice acceptor site in the first intron. The deleted portion of the protein therefore included the conserved 125-aa N-terminal domain (Table 1). Western analysis confirmed that no alternative splice acceptor site was used to create any detectable Hook protein (Figure 3B).

Consistent with the molecular data, phenotypic analysis indicated that hk¹¹ exhibited a stronger mutant phenotype than the previously described allele hk¹. Endocytosed Boss protein was no longer detectable in R7 cells (Figure 4B and Figure 6). With regard to the Boss endocytosis phenotype (Figure 4B), the morphology of the adult eye (Figure 5C), and the hooked-bristle phenotype (Figure 4D), the hk¹¹ homozygous phenotype was indistinguishable from that of hk¹¹ over Df(2L)TW130. On the basis of this genetic evidence, in combination with the molecular analysis of hk¹¹, we concluded that the hk¹¹ mutation resulted in a complete loss of hook function.

View larger version (161K): In this window In a new window Download PPT slide   Figure 4. The hk11 null allele causes pleiotropic defects. Phenotypes of the hk11 null allele were similar but more pronounced than those of the previously described hypomorphic alleles (MOHR 1927 ; KRAMER and PHISTRY 1996 ). Boss accumulation in MVBs of R7 cells that could be observed in wild-type eye discs (arrowheads in A) and at drastically reduced levels in hk1 (see Figure 2) is no longer detectable in hk11 (B). The right edges of A and B correspond to row 3 of eye discs stained for Boss. The inset in A indicates the relative position of R cells in the developing ommatidium. (C) Macrochaete of wild-type flies are straight, while macrochaete of hk11 flies (D) exhibit the characteristic (MOHR 1927 ) hooked-bristle phenotype visualized by scanning electron microscopy. Bar in B represents 4 µm in A and B, 20 µm in C, and 30 µm in D.

View larger version (70K): In this window In a new window Download PPT slide   Figure 5. The hook mutation does not cause retinal degeneration. (A–E) Sections of eyes from adult flies with the following genotypes: (A) wild type, (B) hk1, (C) hk11, (D) hk1/hk11, E) w1118; P[GMR-Hook1-551Myc]4. Changes in ligand endocytosis in hk11 eye discs (see Figure 4 and Figure 6) do not affect cell fate decisions in the developing eye; all photoreceptor cells are present in sections of adult hk11 compound eyes (C) and they do not exhibit any signs of degeneration. Eyes of hk1 flies (B) exhibit strong degeneration as indicated by the general disorganization of the compound eye. Arrows in B indicate individual examples of degenerated rhabdomeres. (D) Flies heterozygous for the hk11 and hk1 alleles do not exhibit this degeneration. (F) An eye imaginal disc expressing the P[GMR-Hook1-551Myc]4 transgene was stained with anti-Myc antibodies. Expression of the Hook1-551Myc protein is initiated at the morphogenetic furrow (arrowhead). In photoreceptor cells, the protein is also localized to axons entering the optic lobes (arrow). Two transgenic lines expressing the Hook1-551Myc protein do not exhibit retinal degeneration (E and data not shown). Bar in D corresponds to 10 µm in A to E, and 40 µm in F. (G) Western blot comparing the expression levels of endogenous wild-type Hook protein and the transgenic Hook1-551Myc protein in extracts of adult eyes. Loading in each lane corresponded to one fly head (lane 1, P[GMR-Hook1-551Myc]14; lane 2, P[GMR-Hook1-551Myc]4; lane 3, Oregon R). Proteins were detected using anti-Hook antibodies at a dilution of 1:3000 (KRAMER and PHISTRY 1996 ).

View larger version (56K): In this window In a new window Download PPT slide   Figure 6. The Hook endocytosis phenotype is not specific for the Boss ligand or the eye disc. Eye discs from wild-type (A, C) or hk11 (B, D) larvae were double stained for Boss and Delta (A, B) or Boss and Scabrous (C, D). Boss staining is visualized in red and Boss internalized into R7 cells is indicated by arrowheads. Note that the apical vesicles in the morphogenetic furrow (MF) of wild-type eye discs were intensely labeled by (A) anti-Delta or (C) anti-Scabrous antibodies (visualized in green; some labeled by arrows). Such staining is not observed in hk11 mutant eye discs (B, D), while the cell surface localization for the transmembrane ligands Boss and Delta appears unaffected. The inset in A indicates the relative position of cells in developing ommatidia. Similar effects on Delta endocytosis were observed in wing discs. In wild-type wing discs (E), a considerable fraction of Delta protein is detected in vesicles, whereas in hk11 mutant wing discs (F) only cell surface staining is detected. Bar in F corresponds to 4 µm in all panels.

In the original hk¹ mutant, we had observed a light-dependent degeneration of the compound eye (Figure 5B and KRAMER and PHISTRY 1996 ). Eye degeneration was weaker, but still detectable, when hk¹ was placed over the deficiency Def(2L)130TW that uncovers hook. Flies carrying the hk¹¹ null allele over Def(2L)130TW did not exhibit any detectable signs of retinal degeneration (Figure 5C). These results indicated that retinal degeneration in the hk¹ mutants was either caused by a different gene on the hk¹ chromosome or due to an effect caused by the expression of the truncated hk¹ protein unrelated to the loss-of-function phenotype of hook (KRAMER and PHISTRY 1996 ).

To test the second possibility we expressed the truncated Hook1-551-Myc protein in adult eyes using the pGMR transformation vector (HAY et al. 1994 ) that provides high-level expression of proteins in the compound eye under control of the glass promoter. Expression of Hook1-551-Myc protein could be observed starting shortly after the initiation of cellular differentiation posterior to the morphogenetic furrow in the eye imaginal disc (Figure 5F). Western analysis of extracts from adult eyes indicated that expression levels of Hook1-551-Myc under control of the GMR enhancer/promoter cassette were considerably higher than that of the wild-type Hook protein (Figure 5G). While expression of this protein effectively interferes with endocytic trafficking of the Boss protein in the eye disc (Figure 2), we could not detect any degeneration of the compound eye in two different lines expressing the GMR-Hook1-551Myc transgene (Figure 5E and data not shown). Given these results and the lack of eye degeneration in the hk¹¹/hk¹ allelic combination (Figure 5D) or in the absence of any hook function (Figure 5C), we concluded that the light-dependent retinal degeneration observed in the hk¹ allele was likely caused by a mutation in a different gene on the hk¹ chromosome.

The hk¹¹ null allele affects localization of multiple endocytosed ligands:
Like Boss, the Delta transmembrane ligand is also internalized across cell boundaries into cells expressing its receptor, Notch (KLUEG et al. 1998 ). In wild-type eye discs that were double-labeled for Boss (red) and Delta (green), the majority of Delta localized to intensely stained vesicles rather than the cell surface (Figure 6A). We have previously shown that hook mutations did not specifically affect endocytosed Boss protein, but also changed Delta localization. The number of vesicles stained with Delta antibodies was reduced by a factor of three in hk¹ eye discs compared to wild-type eye discs (KRAMER and PHISTRY 1996 ). This effect was even more pronounced in the hk¹¹ null allele, in which Delta appeared confined to the apical cell surface (Figure 6B, Delta in green). This effect was not specific for the eye disc. While Delta localization in vesicles could be observed in wild-type wing discs (Figure 6E and HUPPERT et al. 1997 ), such vesicle staining was absent in hk¹¹ wing discs (Figure 6F).

In the original hk¹ mutant, we were not able to discern effects on the internalization of soluble ligands. To assess the effect of the hk¹¹ null allele on endocytosed soluble ligands, we analyzed the distribution of the Scabrous ligand in the eye disc (MLODZIK et al. 1990 ). In wild-type eye discs, the Scabrous protein (green in Figure 6C) localized to intensely stained apical vesicles in the morphogenetic furrow (at the right-hand side of Figure 6C), anterior to the cells that express Boss (red). However, when we analyzed Scabrous localization in hk¹¹ eye discs, we noticed a dramatic reduction of staining in apically localized vesicles (Figure 6D). We concluded from these data that the effects of hook mutations on endocytosis are not specific to any one ligand or type of ligand. Rather, our results point to a general role of the Hook protein in endocytic trafficking.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

In this article we report the identification of a null allele of the hook gene and the analysis of its associated phenotype. The two most important conclusions are that the hook gene is required for normal endocytic trafficking of different classes of ligands, but hook null flies nevertheless are still viable and fertile.

This contrasts sharply with the consequences of mutations in three genes that are required for the initial internalization step at the plasma membrane. Mutations in the Drosophila clathrin heavy chain gene are cell-lethal (BAZINET et al. 1993 ). Similarly, -Adaptin is required for viability (GONZALEZ-GAITAN and JACKLE 1997 ). The -Adaptin protein is widely expressed, preferentially in cell types that exhibit high rates of endocytosis (DORNAN et al. 1997 ), and is required for synaptic vesicle recycling (GONZALEZ-GAITAN and JACKLE 1997 ). Similarly, a defect in synaptic transmission causes fast paralysis in temperature-sensitive alleles of the shibire gene (GRIGLIATTI et al. 1973 ; RAMASWAMI et al. 1994 ), which encodes the Drosophila Dynamin protein (CHEN et al. 1991 ; VAN DER BLIEK and MEYEROWITZ 1991 ).

A role for these essential genes or their mammalian homologs in the initial internalization step of endocytosis has been amply documented (ROBINSON 1994 ), providing proof that endocytosis is strictly required for viability in Drosophila. Two models could explain our finding that a hook null allele is viable: (i) hook gene function is partially redundant and loss of hook function can be compensated for by a second gene, or alternatively, (ii) the hook gene affects a different step in the endocytosis pathway from the three genes mentioned above, and this step is not crucial for viability.

Although at this point we cannot make a final distinction between these two possibilities, we strongly favor the second one. Our attempts to identify additional hook-like genes by sequence homology so far have been unsuccessful; but such negative results are not conclusive. More importantly, localization of the Hook protein to endocytic vesicles and vacuoles points to a role of hook later in the endocytic pathway than the initial internalization step (KRAMER and PHISTRY 1996 ). This differs from proteins such as Adaptin, Dynamin, and Clathrin that are required for the internalization step at the plasma membrane (ESTES et al. 1996 ; GONZALEZ-GAITAN and JACKLE 1997 ). While the lack of specific markers for subsets of endocytic vesicles in Drosophila precluded the precise identification of the Hook-positive vesicles, they could be clearly separated from lysosomes, the final destination of internalized dextran particles (KRAMER and PHISTRY 1996 ).

An additional indication that the hook mutation affects endocytic trafficking at a stage other than the initial internalization step comes from the comparison of its effects on signaling pathways to that of the shibire mutant. Recently it has been demonstrated that shibire function is required for proper signaling through the Notch pathway (REDDY et al. 1997 ; SEUGNET et al. 1997 ). However, although the hook mutation dramatically changes the level of detectable internalized Delta protein in eye discs as well as wing discs (see Figure 6), hook mutants did not exhibit any mutant phenotypes similar to the developmental defects in shibire mutants (RAMASWAMI et al. 1993 ).

If hook mutations do not effect the actual internalization of ligands, why then are they no longer detected within cells after their internalization? An interesting possibility that would explain these findings is that the hook mutation may cause changes in endocytic trafficking that result in premature degradation of internalized proteins. This could be caused by a mistargeting of lysosomal hydrolases to early endocytic compartments normally devoid of these enzymes. Alternatively, in cells lacking hook activity, internalized ligands may be transported more quickly to lysosomes, where they are degraded.

Such an increased rate of transport of internalized cargo to lysosomes recently has been demonstrated in fibroblasts derived from patients suffering from mucolipidosis, type IV (CHEN et al. 1998 ). These patients exhibit symptoms typically associated with lysosomal storage diseases. The initial internalization step as well as degradation by lysosomal enzymes appears to be normal in these fibroblasts. Instead, the underlying cause appears to be a defect in vesicular trafficking late in the endocytosis pathway (CHEN et al. 1998 ). It will be interesting to see whether the mutations causing mucolipidosis, type IV, in humans and the hook mutation in Drosophila disrupt similar steps in the control of lysosomal delivery of proteins.

	ACKNOWLEDGMENTS

We thank Arisa Sunio for technical assistance, John Tamkun for the D. virilis library, and Marc Muskavitch and Nick Baker for antibodies. We are grateful to Dennis McKearin, Bruce Horazdovsky, Ellen Lumpkin, Richard Anderson, and the members of our lab for insightful comments on early versions of this manuscript and to Hari Rajagopal for help with cloning the D. virilis homolog. This work was supported by grants I-1300 from The Welch Foundation and EY-10199 from the National Institutes of Health.

Manuscript received July 17, 1998; Accepted for publication October 20, 1998.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

ALTSCHUL, S. F., W. GISH, W. MILLER, E. W. MYERS, and D. J. LIPMAN, 1990  Basic local alignment search tool. J. Mol. Biol. 215:403-410[Medline].

ASHBURNER, M., 1989 Drosophila: A Laboratory Handbook. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

AUSUBEL, F. M., R. BRENT, R. E. KINGSTON, D. D. MOORE, J. G. SEIDMAN et al., 1994 Current Protocols in Molecular Biology. John Wiley & Sons, New York.

BASLER, K., B. CHRISTEN, and E. HAFEN, 1991  Ligand-independent activation of the sevenless receptor tyrosine kinase changes the fate of cells in the developing Drosophila eye. Cell 64:1069-1081[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].

CAGAN, R. L., H. KRÄMER, A. C. HART, and S. L. ZIPURSKY, 1992  The bride of sevenless and sevenless interaction: internalization of a transmembrane ligand. Cell 69:393-399[Medline].

CHEN, M. S., R. A. OBAR, C. C. SCHROEDER, T. W. AUSTIN, and C. A. POODRY et al., 1991  Multiple forms of dynamin are encoded by shibire, a Drosophila gene involved in endocytosis. Nature 351:583-586[Medline].

CHEN, C. S., G. BACH, and R. E. PAGANO, 1998  Abnormal transport along the lysosomal pathway in mucolipidosis, type IV disease. Proc. Natl. Acad. Sci. USA 95:6373-6378[Abstract/Free Full Text].

DORNAN, S., A. P. JACKSON, and N. J. GAY, 1997  Alpha-adaptin, a marker for endocytosis, is expressed in complex patterns during Drosophila development. Mol. Biol. Cell 8:1391-1403[Abstract].

ESTES, P. S., J. ROOS, A. VAN DER BLIEK, R. B. KELLY, and K. S. KRISHNAN et al., 1996  Traffic of dynamin within individual Drosophila synaptic boutons relative to compartment-specific markers. J. Neurosci. 16:5443-5456[Abstract/Free Full Text].

FUTTER, C. E., A. PEARSE, L. J. HEWLETT, and C. R. HOPKINS, 1996  Multivesicular endosomes containing internalized EGF-EGF receptor complexes mature and then fuse directly with lysosomes. J. Cell Biol. 132:1011-1123[Abstract/Free Full Text].

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

GRIGLIATTI, T. A., L. HALL, R. ROSENBLUTH, and D. T. SUZUKI, 1973  Temperature-sensitive mutations in Drosophila melanogaster. XIV. A selection of immobile adults. Mol. Gen. Genet. 120:107-114[Medline].

GRUENBERG, J. and F. R. MAXFIELD, 1995  Membrane transport in the endocytic pathway. Curr. Opin. Cell Biol. 7:552-563[Medline].

HAY, B. A., T. WOLFF, and G. M. RUBIN, 1994  Expression of baculovirus P35 prevents cell death in Drosophila.. Development 120:2121-2129[Abstract].

HUPPERT, S. S., T. L. JACOBSEN, and M. A. MUSKAVITCH, 1997  Feedback regulation is central to Delta-Notch signalling required for Drosophila wing vein morphogenesis. Development 124:3283-3291[Abstract].

KLUEG, K. M., T. R. PARODY, and M. A. T. MUSKAVITCH, 1998  Complex proteolytic processing acts on Delta, a transmembrane ligand for Notch, during Drosophila development. Mol. Biol. Cell 9:1709-1723[Abstract/Free Full Text].

KOSAKA, T. and K. IKEDA, 1983  Reversible blockage of membrane retrieval and endocytosis in the garland cell of the temperature-sensitive mutant of Drosophila melanogaster, shibire^ts1. J. Cell Biol. 97:499-507[Abstract/Free Full Text].

KRÄMER, H. and M. PHISTRY, 1996  Mutations in the Drosophila hook gene inhibit endocytosis of the Boss transmembrane ligand into multivesicular bodies. J. Cell Biol. 133:1205-1216[Abstract/Free Full Text].

KRÄMER, H., R. L. CAGAN, and S. L. ZIPURSKY, 1991  Interaction of bride of sevenless membrane-bound ligand and the sevenless tyrosine-kinase receptor. Nature 352:207-212[Medline].

LEE, E.-C., X. HU, S. Y. YU, and N. E. BAKER, 1996  The scabrous gene encodes a secreted glycoprotein dimer and regulates proneural development in Drosophila eyes. Mol. Cell. Biol. 16:1179-1188[Abstract].

LENNON, G. G., C. AUFFRAY, M. POLYMEROPOULOS, and M. B. SOARES, 1996  The I.M.A.G.E. Consortium: an integrated molecular analysis of genomes and their expression. Genomics 33:151-152[Medline].

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

MLODZIK, M., N. E. BAKER, and G. M. RUBIN, 1990  Isolation and expression of scabrous, a gene regulating neurogenesis in Drosophila.. Genes Dev. 4:1848-1861[Abstract/Free Full Text].

MOHR, O., 1927  The second chromosome recessive hook bristles in Drosophila melanogaster.. Hereditas 9:169-179.

MUKHERJEE, S., R. N. GHOSH, and F. R. MAXFIELD, 1997  Endocytosis. Physiol. Rev. 77:759-803[Abstract/Free Full Text].

PARKS, A. L., F. R. TURNER, and M. A. MUSKAVITCH, 1995  Relationships between complex Delta expression and the specification of retinal cell fates during Drosophila eye development. Mech. Dev. 50:201-216[Medline].

RAMASWAMI, M., S. RAO, A. VAN DER BLIEK, R. B. KELLY, and K. S. KRISHNAN, 1993  Genetic studies on dynamin function in Drosophila.. J. Neurogenet. 9:73-87[Medline].

RAMASWAMI, M., K. S. KRISHNAN, and R. B. KELLY, 1994  Intermediates in synaptic vesicle recycling revealed by optical imaging of Drosophila neuromuscular junctions. Neuron 13:363-375[Medline].

REDDY, G. V., B. GUPTA, K. RAY, and V. RODRIGUES, 1997  Development of the Drosophila olfactory sense organs utilizes cell-cell interactions as well as lineage. Development 124:703-712[Abstract].

ROBINSON, M. S., 1994  The role of clathrin, adaptors and dynamin in endocytosis. Curr. Opin. Cell Biol. 6:538-544[Medline].

RUBIN, G. M. and A. C. SPRADLING, 1982  Genetic transformation of Drosophila with transposable element vectors. Science 218:348-353[Abstract/Free Full Text].

SCHULER, G. D., S. F. ALTSCHUL, and D. J. LIPMAN, 1991  A workbench for multiple alignment construction and analysis. Proteins 9:180-190[Medline].

SEUGNET, L., P. SIMPSON, and M. HAENLIN, 1997  Requirement for dynamin during Notch signaling in Drosophila neurogenesis. Dev. Biol. 192:585-598[Medline].

SEVRIOUKOV, E., J. H. WALENTA, M. PHISTRY, A. SUNIO, and H. KRÄMER, 1998  Oligomerization of the extracellular domain of Boss enhances its binding to the Sevenless receptor and its antagonistic effect on R7 induction. J. Cell Sci. 111:737-747[Abstract].

STATHAKIS, D. G., E. S. PENTZ, M. E. FREEMAN, J. KULLMAN, and G. R. HANKINS et al., 1995  The genetic and molecular organization of the DOPA Decarboxylase gene cluster of Drosophila melanogaster.. Genetics 141:629-655[Abstract].

VAN DER BLIEK, A. M. and E. M. MEYEROWITZ, 1991  Dynamin-like protein encoded by the Drosophila shibire gene associated with vesicular traffic. Nature 351:411-414[Medline].

VAN VACTOR, D. L. J., R. L. CAGAN, H. KRÄMER, and S. L. ZIPURSKY, 1991  Induction in the developing compound eye of Drosophila: multiple mechanisms restrict R7 induction to a single retinal precursor cell. Cell 67:1145-1155[Medline].

ZIPURSKY, S. L. and G. M. RUBIN, 1994  Determination of neuronal cell fate: lessons from the R7 neuron of Drosophila.. Annu. Rev. Neurosci. 17:373-397[Medline].

This article has been cited by other articles:

				A. Simonsen, R. C. Cumming, K. Lindmo, V. Galaviz, S. Cheng, T. E. Rusten, and K. D. Finley Genetic Modifiers of the Drosophila Blue Cheese Gene Link Defects in Lysosomal Transport With Decreased Life Span and Altered Ubiquitinated-Protein Profiles Genetics, June 1, 2007; 176(2): 1283 - 1297. [Abstract] [Full Text] [PDF]

				H. Sano, M. Ishino, H. Kramer, T. Shimizu, H. Mitsuzawa, C. Nishitani, and Y. Kuroki The Microtubule-binding Protein Hook3 Interacts with a Cytoplasmic Domain of Scavenger Receptor A J. Biol. Chem., March 16, 2007; 282(11): 7973 - 7981. [Abstract] [Full Text] [PDF]

				K. Wilhelmsen, M. Ketema, H. Truong, and A. Sonnenberg KASH-domain proteins in nuclear migration, anchorage and other processes J. Cell Sci., December 15, 2006; 119(24): 5021 - 5029. [Abstract] [Full Text] [PDF]

				A. L. Parks, J. R. Stout, S. B. Shepard, K. M. Klueg, A. A. Dos Santos, T. R. Parody, M. Vaskova, and M. A. T. Muskavitch Structure-Function Analysis of Delta Trafficking, Receptor Binding and Signaling in Drosophila Genetics, December 1, 2006; 174(4): 1947 - 1961. [Abstract] [Full Text] [PDF]

				A. A. Lanahan, T. W. Chittenden, E. Mulvihill, K. Smith, S. Schwartz, and M. Simons Synectin-dependent gene expression in endothelial cells Physiol Genomics, November 21, 2006; 27(3): 380 - 390. [Abstract] [Full Text] [PDF]

				C.-M. Lin, H.-J. Chen, C. L. Leung, D. A. D. Parry, and R. K. H. Liem Microtubule actin crosslinking factor 1b: a novel plakin that localizes to the Golgi complex J. Cell Sci., August 15, 2005; 118(16): 3727 - 3738. [Abstract] [Full Text] [PDF]

				E. A. Sevrioukov, N. Moghrabi, M. Kuhn, and H. Kramer A Mutation in dVps28 Reveals a Link between a Subunit of the Endosomal Sorting Complex Required for Transport-I Complex and the Actin Cytoskeleton in Drosophila Mol. Biol. Cell, May 1, 2005; 16(5): 2301 - 2312. [Abstract] [Full Text] [PDF]

				K. Luiro, K. Yliannala, L. Ahtiainen, H. Maunu, I. Jarvela, A. Kyttala, and A. Jalanko Interconnections of CLN3, Hook1 and Rab proteins link Batten disease to defects in the endocytic pathway Hum. Mol. Genet., December 1, 2004; 13(23): 3017 - 3027. [Abstract] [Full Text] [PDF]

				J. A. Fischer, S. Acosta, A. Kenny, C. Cater, C. Robinson, and J. Hook Drosophila Klarsicht Has Distinct Subcellular Localization Domains for Nuclear Envelope and Microtubule Localization in the Eye Genetics, November 1, 2004; 168(3): 1385 - 1393. [Abstract] [Full Text] [PDF]

				S. C. W. Richardson, S. C. Winistorfer, V. Poupon, J. P. Luzio, and R. C. Piper Mammalian Late Vacuole Protein Sorting Orthologues Participate in Early Endosomal Fusion and Interact with the Cytoskeleton Mol. Biol. Cell, March 1, 2004; 15(3): 1197 - 1210. [Abstract] [Full Text] [PDF]

				E. L. Rawlins, B. Lovegrove, and A. P. Jarman Echinoid facilitates Notch pathway signalling during Drosophila neurogenesis through functional interaction with Delta Development, December 29, 2003; 130(26): 6475 - 6484. [Abstract] [Full Text] [PDF]

				A. S. Rawls and T. Wolff Strabismus requires Flamingo and Prickle function to regulate tissue polarity in the Drosophila eye Development, May 1, 2003; 130(9): 1877 - 1887. [Abstract] [Full Text] [PDF]

				I. Mendoza-Lujambio, P. Burfeind, C. Dixkens, A. Meinhardt, S. Hoyer-Fender, W. Engel, and J. Neesen The Hook1 gene is non-functional in the abnormal spermatozoon head shape (azh) mutant mouse Hum. Mol. Genet., July 1, 2002; 11(14): 1647 - 1658. [Abstract] [Full Text] [PDF]

				J. H. Walenta, A. J. Didier, X. Liu, and H. Kramer The Golgi-associated Hook3 Protein Is a Member of a Novel Family of Microtubule-binding Proteins J. Cell Biol., February 26, 2001; 152(5): 923 - 934. [Abstract] [Full Text] [PDF]

R. Roepman, N. Bernoud-Hubac, D. E. Schick, A. Maugeri, W. Berger, H.-H. Ropers, F. P.M. Cremers, and P. A. Ferreira The retinitis pigmentosa GTPase regulator (RPGR) interacts with novel transport-like proteins in the outer segments of rod photoreceptors Hum. Mol. Genet., September 1, 2000; 9(14): 2095 - 2105. [Abstract] [Full Text] [PDF]