Mutations in chs1/beige result in a deficiency in intracellular transport of vesicles that leads to a generalized immunodeficiency in mice and humans. The function of NK cells, CTL, and granulocytes is impaired by these mutations, indicating that polarized trafficking of vesicles is controlled by CHS1/beige proteins. However, a molecular explanation for this defect has not been identified. Here we describe a novel gene with orthologues in mice, humans, and flies that contains key features of both chs1/beige and A kinase anchor genes. We designate this novel gene lba for LPS-responsive, beige-like anchor gene. Expression of lba is induced after LPS stimulation of B cells and macrophages. In addition, lba is expressed in many other tissues in the body and has three distinct mRNA isoforms that are differentially expressed in various tissues. Strikingly, LBA-green-fluorescent protein (GFP) fusion proteins are localized to vesicles after LPS stimulation. Confocal microscopy indicates this protein is colocalized with the trans-Golgi complex and some lysosomes. Further analysis by immunoelectron microscopy demonstrates that LBA-GFP fusion protein can localize to endoplasmic reticulum, plasma membrane, and endocytosis vesicles in addition to the trans-Golgi complex and lysosomes. We hypothesize that LBA/CHS1/BG proteins function in polarized vesicle trafficking by guiding intracellular vesicles to activated receptor complexes and thus facilitate polarized secretion and/or membrane deposition of immune effector molecules.

Lipopolysaccharide is a potent inducer of maturation in B cells, monocytes, and dendritic cells that facilitates production of inflammatory cytokines, NO, and Ag presentation so that these cells can participate in the immune response to bacterial pathogens (1, 2, 3). In an attempt to identify genes involved in the maturation of immune cells, we developed a gene-trapping strategy to identify mammalian genes the expression of which is altered by cellular stimuli (4). We successfully trapped several novel LPS-responsive genes (5) including the SHIP gene, which plays a role in controlling the maturation and proliferation of B cells and monocyte/macrophages in vivo (6, 7, 8).

Chediak-Higashi syndrome (CHS)3 patients suffer from a systematic immunodeficiency characterized by a severe immune defect, hypopigmentation, progressive neurologic dysfunction, and a bleeding diathesis (9). Specific defects in immune cells include defects in T cell cytotoxicity (10, 11), killing by NK cells (12), defective bactericidal activity, and chemotaxis by granulocytes and monocytes (13). CHS and beige lysosomes also exhibit compartmental missorting of proteins (14). Other studies have found that beige macrophages are defective for class II surface presentation (15, 16) and that T cells in CHS patients are defective for CTLA4 surface expression (17). All cells in beige mice and CHS patients bear giant vesicles that cluster around the nucleus. Affected vesicles include lysosomes, platelet-dense granules, endosomes, and cytolytic granules. These giant vesicles seem normal in several aspects except for their failure to release their contents, probably resulting from the inability of the giant granules to mobilize and/or fuse with the membrane on stimulation (11). However, despite these very provocative findings, there is still no direct evidence that BG(beige)/CHS1 proteins associate with intracellular vesicles; thus, a molecular explanation for defective vesicle trafficking and protein missorting in these diseases is still sought.

The cAMP-dependent protein kinase A holoenzyme is a tetramer composed of two regulatory subunits (R unit) and two catalytic subunits (C unit). A kinase anchor proteins (AKAPs) including their R binding sites are functionally, but not structurally, related in that they share no sequence homology with each other. The R (RII, RI) binding site is an amphipathic helix of 14–18 aa. AKAPs bind protein kinase A (PKA) by inserting the hydrophobic side of the helix into the hydrophobic pocket formed by the two regulatory subunits of a PKA (18). This leads to translocation of PKA to a distinct subcellular location where it is activated by cAMP binding that frees its catalytic subunits to phosphorylate substrates. AKAPs associate with a variety of subcellular structures including centrosomes, dendrites, endoplasmic reticulum (ER), mitochondria, nuclear membrane, plasma membrane, and vesicles (19).

Here we describe a novel LPS-inducible gene in humans and mice that contains features of both chs1/beige and AKAP genes. Previously, we demonstrated that this novel gene in mouse B cells is up-regulated by LPS simulation when lacZ is fused with the endogenous lba gene by gene trapping (5). We designate this gene lba for LPS-responsive and beige-like Anchor gene. The lba gene contains a tandem array of WDL (for WD-like, defined in this paper) repeats, a BEACH (BEige And CHS (20)) domain, and WD repeats shared by chs1/beige, factor associated with neutral sphingomyelinase activation (FAN), lvsA genes, and some anonymous open reading frames (ORFs). The lba gene also has an orthologue in flies, the DAKAP550 gene (21). Expression of lba is induced 2- to 4-fold in B cells and macrophages after LPS stimulation. We find that lba mRNA is present in three different isoforms and that the ratio of these isoforms varies dramatically in different tissues. When an LBA-green-fluorescent protein (GFP) fusion protein is expressed in unstimulated macrophages, GFP fluorescence is cytosolic in most of the cells. However, on LPS stimulation the GFP fluorescence is dramatically targeted to intracellular vesicles in nearly all cells. Confocal microscopy shows that the BEACH-WD domains of lba associate with the Golgi complex and some lysosomes. It is also localized to ER, plasma membrane, and endocytosis vesicles in addition to the trans-Golgi complex and some lysosomes as demonstrated by immunoelectron microscopy. These results provide the first direct evidence that CHS1/BG-like proteins can associate with the intracellular vesicle system and that this localization is increased by activation of cells in the immune system. These results also suggest that PKA may play a role in mediating polarized trafficking of intracellular vesicles. Our studies shed light on understanding the molecular mechanism of the chs1/beige diseases in mice and humans as well as the polarized responses of immune effector cells.

Total RNA was prepared using the RNeasy kit (Qiagen, Valencia, CA). Poly(A)+ RNA was prepared using the Fast Track mRNA isolation kit (Invitrogen, Carlsbad, CA). RNA was prepared from murine cell lines as well as liver and thymus of C57BL6/J mice per the manufacturers’ instructions. RNAs were treated with RNase-free DNase I (Amersham Pharmacia Biotech, Piscataway, NJ) at 10 U/μg RNA for 30 min at 37°C to destroy genomic DNA. First-strand cDNA synthesis was primed with random DNA hexamers or oligo(dT) primers at 42°C for 1 h using the Superscript II RNase H Reverse Transcriptase cDNA Synthesis System (Life Technologies, Gaithersburg, MD).

Primers (forward: 5′-AGAGAAGAGGAGAAGATGTGTGATC-3′; reverse: 5′-CCAGGCTCCATGCTTGTCTGTGAG-3′) were designed from a 143-bp cDNA fragment obtained from our previous gene trap work (5) and combined with λGT10 forward (5′-AGCAAGTTCAGCCTGGTTAAGT3-′) and reverse (5′-TTATGAGTATTTCTTCCAGGG3-′) primers to amplify the lba gene cDNA from a mouse B lymphocyte cDNA library (mouse lymphocyte 5′ stretch cDNA library; Clontech, Palo Alto, CA). These PCR products were then cloned and sequenced. New primers were then designed from these sequences, and further RT-PCR were conducted to extend the cDNA sequence to the 5′ or 3′ direction. The SMART RACE amplification kit (Clontech) was used to amplify 5′-cDNA ends using the following lba-specific primers: 5′-ACTGCAGCAAGCTCCTCCTGTTTTCTC-3′ and a nested primer 5′-TGGGCGAAGAGCGGAAACAGAAC3′, whereas for 3′-cDNA clones, the following primers were used: 5′-AGAGAAGAGGAGAAGATGTGTGATC-3′ and a nested primer 5′-GAGTGATGGATGATGGGACAGTGGTG-3′. PCR conditions for the 5′-rapid amplification of cDNA end (RACE) and 3′-RACE were as follows using the Advantage polymerase mix (Clontech): 94°C for 30 s, followed by 5 cycles at 94°C for 30 s, 70°C for 30 s, and 72°C for 3–5 min; 5 cycles at 94°C for 30 s, 68°C for 30 s, and 72°C for 3–5 min; 20 cycles at 94°C for 30 s, 65°C for 30 s, and 72°C for 3–5 min; and a final extension at 72°C for 30 min. After the full-length cDNA sequence of the lba gene was obtained, several primers were designed to amplify the region of the lba gene cDNA that contains its major ORF. The region containing the major ORF of the lba gene was then amplified from a single source of C57BL6/J liver mRNA and resequenced to confirm that the lba cDNAs obtained from liver cells are identical with that amplified from the aligned cDNA fragments amplified from primary and transformed B lymphocytes, indicating that these represent the major mRNAs expressed from the lba locus. All RT-PCR and RACE products were isolated and purified from agarose gels using the QIAEX II Gel Extraction Kit (Qiagen, Valencia, CA). The purified products were sequenced directly to avoid detecting the mutations introduced during PCR. Both strands of each template were sequenced, and the sequence was confirmed by sequence analysis of at least two independent PCR products. PCR products and RACE products were cloned into PCRII vector (TA cloning kit; Invitrogen), and multiple clones were sequenced. Plasmids were purified from liquid cultures using the Qiagen plasmid Maxi preparation kit.

A search of GenBank found that the murine lba gene has a high degree of homology to a 7.3-kb human partial cDNA sequence (GenBank accession number M83822) called BGL (22), which may be the homologue of the mouse lba gene, and we designate it human lba gene. The 5′-end of the human lba gene was obtained by using a 5′-primer (5′-GCCACCTCCGTCTCGCTGC-3′) from the mouse lba gene cDNA sequence and a 3′-primer (5′-GGGCACTGGGGAGAATTTCGAAGTAGG-3′) from the human BGL sequence. Human lung, brain, and kidney cDNA libraries (Marathon cDNA libraries; Clontech) were used as templates for the amplification of the 5′- and 3′-ends of the human cDNA under the following PCR conditions: 35 cycles at 95°C for 45 s; 60°C for 15 s; 72°C for 3 min. The PCR products were cloned into a TA cloning vector, and multiple clones were sequenced. Additional PCRs were conducted with the primers from the 3′-cDNA clones obtained as described above to complete the sequence of the human lba cDNA. The primer pairs used for these additional 3′-cDNA clones were 5′-TTCAGGCAGTTTTCAGGACCCTCCAAG-3′ and 5′-TAGTGTCTGATGTTGAACTTCCTCCTG-3′. Overlapping regions of the 5′ and 3′ human lba cDNAs were compared and merged with the human BGL cDNA in GenBank to construct a complete sequence for the human lba gene.

70Z/3 B lymphoma cells were maintained in RPMI 1640 supplemented with 10−5 M 2-ME and 10% FBS. J774 cells were maintained in DMEM supplemented with 10% FBS. 70Z/3 cells were stimulated with 10 ng/ml LPS (Sigma, St. Louis, MO), and J774 cells were stimulated with 1 ng/ml LPS for 20 h. Poly(A)+ RNA was prepared from 108 stimulated or unstimulated cells using the FastTrack isolation kit (Invitrogen). Poly(A)+ RNA (5 μg/lane) was size-fractionated by electrophoresis on a 6% formaldehyde/1% agarose gel buffered with MOPS, transferred to a nylon membrane (Stratagene, La Jolla, CA) by capillary action in 20× SSC, and immobilized by UV cross-linking. The filter was probed with a uniformly labeled 32P probe using the Ready-To-Go DNA labeling kit (Amersham Pharmacia Biotech). The probe corresponds to a 2.5-kb PCR product that spans nucleotides 3545–6040 of the murine lba cDNA. The filter was hybridized with the probe in 2×SSC, 0.5% SDS, 5× Denhardt’s containing 100 μg/ml heat denatured salmon sperm DNA at 68°C overnight. Filters were washed twice for 5 min at room temperature in 2× SSC, 0.5% SDS and twice for 30 min at 68°C in 0.1× SSC, 0.1% SDS. Hybridization signals were detected and quantitated using a Molecular Dynamics PhosphorImager and Imagequant software.

The cell lines (70Z/3, BAL17, A20, WEHI231, and S194) used for the RT-PCR were obtained from American Type Culture Collection (Manassas, VA). Spleen, brain, lung, and bone marrow were obtained from C57BL6/J mice. The preparation of total RNA and cDNA synthesis were conducted as described above. First-strand cDNA reaction products (2 μl) were amplified in a 25-μl PCR using primers that detect the three lba isoforms (5′-GGCACAACCTTCCTGCTCAC-3′ and 5′-CCTGTCCCCCATTTGAACCC-3′ for the α form; 5′-ACGGCTGCTTCTGCACCTTC-3′ and 5′-TTTTGGGACAGGGCTTCTCTG-3′ for the β form; 5′-GGCACAACCTTCCTGCTCAC-3′ and 5′-GCAGATGCTCTCCTCGCTCC-3′ for the γ form). The cycling program was: 94°C for 30 s, followed by 5 cycles at 94°C for 30 s, 70°C for 30 s, and 72°C for 4 min; 5 cycles at 94°C for 30 s, 68°C for 30 s, and 72°C for 4 min; 30 cycles at 94°C for 30 s, 62°C for 30 s, and 72°C for 4 min; and a final extension at 72°C for 10 min.

Analyses of the nucleotide and amino acid sequences for the murine and human lba gene were performed using MacVector (Oxford Molecular Group, Oxford, U.K.). Nucleotide sequence alignments and other analyses were conducted using BLAST (23). SMART (24) and ClustlX (25) were used for protein secondary structure predictions. For WD repeat prediction, we used an algorithm developed by Neer et al. (26, 27, 28, 29, 30).

A region from the murine lba cDNA that includes the BEACH and the WD domains 3′ to the BEACH domain were inserted “in-frame” and upstream of the coding region of a modified GFP gene cloned in a proprietary mammalian expression vector pEGFP-N2 (Clontech). Recombinant clones (called pBWEGFP) were picked, plasmid DNAs prepared and sequenced to confirm that no mutations were introduced during these manipulations. Murine 3T3 cells, the macrophage RAW264.7 cells, and human 293 cells were transfected by the FuGEN transfection kit (Roche Molecular Biochemicals, Indianapolis, IN) or by electroporation (Gene Pulser; Bio-Rad Laboratories, Hercules, CA) with 20 μg linearized recombinant plasmid pBWEGFP DNA as well as the control vector pEGFP at 250 V and 500 μF. One day later, cells were cultured in DMEM containing 0.8 μg/ml G418 (Life Technologies). This medium was changed every day for the first 4 days. The surviving G418-resistant colonies were isolated and used for further experimentation. For subcellular localization, cells were plated in glass-covered plates at 2.5 × 105 cells/ml in 2 ml DMEM media with or without LPS at 100 ng/ml. After 12 h, cells were directly examined by fluorescence microscopy using a FITC filter to detect expression of GFP fusion proteins. Fluorescent photomicrography was performed using Nikon model H-III photomicrographic equipment and image software (Nikon, Tokyo, Japan).

The RAW 264.7 cells stably transfected with the pBWEGFP construct were grown on glass coverslips and stimulated with 100 ng/ml LPS for 24 h. Golgi and lysosomes were specifically labeled with BODIPY TR ceramide and LysoTracker Red DND-99 (Molecular Probes, Eugene, OR), respectively, following the manufacturer’s protocols. Briefly, for Golgi labeling, cells were washed three times with PBS and incubated for 30 min at 4°C with 5 μM BODIPY TR ceramide, rinsed several times with ice-cold medium, and then incubated in fresh medium at 37°C for another 30 min. For lysosome labeling, medium was changed with prewarmed fresh medium containing 60–75 nM lysosome probe, and the cell sample was incubated for 30 min. Finally, the medium was removed, washed three times with PBS, fixed with 3.7% formaldehyde for 10–20 min, and washed again; the slides were mounted with 4′,6′-diamidino-2-phenylindole-containing Vectashield medium (Vector Laboratories, Burlingame, CA). Cells were observed on an inverted Axiovert 100M laser scanning confocal microscope (Zeiss, Oberkochen, Germany). GFP fluorescence was excited using a 458/488 nm argon/krypton laser, and emitted fluorescence was detected with 505–530 nm band pass filter. For LysoTracker Red and BODIPY TR, a 633 nm helium/neon laser was used for excitation, and fluorescence was detected with a 585 nm band pass filter, using a ×100 oil immersion lens. The colocalization function of LSM510 software (EMBO Laboratory, Heidelberg, Germany) allows for a reliability of 99% for actual pixels with both fluorophores. The colocalization mask pixels were converted to white color for obviousness.

The RAW 264.7 cells stably transfected with the pBWEGFP construct were grown in the presence of 100 ng/ml LPS for 24 h, washed three times with PBS, fixed with 2% paraformaldehyde in phosphate buffer for 1 h at 4°C, and processed for postembedding immunocytochemistry. The cells were scraped from the dishes in which they were grown and pelleted by low speed centrifugation. The pellets were dehydrated in a graded series of ethanol dilutions and embedded in gelatin capsules in LR White resin. The resin was polymerized for 48 h at 50°C. Ultrathin sections of LR White-embedded cells were collected on nickel grids and immunolabeled according to the technique of Haller et al. (31)with rabbit anti-GFP (Clontech) at a 1:20 ratio for 1 h at room temperature, followed by extensive rinsing and then by labeling with 10 nm goat anti-rabbit IgG-gold (Aurion, Wageningen, The Netherlands) for 1 h at room temperature. Control grids were labeled by replacing the primary Ab with normal rabbit serum. After extensive washing, thin sections were stained with uranyl acetate and lead citrate before examination with immunoelectron microscopy.

A novel LPS-inducible gene was identified by integration of Gensr1 gene trap retrovirus (5). A partial cDNA sequence of the LPS-inducible gene trap cell clone, 7a65, was used to design PCR primers to amplify the upstream and downstream regions of cDNA from a mouse B lymphocyte library. Initially, we obtained a 1.6-kb cDNA sequence by this strategy. Sequence analysis confirmed that this 1.6-kb cDNA sequence contained the original 142-bp sequence obtained by gene trapping (5). 5′-RACE reactions using antisense primers from the 5′-end of this 1.6-kb region yielded additional 5′-cDNA sequences including the 5′-untranslated sequences of the lba gene as well as the ATG of its major ORF. Sense strand primers were also designed from the 1.6-kb cDNA sequence, and three 3′-RACE fragments of 2.5, 2, and 1.4 kb were obtained that have identical 5′-end sequences; however, their 3′-ends differ dramatically. The amino acid sequence of the major ORF in the murine lba cDNA is shown in Fig. 1 A. The human lba orthologue was obtained as described in Materials and Methods.

FIGURE 1.

Sequence and structure of the lba gene. A, Predicted full-length amino acid sequence of the lba-α and lba-β (stopped at the boxed “R” with the additional sequence VSAVGSTLFLLLGSSK) and lba-γ cDNAs (stopped at the boxed “I” with the additional sequence GLPLLSLFAIH). Bold amino acids indicate the proposed BEACH domain (2276–2490) based on alignment with 20 other BEACH domains (data not shown). Eight WD repeats predicted by an algorithm available at http://bmerc-www.bu.edu/psa/request.htm are underlined or dotted-underlined. The first three WD repeats were not predicted by other programs but resemble WD repeats; thus, we refer to them as WDL repeats. Two putative protein kinase A RII-binding sites are shaded. The sequences of the lba cDNAs have been deposited in GenBank with the following GenBank accession numbers: lba-α, AF187731; lba-β, AF188506; lba-γ, AF188507.B, Schematic diagram of mLBA protein and alignment of the predicted mlBA protein with its orthologues and some paralogues. Some part of this figure was taken from a National Center for Biotechnology Information BLAST search result and modified appropriately. The stop sites for lba-β and lba-γ are indicated by dash lines. The hLBA protein was predicted from a 9.9-kb “hybrid” cDNA sequence with the first 5′ 2577 nucleotides from this work (GenBank accession number AF216648) and the rest from the CDC4L partial cDNA sequence (GenBank accession number M83822) (22 ) except one G was added after position 5696 for two reasons: 1) the G base is present in our cDNA sequence (GenBank accession number AF217149); and 2) this addition extended the CDC4L ORF by an additional 165 aa that has high homology with mLBA and other proteins shown in this figure. The dLBA was predicted from the Drosophila melanogaster genomic sequence (GenBank accession number AE003433). cLBA (GenBank accession number T20719, C. elegans), aCDC4L(GenBank accession number T00867; Arabidopsis thaliana), LVSA (GenBank accession number AAD52096, Dictyostelium discoideum), human FAN (hFAN; GenBank accession number NP_0035711; Homo sapiens), CHS1 (Chediak-Higashi syndrome 1; GenBank accession number NP_000072, H. sapiens), murine BG (mBG; GenBank accession number AAB60778; Mus musculus).

FIGURE 1.

Sequence and structure of the lba gene. A, Predicted full-length amino acid sequence of the lba-α and lba-β (stopped at the boxed “R” with the additional sequence VSAVGSTLFLLLGSSK) and lba-γ cDNAs (stopped at the boxed “I” with the additional sequence GLPLLSLFAIH). Bold amino acids indicate the proposed BEACH domain (2276–2490) based on alignment with 20 other BEACH domains (data not shown). Eight WD repeats predicted by an algorithm available at http://bmerc-www.bu.edu/psa/request.htm are underlined or dotted-underlined. The first three WD repeats were not predicted by other programs but resemble WD repeats; thus, we refer to them as WDL repeats. Two putative protein kinase A RII-binding sites are shaded. The sequences of the lba cDNAs have been deposited in GenBank with the following GenBank accession numbers: lba-α, AF187731; lba-β, AF188506; lba-γ, AF188507.B, Schematic diagram of mLBA protein and alignment of the predicted mlBA protein with its orthologues and some paralogues. Some part of this figure was taken from a National Center for Biotechnology Information BLAST search result and modified appropriately. The stop sites for lba-β and lba-γ are indicated by dash lines. The hLBA protein was predicted from a 9.9-kb “hybrid” cDNA sequence with the first 5′ 2577 nucleotides from this work (GenBank accession number AF216648) and the rest from the CDC4L partial cDNA sequence (GenBank accession number M83822) (22 ) except one G was added after position 5696 for two reasons: 1) the G base is present in our cDNA sequence (GenBank accession number AF217149); and 2) this addition extended the CDC4L ORF by an additional 165 aa that has high homology with mLBA and other proteins shown in this figure. The dLBA was predicted from the Drosophila melanogaster genomic sequence (GenBank accession number AE003433). cLBA (GenBank accession number T20719, C. elegans), aCDC4L(GenBank accession number T00867; Arabidopsis thaliana), LVSA (GenBank accession number AAD52096, Dictyostelium discoideum), human FAN (hFAN; GenBank accession number NP_0035711; Homo sapiens), CHS1 (Chediak-Higashi syndrome 1; GenBank accession number NP_000072, H. sapiens), murine BG (mBG; GenBank accession number AAB60778; Mus musculus).

Close modal

Sequence analysis of the lba cDNAs indicates that there are three isoforms with identical 5′-ends that differ at their 3′-termini. These isoforms include a 9903-bp form (lba-α), a 9396-bp form (lba-β), and a 8854-bp form (lba-γ) encoding proteins of 2856, 2792, and 2779 aa, respectively. All three ORFs begin with the same Kozak consensus ATG at nucleotide 308. The first 2776aa of the β form are identical with the first 2776 aa of the α form, whereas the 16 aa at its C terminus are unique to it. The first 2769 aa of the γ form are identical with the first 2769 aa of the α and β forms with its C-terminal 10 aa unique to it; the α form has its C-terminal 80 aa unique to it (Fig. 1). Homology search indicates that all lba isoforms have a BEACH domain (20); however, the lba-α isoform has 5 WD repeats, lba-β has 3 WD repeats, and lba-γ lacks WD repeats (Fig. 1 B). The isoform-specific unique coding sequences and the associated 3′-untranslated sequence (totally 1267 bp for α form, 761 bp for β form, and 845 bp for γ form) show no significant homology with each other. Interestingly, only the α form has an AATAAA sequence for poly(A) recognition and a TGA stop codon, whereas the β and γ forms have TAA stop codons.

Homology analysis revealed that lba has significant homology with the partial protein sequence DAKAP550 (21), which is an AKAP, and with AKAP550 (GenBank accession number AAF46011) predicted from the Drosophila genomic sequence (GenBank accession number AE003433). BLAST search indicates the DAKAP550 and AKAP550 are identical and thus are the same gene (data not shown). We predicted a longer sequence for this gene from the genomic sequence and designated it dLBA, which is identical with the AKAP550 except that it has an additional 160 aa at its N terminus. Here and later, we place the first letter of the genus before the name of the gene to distinguish the lba genes of different species. Therefore DAKAP550 is a partial sequence of dLBA and AKAP550. Amino acid alignment analysis shows that the murine LBA protein has 85% amino acid identity with human LBA, 51% amino acid identity with dLBA, and 35% amino acid identity with the Caenorhabditis elegans CDC4L gene (GenBank accession number T20719) (designated cLBA for clarity) (Fig. 1,B). This homology analysis shows that the lba and DAKAP550 genes are orthologues based on their high homology, which extends from their N terminus to the C terminus ( Figs. 1–3 and Table I). Furthermore, two putative PKA-binding sites are found in all lba orthologues (Fig. 2) and are structurally similar to the B1 and B2 RII binding sites of DAKAP550, a protein that has been demonstrated to bind PKA in vitro and in vivo (21). This region is highly conserved in lba orthologues in mice, humans, Drosophila, and C. elegans (Fig. 2 A) and potentially provides another two PKA binding sites for DAKAP550. Unexpectedly, the B1 and B2 sites of DAKAP550 are not found in other LBA proteins; they may be species specific, and these potential RII binding sites must be confirmed by biochemical studies.

FIGURE 2.

Putative PKA binding sites in LBA. A, Conservation of hydrophobic amino acids of putative PKA-binding sites in mLBA, hLBA, dLBA, and C. elegans LBA cLBA are shown by aligning with the known B1 and B2 PKA RII tethering sites (underlined) in DAKAP550 (a partial cDNA sequence for dLBA) along with other sequence in these regions (for the source of these sequences, see the legend of Fig. 1). B, Predicted secondary structure of the putative PKA binding sites in murine lBA (mLBAB1, mLBAB2). The hydrophobic amino acids on the hydrophobic side of the predicted amphipathic helices are boxed.

FIGURE 2.

Putative PKA binding sites in LBA. A, Conservation of hydrophobic amino acids of putative PKA-binding sites in mLBA, hLBA, dLBA, and C. elegans LBA cLBA are shown by aligning with the known B1 and B2 PKA RII tethering sites (underlined) in DAKAP550 (a partial cDNA sequence for dLBA) along with other sequence in these regions (for the source of these sequences, see the legend of Fig. 1). B, Predicted secondary structure of the putative PKA binding sites in murine lBA (mLBAB1, mLBAB2). The hydrophobic amino acids on the hydrophobic side of the predicted amphipathic helices are boxed.

Close modal
FIGURE 3.

Alignment of the C-terminal sequences of mLBA, hLBA, dLBA, CHS1, and human FAN (hFAN), which include the BEACH domains (in the middle, boxed), 5 WD repeats, and the 3 WDL repeats predicted in mLBA and hLBA. The predicted SH3, SH2 binding sites and tyrosine kinase recognition sites are also boxed. The C-terminal difference of the three isoforms of the mLBA is shown here (for detail see Fig. 1).

FIGURE 3.

Alignment of the C-terminal sequences of mLBA, hLBA, dLBA, CHS1, and human FAN (hFAN), which include the BEACH domains (in the middle, boxed), 5 WD repeats, and the 3 WDL repeats predicted in mLBA and hLBA. The predicted SH3, SH2 binding sites and tyrosine kinase recognition sites are also boxed. The C-terminal difference of the three isoforms of the mLBA is shown here (for detail see Fig. 1).

Close modal
Table I.

Protein homology between LBA and dLBA, mBG, and hFAN, showing the percentage of identity, positives, and gaps as well as the positions of each fragment in each gene

Fragments and PositionsIdentities (%)Positives (%)Length (aa)
mLBAdLBA
92–405 47–394 51 73 314 
405–959 601–1,160 55 75 555 
998–1,576 1,542–2,127 36 53 579 
1,793–2,856 2,642–3,727 56 74 1,064 
     
mLBA cLBA    
65–946 164–1,057 42 61 882 
1,300–1,571 1,065–1,333 39 59 271 
1,787–2,856 1,436–22,512 47 64 1,070 
     
mLBA hLBA    
1–2,856 1–2,863 85 88 2,856 
     
mLBA mBG    
1,934–2,839 1,460–2,335 27 43 906 
     
mLBA hFAN    
2,038–2,841 163–913 29 45 803 
Fragments and PositionsIdentities (%)Positives (%)Length (aa)
mLBAdLBA
92–405 47–394 51 73 314 
405–959 601–1,160 55 75 555 
998–1,576 1,542–2,127 36 53 579 
1,793–2,856 2,642–3,727 56 74 1,064 
     
mLBA cLBA    
65–946 164–1,057 42 61 882 
1,300–1,571 1,065–1,333 39 59 271 
1,787–2,856 1,436–22,512 47 64 1,070 
     
mLBA hLBA    
1–2,856 1–2,863 85 88 2,856 
     
mLBA mBG    
1,934–2,839 1,460–2,335 27 43 906 
     
mLBA hFAN    
2,038–2,841 163–913 29 45 803 

These lba orthologues also have a highly conserved long C-terminal region (∼1000 aa) shared with a group of proteins including CHS1/BG (32, 33), FAN (34), large volume sphere A (LVSA (35)) proteins (Fig. 2), and a number of anonymous ORFs. They may constitute a new gene family. The conserved region contains an unidentified region followed by one BEACH domain and several WD repeats. Several WD repeats were found in the unidentified region of homology in these genes when ∼1000 aa of C-terminal sequence were searched for WD repeats; however, no WD repeat is predicted when this region is analyzed alone (data not shown). Thus, we designate this region WD repeat-like domain (WDL). In aggregate, the entire WDL-BEACH-WD (WBW) structure may have a precise functional role because the WD repeats found in the WBW structures of different beige-like genes have a higher degree of homology with each other than with other WD repeats in proteins that lack a BEACH domain (Fig. 3). This homology analysis suggests the evolutionary conservation of the WBW structure in a gene family that includes lba, chs1/beige, FAN, lvsA, and other unidentified ORFs in GenBank. However, the BEACH domain can exist without WD motifs as in the case of lba-γ (Figs. 1 and 3). We show here that all BEACH domains have an SH3 binding site (consensus sequence PXXP), an SH2 binding site (consensus sequence YXXhy) (36), and a tyrosine kinase phosphorylation site (consensus sequence: (RK)-X(2, 3)-(DE)-X(2, 3)-Y) (37, 38, 39) (Fig. 3). These putative binding sites suggest that WBW proteins may interact with multiple signal transduction components.

Northern blot analysis indicates that a single ∼10-kb mRNA encoding the lba gene is present in LPS-induced J774 macrophages and 70Z/3 B cells (Fig. 4,A), as well as in other B cell lines (WEHI231, BCL1) and the macrophage cell line, RAW267.4 (data not shown). The size (∼10 kb) of the transcript is consistent with our cDNA sequence analysis (9903 bp for lba-α). The expression of the lba gene is significantly up-regulated in LPS-induced J774 macrophage cells given that the lba mRNA is nearly undetectable in J774 cells in the absence of LPS stimulation. The level of lba mRNA is increased by 3-fold in 70Z/3 B cells (Fig. 4 A) using β-actin mRNA as an internal standard. The up-regulation of lba expression in the B cell lines is entirely consistent with the FACS analysis of lacZ induction in the 7a65 gene trap cell clone in which lba gene was first identified (5).

FIGURE 4.

Expression of lba is inducible in B cells and macrophages. A, Northern blot hybridization of mRNA from B cell line 70Z/3 and the macrophage cell line J774. Both cell lines were cultured with or without LPS for 20 h. The poly(A)+ RNA was purified from these cells, run on a denaturing formaldehyde agarose gel and transferred to a Hybond-N+ filter. The filter was hybridized with the 2.5-kb probe that corresponds to the coding region of the lba gene including the BEACH and WD domains, as described in Materials and Methods. The hybridized filter was exposed to x-ray film for 24 h. Similar amounts of β-actin mRNA were found in all mRNA tested (actin panels). B, Expression of mRNA of the three lba isoforms in B cell lines (a) and tissues (b). Three isoform-specific primer pairs were used to detect the expression of the three isoforms by RT-PCR; the expected product size of the RT-PCR product for the α form is 1344 bp, for the β form it is 836 bp, and for the γ form it is 787 bp. Total RNA was analyzed. Aliquots (10 μl) of the PCR products were resolved on 0.8% agarose gels. Three independent experiments were performed and gave similar results.

FIGURE 4.

Expression of lba is inducible in B cells and macrophages. A, Northern blot hybridization of mRNA from B cell line 70Z/3 and the macrophage cell line J774. Both cell lines were cultured with or without LPS for 20 h. The poly(A)+ RNA was purified from these cells, run on a denaturing formaldehyde agarose gel and transferred to a Hybond-N+ filter. The filter was hybridized with the 2.5-kb probe that corresponds to the coding region of the lba gene including the BEACH and WD domains, as described in Materials and Methods. The hybridized filter was exposed to x-ray film for 24 h. Similar amounts of β-actin mRNA were found in all mRNA tested (actin panels). B, Expression of mRNA of the three lba isoforms in B cell lines (a) and tissues (b). Three isoform-specific primer pairs were used to detect the expression of the three isoforms by RT-PCR; the expected product size of the RT-PCR product for the α form is 1344 bp, for the β form it is 836 bp, and for the γ form it is 787 bp. Total RNA was analyzed. Aliquots (10 μl) of the PCR products were resolved on 0.8% agarose gels. Three independent experiments were performed and gave similar results.

Close modal

We also developed a multiplex RT-PCR assay that can simultaneously detect the expression of all three lba mRNA isoforms. RT-PCR analysis of lba mRNA (Fig. 4 B) shows that lba-β mRNA is expressed in all cell lines and tissues analyzed; however, lba-α mRNA is absent in 70Z/3, lung, and bone marrow and is less abundant in BAL17, A20, and spleen. The lba-γ mRNA is absent in 70Z/3 and bone marrow and is less abundant in spleen and lung, suggesting that these different isoforms may have discrete functions in different tissues.

All mutations in beige or chs1 genes result in truncated proteins that lack the BEACH and COOH-terminal WD repeats (40). We speculate that this region contains sequences critical to the function of chs1/beige and lba genes. In particular, the ability of their gene products to associate with intracellular vesicles to influence their trafficking may be lost in these truncated mutants. Therefore, we created a GFP fusion with the BEACH-WD region of lba called BW-GFP. Fluorescence microscopy (Fig. 5) of RAW 267.4 cells stably transfected with an expression vector encoding the BW-GFP fusion showed that the BW-GFP protein is present in the cytosol with rare cells showing a vesicular staining pattern in the absence of LPS stimulation (Fig. 5,A). However, this vesicular staining pattern is substantially increased in these cells after LPS stimulation (Fig. 5,B). Both the percentage of cells and the degree of vesicular staining in each cell are increased after LPS stimulation. RAW267.4 cells stably transfected with a GFP control construct show no change in their GFP fluorescence pattern upon LPS stimulation (Fig. 5 C).

FIGURE 5.

Subcellular localization of GFP-LBA fusion proteins revealed by UV fluorescence microscopy and laser scan confocal microscopy. A, RAW 267.4 macrophage cell line (R7) stably transfected with a BEACH-WD-GFP fusion construct. Most cells have diffuse, cytosolic GFP fluorescence, but some cells show vesicle association of the GFP fusion protein. B, The same cell line from A was plated on glass-covered plates and stimulated with LPS (100 ng/ml) for 24 h. Extensive vesicle association of the fusion protein was observed. C, RAW 267.4 macrophages stably transfected with the control vector pEGFP-N2 were cultured with 100 ng/ml LPS stimulation. No obvious vesicle association of native GFP is observed. ×400. D, Part of an R7 macrophage cell, showing GFP fluorescence. E, The same part of an R7 macrophage cell as in D, showing acidic lysosomes specifically labeled by LysoTracker Red in living cells. F, Lysosome colocalization (white part) of GFP fusion protein by overlapping pictures D and E; N, nucleus. G, R7 macrophage cells, showing GFP fluorescence. H, The same R7 macrophage cells as in G, showing prominent labeling of the Golgi complex (between the two nuclei) specifically labeled by BODIPY TR ceramide. Other intracellular membranes were weakly labeled. I, Golgi colocalization (white part) of GFP fusion protein by overlapping pictures G and H. Colocalization was determined by Zeiss LSM 510 software which allows for a reliability of 99% for actual pixels with both fluorophores. Colocalization mask pixels were converted to white color for obviousness. All cells were stimulated with LPS (100 ng/ml) for 24 h except for A.

FIGURE 5.

Subcellular localization of GFP-LBA fusion proteins revealed by UV fluorescence microscopy and laser scan confocal microscopy. A, RAW 267.4 macrophage cell line (R7) stably transfected with a BEACH-WD-GFP fusion construct. Most cells have diffuse, cytosolic GFP fluorescence, but some cells show vesicle association of the GFP fusion protein. B, The same cell line from A was plated on glass-covered plates and stimulated with LPS (100 ng/ml) for 24 h. Extensive vesicle association of the fusion protein was observed. C, RAW 267.4 macrophages stably transfected with the control vector pEGFP-N2 were cultured with 100 ng/ml LPS stimulation. No obvious vesicle association of native GFP is observed. ×400. D, Part of an R7 macrophage cell, showing GFP fluorescence. E, The same part of an R7 macrophage cell as in D, showing acidic lysosomes specifically labeled by LysoTracker Red in living cells. F, Lysosome colocalization (white part) of GFP fusion protein by overlapping pictures D and E; N, nucleus. G, R7 macrophage cells, showing GFP fluorescence. H, The same R7 macrophage cells as in G, showing prominent labeling of the Golgi complex (between the two nuclei) specifically labeled by BODIPY TR ceramide. Other intracellular membranes were weakly labeled. I, Golgi colocalization (white part) of GFP fusion protein by overlapping pictures G and H. Colocalization was determined by Zeiss LSM 510 software which allows for a reliability of 99% for actual pixels with both fluorophores. Colocalization mask pixels were converted to white color for obviousness. All cells were stimulated with LPS (100 ng/ml) for 24 h except for A.

Close modal

To determined to which vesicular compartments the BW-GFP fusion localizes, the RAW 264.7 cells stably transfected with the pBWEGFP construct stained with a lysosome-specific dye (Fig. 5,E) and trans-Golgi-specific dye (Fig. 5,H) were analyzed with confocal microscopy. The merged pictures show that some LBA-GFP proteins are colocalized with lysosomes (Fig. 5,F, white area) and colocalized with the trans-Golgi complex (Fig. 5 I, white perinuclear area).

We also performed immunogold labeling experiments that show that the LBA-GFP fusion protein can be found in association with the Golgi complex (Fig. 6,D), lysosomes (Fig. 6, B and F), ER (Fig. 6,C), plasma membrane (Fig. 6,E), perinuclear ER (Fig. 6,E), and endocytic vacuole (Fig. 6 A, because the gold particles are labeling a clathrin-coated endocytic vacuole, which indicates that it is involved in endocytosis and not exocytosis). The immunoelectron microscopy results agree well with our observations made by fluorescence microscopy and confocal fluorescence microscopy.

FIGURE 6.

Immunoelectron microscopy of LBA-GFP fusion protein. The LPS-stimulated R7 macrophage cells were fixed and processed for postembedding immunocytochemistry. The cells were dehydrated and embedded in gelatin capsules in LR White resin. Ultrathin sections of LR White-embedded cells were collected on nickel grids and immunolabeled with rabbit anti-GFP followed by labeling with anti-rabbit IgG-gold secondary Ab, finally stained with uranyl acetate and lead citrate before examination with immunoelectron microscopy. A, Clathrin-coated pit (endocytic, or coated vesicle) labeled with gold particles (white arrowhead). This is a vesicle forming on the cell surface. The fact that there is clathrin around this vacuole indicates that it is involved in endocytosis and not exocytosis. B, Intense labeling of a primary lysosome (white arrowhead) and a vesicle on the cell surface (black arrowhead). C, Black arrowheads show ribosomes lining a profile of ER. There are three gold particles labeling the ER (white arrowhead). The gray structure next to the ER is a mitochondrion (m) which is not labeled. D, Golgi region of a cell labeled for GFP. White arrowheads, gold particles on a Golgi cisterna. E, Labeling of ER comprising the perinuclear cisterna (white arrowheads), and labeling of the plasma membrane of the cell (black arrowheads). F, Gold particles surrounding a secondary lysosome in a cell (∗). At the top of the lysosome is a coated vesicle (black arrowhead) fusing with the lysosome. A portion of ER surrounds the bottom of the lysosome, which is also labeled with gold particles (white arrowhead). Labeling of the perimeter of the secondary lysosome shows routing of GFP from the cell surface to the lysosome-limiting membrane. e, Extracellular space; n, nucleus; g, Golgi; m, mitochondrion; c, cytoplasm. The size of the gold particles is 10 nm.

FIGURE 6.

Immunoelectron microscopy of LBA-GFP fusion protein. The LPS-stimulated R7 macrophage cells were fixed and processed for postembedding immunocytochemistry. The cells were dehydrated and embedded in gelatin capsules in LR White resin. Ultrathin sections of LR White-embedded cells were collected on nickel grids and immunolabeled with rabbit anti-GFP followed by labeling with anti-rabbit IgG-gold secondary Ab, finally stained with uranyl acetate and lead citrate before examination with immunoelectron microscopy. A, Clathrin-coated pit (endocytic, or coated vesicle) labeled with gold particles (white arrowhead). This is a vesicle forming on the cell surface. The fact that there is clathrin around this vacuole indicates that it is involved in endocytosis and not exocytosis. B, Intense labeling of a primary lysosome (white arrowhead) and a vesicle on the cell surface (black arrowhead). C, Black arrowheads show ribosomes lining a profile of ER. There are three gold particles labeling the ER (white arrowhead). The gray structure next to the ER is a mitochondrion (m) which is not labeled. D, Golgi region of a cell labeled for GFP. White arrowheads, gold particles on a Golgi cisterna. E, Labeling of ER comprising the perinuclear cisterna (white arrowheads), and labeling of the plasma membrane of the cell (black arrowheads). F, Gold particles surrounding a secondary lysosome in a cell (∗). At the top of the lysosome is a coated vesicle (black arrowhead) fusing with the lysosome. A portion of ER surrounds the bottom of the lysosome, which is also labeled with gold particles (white arrowhead). Labeling of the perimeter of the secondary lysosome shows routing of GFP from the cell surface to the lysosome-limiting membrane. e, Extracellular space; n, nucleus; g, Golgi; m, mitochondrion; c, cytoplasm. The size of the gold particles is 10 nm.

Close modal

We have identified and completed the initial characterization of the lba gene and found that it has three isoforms with different sequences at their C terminus. Northern blot experiments show that expression of lba is up-regulated 2- to 4-fold after LPS stimulation of B cells and macrophages. A homology search of GenBank reveals that lba gene has orthologues in C. elegans, Drosophila, mice, and humans and paralogues in diverse species ranging from yeast to human. These genes define a new protein family that we designate the WBW gene family because they share an evolutionarily conserved structure over a long protein sequence (∼1000 aa). Our analysis of subcellular localization with a BEACH-WD-GFP fusion protein provides the first direct evidence that the lba member of the WBW family can physically associate with various vesicular compartments in cells. Furthermore, we propose that the lba gene is also an AKAP, suggesting that WBW family proteins may have microtubule- and PKA-binding properties like AKAPs (19). Studies of FAN suggest that WBW proteins can bind to cytoplasmic tails of activated receptors via their WD repeats (34).

Previous evidence suggests that WBW proteins are involved in intracellular vesicle trafficking. For example, the strikingly enlarged vesicles in beige/CHS cells occur in membrane-bound organelles. The CHS1/BG protein has a similar modular architecture to the VPS15 and Huntington proteins that are associated with the membrane fraction (20) and the lvsA gene that is essential for cytokinesis (35), a process that also involves fusion of intracellular vesicles (41, 42). FAN may also be involved in vesicle trafficking because FAN-deficient mice, after cutaneous barrier disruption, have delayed kinetics of skin recovery that requires secretion of vesicles (43, 44). However, there is no direct evidence that these WBW proteins directly associate with vesicles. In contrast, others found unexpectedly by Western blot that the BG, LVSA, and DAKAP550 proteins are present in the cytosolic fraction of cells and not in the membrane fraction (35, 45) or cytoskeleton (21). This paradox can be explained by hypothesizing that these proteins are not constitutively associated with vesicles, but rather associate with vesicles under certain conditions like LPS stimulation. This hypothesis agrees well with our observation that an LBA-GFP fusion protein is located in the cytosol; however, it becomes associated with vesicles after activation of the cells by LPS stimulation. Confocal microscopy also shows this fusion protein colocalizes with the trans-Golgi and lysosomes. Immunoelectron microscopy further demonstrated that it is also localized to ER and the plasma membrane as well as the trans-Golgi complex and lysosomes. Therefore, our experiments clearly show that the BEACH-WD-GFP fusion protein is associated with the vesicular system . This may be true for the intact LBA protein as well as for other WBW proteins like CHS1/BG, LVSA, and FAN, because they share high homology with the region in mouse lba that we used for the GFP fusion experiment. However, this must be determined experimentally. Our activation-triggered vesicle trafficking hypothesis is further supported by the following: 1) BEACH domain contains a tyrosine phosphorylation site; 2) the WD repeats binding site of FAN contains a serine residue (34), and it is possible that this serine is a target of serine kinases, because some experiments suggest that the WD repeats binding requires phosphorylation of the WD binding sites (46); and 3) mitogen-activated protein kinase was suggested to control the movement of lytic granules of NK cells (47). Potentially, WBW protein functions are activated by tyrosine and/or serine/threonine kinases after stimulation by agents like LPS. Although the GFP fusion experiment that we describe does not demonstrate that the BEACH domain and/or the WD repeats in LBA directly associate with intracellular vesicles, we propose that the BEACH domain binds to vesicles whereas the WD repeat domains bind to a membrane-associated protein. We propose that because BEACH domains and WD repeats exist separately in some proteins, they have separate functions. For instance, the WD repeats of the FAN protein bind to the cytoplasmic tail of the TNFR55 receptor independent of the BEACH domain (34). The FAN gene is made up almost entirely of the sequence in the highly conserved WBW structure (Fig. 3); therefore, other WBW-containing proteins may act like FAN and bind the cytoplasmic tails of TNFR55 or TNFR55-like receptors.

Another provocative finding is that lba is a potential AKAP. The recently completed genomic sequence of Drosophila indicates that lba has an orthologue in Drosophila (DAKAP550) that is capable of binding to protein kinase A (21). The DAKAP550 gene is expressed in all tissues throughout development and is the principal A kinase anchor protein in adult flies. It is enriched in secretory tissues such as neurons and salivary glands and is found concentrated in the apical cytoplasm of some cells (21), in agreement with the proposed function in secretion for lba. Although the B1 and B2 RII-binding sites of DAKAP550 are not present in murine LBA (mLBA), hLBA, and cLBA, we do find two sequences that are very similar to the B1 and B2 RII-binding sites in all lba orthologues. The two sequences are predicted to form two adjacent amphipathic helices characteristic of PKA-binding sites, satisfying the requirement of the hydrophobic interaction mechanism of RII peptide binding to the RII subunits of PKA revealed recently (18). Thus, lba may serve as an AKAP that is involved in cAMP-mediated signaling secretory processes by translocating PKA to specific membrane sites. This translocation may require microtubule binding as suggested by the recent finding that another WBW protein, human CHS1, can associate with microtubules (15). On the basis of these findings, we propose a two-signal model for the function of the WBW protein family using the lba gene as a prototype. LBA is constitutively associated with PKA like other AKAPs and after LPS stimulation (signal 1), the BEACH domain is phosphorylated. This enables the LBA-PKA complex to bind to intracellular vesicles and tether vesicles to microtubules for transport to the plasma membrane. At the membrane a second signal is required that generates cAMP. Binding of locally generated cAMP to the LBA/PKA complex releases PKA, allowing it to phosphorylate cytoplasmic tails of activated receptors to enable binding of LBA via its WD repeats. This final step would result in vesicle fusion with the plasma membrane (Fig. 7). Many immune processes need a second signal such as in the case of costimulators. We speculate that a first signal activates an immune cell to transport enough vesicles to the plasma membrane area that contact another cell. A second signal generated by the contact with the target cell produces cAMP that stimulates PKA activity resulting in membrane fusion of vesicles. Thus, LBA and other WBW proteins may provide a means for eukaryotic cells to direct the fusion of membrane-bound vesicles in a polarized fashion, in coordination with signal transduction complexes at the plasma membrane as is required of many different effector cell types in the immune system (48).

FIGURE 7.

Model of vesicle secretion for WBW protein family using the lba gene as a prototype. After immune cell activation, the BEACH domain binds to vesicles containing cargo proteins and membrane proteins for secretion or deposition in the plasma membrane. The anchor domain binds to microtubules to move the vesicles to the membrane where the WD domain binds to a phosphorylated sequences of membrane receptor complexes to mediate the fusion of the vesicles with the membrane, thus releasing the cargo proteins or depositing membrane proteins on the plasma membrane of immune cells. SP, phosphorylated serine residue; S, serine residue; C, catalytic subunit of PKA.

FIGURE 7.

Model of vesicle secretion for WBW protein family using the lba gene as a prototype. After immune cell activation, the BEACH domain binds to vesicles containing cargo proteins and membrane proteins for secretion or deposition in the plasma membrane. The anchor domain binds to microtubules to move the vesicles to the membrane where the WD domain binds to a phosphorylated sequences of membrane receptor complexes to mediate the fusion of the vesicles with the membrane, thus releasing the cargo proteins or depositing membrane proteins on the plasma membrane of immune cells. SP, phosphorylated serine residue; S, serine residue; C, catalytic subunit of PKA.

Close modal

Increasing evidence suggests that all clinical symptoms of CHS/beige patients could be explained by a secretion malfunction. The cytolytic proteins (granzymes A/B and perforin) in CHS CTL are expressed normally but are not secreted on stimulation (11). Secretion of other enzymes are also defective in macrophages and neutrophils (49), as are the membrane deposition of class II molecules (15) and CTL-4 (17). However, there is a dispute over whether giant lysosomes in beige/CHS disease are a result of abnormalities in the fusion or fission of lysosomes. (11, 17, 45, 50, 51, 52). How the secretion pathway is impaired is unclear. Our characterization of the lba gene and our model for its function may provide a molecular explanation for these two major cellular dysfunctions of CHS/beige, giant vesicles and secretion malfunction. In our opinion, vesicles may require association with the BEACH domain of CHS1 for fission and/or movement to the plasma membrane. After reaching the plasma membrane, they then require recognition of certain membrane proteins by the WD repeats to mediate fusion with the plasma membrane. This requires CHS1 proteins to be full length for proper function because the WD repeats are at the C terminus. Thus, truncated beige/CHS protein molecules (or perhaps LBA proteins) that lack the C-terminal WD repeats would be expected to cause disease (40). The giant lysosomes in the affected cells may come from the failure of vesicle movement and/or fusion with the membrane. Similar disorders of beige/CHS have also been described in mink, cattle, cats, and killer whales. Given the structural similarity of the WBW gene family, we propose that the genetic mutations in these species also involve other WBW genes. There are also other lysosomal trafficking mutants in mice with phenotypes similar to beige that may also involve mutation of other WBW gene family members.

In summary, we demonstrate the existence of a novel gene family, the WBW family, which includes the lba gene that can: 1) associate with the vesicular system, including the Golgi complex, lysosomes, ER, plasma membrane, and perinuclear ER; 2) is LPS inducible; 3) is potential AKAP; and 4) has three different isoforms that differ in WD repeat number. These findings suggest an important role for lba in coupling signal transduction and vesicle trafficking to enable polarized secretion and/or membrane deposition of immune effector molecules. We hope that this work will further the understanding of the mechanism of CHS and other related diseases as well as general immune cell function.

We thank Maria Lemos for technical assistance, Dr. Nikola Valkov for excellent confocal help, and the other members of the Kerr laboratory for their support and discussion of the manuscript.

1

This work was supported by grants from the National Institutes of Health to W.G.K. and by academic development funds from H. Lee Moffit Cancer Center.

3

Abbreviations used in this paper: CHS, Chediak-Higashi syndrome; RACE, rapid amplification of cDNA end; AKAP, A kinase anchor protein; ORF, open reading frame; PKA, protein kinase A; FAN, factor associated with neutral sphingomyelinase activation; GFP, green-fluorescent protein; BEACH domain, beige and CHS domain; LSVA, large volume sphere A; WDL, WD repeat-like domain; LBA, LPS-responsive, beige-like anchor; ER, endoplasmic reticulum; dLBA, Drosophila LBA; mlBA, murine lBA; hLBA, human LBA; cLBA, C. elegans LBA.

1
Harris, M. R., C. S. Kindle, A. F. Abruzzini, C. W. Pierce, S. E. Cullen.
1984
. Antigen presentation by the BCL1 murine B cell line: in vitro stimulation by LPS.
J. Immunol.
133
:
1202
2
Tobias, P. S., J. Gegner, J. Han, T. Kirkland, V. Kravchenko, D. Leturcq, J. D. Lee, A. Moriarty, J. C. Mathison, J. Pugin, et al
1994
. LPS binding protein and CD14 in the LPS dependent activation of cells.
Prog. Clin. Biol. Res.
388
:
31
3
Inazawa, M., C. Shiozawa, R. Fotedar, E. Diener.
1985
. How relevant are growth and maturation factors to the B lymphocyte response induced by LPS?.
Lymphokine Res.
4
:
343
4
Kerr, W. G., G. P. Nolan, A. T. Serafini, L. A. Herzenberg.
1989
. Transcriptionally defective retroviruses containing lacZ for the in situ detection of endogenous genes and developmentally regulated chromatin.
Cold Spring Harbor Symp. Quant. Biol.
54
:
767
5
Kerr, W. G., M. Heller, L. A. Herzenberg.
1996
. Analysis of lipopolysaccharide-response genes in B-lineage cells demonstrates that they can have differentiation stage-restricted expression and contain SH2 domains.
Proc. Natl. Acad. Sci. USA
93
:
3947
6
Huber, M., C. D. Helgason, J. E. Damen, M. Scheid, V. Duronio, L. Liu, M. D. Ware, R. K. Humphries, G. Krystal.
1999
. The role of SHIP in growth factor induced signalling.
Prog. Biophys. Mol. Biol.
71
:
423
7
Ono, M., S. Bolland, P. Tempst, J. V. Ravetch.
1996
. Role of the inositol phosphatase SHIP in negative regulation of the immune system by the receptor Fc(γ)RIIB.
Nature
383
:
263
8
Ono, M., H. Okada, S. Bolland, S. Yanagi, T. Kurosaki, J. V. Ravetch.
1997
. Deletion of SHIP or SHP-1 reveals two distinct pathways for inhibitory signaling.
Cell
90
:
293
9
Spritz, R. A..
1998
. Genetic defects in Chediak-Higashi syndrome and the beige mouse.
J. Clin. Immunol.
18
:
97
10
Abo, T., J. C. Roder, W. Abo, M. D. Cooper, C. M. Balch.
1982
. Natural killer (HNK-1+) cells in Chediak-Higashi patients are present in normal numbers but are abnormal in function and morphology.
J. Clin. Invest.
70
:
193
11
Baetz, K., S. Isaaz, G. M. Griffiths.
1995
. Loss of cytotoxic T lymphocyte function in Chediak-Higashi syndrome arises from a secretory defect that prevents lytic granule exocytosis.
J. Immunol.
154
:
6122
12
Haliotis, T., J. Roder, M. Klein, J. Ortaldo, A. S. Fauci, R. B. Herberman.
1980
. Chediak-Higashi gene in humans I. Impairment of natural-killer function.
J. Exp. Med.
151
:
1039
13
Clark, R. A., H. R. Kimball.
1971
. Defective granulocyte chemotaxis in the Chediak-Higashi syndrome.
J. Clin. Invest.
50
:
2645
14
Takeuchi, K., H. Wood, R. T. Swank.
1986
. Lysosomal elastase and cathepsin G in beige mice. Neutrophils of beige (Chediak-Higashi) mice selectively lack lysosomal elastase and cathepsin G.
J. Exp. Med.
163
:
665
15
Faigle, W., G. Raposo, D. Tenza, V. Pinet, A. B. Vogt, H. Kropshofer, A. Fischer, G. de Saint-Basile, S. Amigorena.
1998
. Deficient peptide loading and MHC class II endosomal sorting in a human genetic immunodeficiency disease: the Chediak-Higashi syndrome.
J. Cell Biol.
141
:
1121
16
Lem, L., D. A. Riethof, M. Scidmore-Carlson, G. M. Griffiths, T. Hackstadt, F. M. Brodsky.
1999
. Enhanced interaction of HLA-DM with HLA-DR in enlarged vacuoles of hereditary and infectious lysosomal diseases.
J. Immunol.
162
:
523
17
Barrat, F. J., F. Le Deist, M. Benkerrou, P. Bousso, J. Feldmann, A. Fischer, G. de Saint Basile.
1999
. Defective CTLA-4 cycling pathway in Chediak-Higashi syndrome: a possible mechanism for deregulation of T lymphocyte activation.
Proc. Natl. Acad. Sci. USA
96
:
8645
18
Newlon, M. G., M. Roy, D. Morikis, Z. E. Hausken, V. Coghlan, J. D. Scott, P. A. Jennings.
1999
. The molecular basis for protein kinase A anchoring revealed by solution NMR.
Nat. Struct. Biol.
6
:
222
19
Colledge, M., J. D. Scott.
1999
. AKAPs: from structure to function.
Trends Cell Biol.
9
:
216
20
Nagle, D. L., M. A. Karim, E. A. Woolf, L. Holmgren, P. Bork, D. J. Misumi, S. H. McGrail, B. J. Dussault, Jr, C. M. Perou, R. E. Boissy, et al
1996
. Identification and mutation analysis of the complete gene for Chediak-Higashi syndrome.
Nat. Genet.
14
:
307
21
Han, J. D., N. E. Baker, C. S. Rubin.
1997
. Molecular characterization of a novel A kinase anchor protein from Drosophila melanogaster.
J. Biol. Chem.
272
:
26611
22
Feuchter, A. E., J. D. Freeman, D. L. Mager.
1992
. Strategy for detecting cellular transcripts promoted by human endogenous long terminal repeats: identification of a novel gene (CDC4L) with homology to yeast CDC4.
Genomics
13
:
1237
23
Altschul, S. F., E. V. Koonin.
1998
. Iterated profile searches with PSI-BLAST: a tool for discovery in protein databases.
Trends Biochem. Sci.
23
:
444
24
Schultz, J., R. R. Copley, T. Doerks, C. P. Ponting, P. Bork.
2000
. SMART: a web-based tool for the study of genetically mobile domains.
Nucleic Acids Res.
28
:
231
25
Thompson, J. D., J. B. Benjamin, and J. A. Szivek. 1997. Pullout strengths of cannulated and noncannulated cancellous bone screws. Clin. Orthop. Rel. Res. :241.
26
Neer, E. J., T. F. Smith.
1996
. G protein heterodimers: new structures propel new questions.
Cell
84
:
175
27
Garcia-Higuera, I., J. Fenoglio, Y. Li, C. Lewis, M. P. Panchenko, O. Reiner, T. F. Smith, E. J. Neer.
1996
. Folding of proteins with WD-repeats: comparison of six members of the WD-repeat superfamily to the G protein β subunit.
Biochemistry
35
:
13985
28
Neer, E. J., C. J. Schmidt, R. Nambudripad, and T. F. Smith. 1994. The ancient regulatory-protein family of WD-repeat proteins [Published erratum appears in Nature 1994;371:812.]. Nature 371:297.
29
Smith, T. F., C. Gaitatzes, K. Saxena, E. J. Neer.
1999
. The WD repeat: a common architecture for diverse functions.
Trends Biochem. Sci.
24
:
181
30
Neer, E. J., T. F. Smith.
2000
. A groovy new structure.
Proc. Natl. Acad. Sci. USA
97
:
960
31
Haller, E. M., S. A. Shelley, M. R. Montgomery, J. U. Balis.
1992
. Immunocytochemical localization of lysozyme and surfactant protein A in rat type II cells and extracellular surfactant forms.
J. Histochem. Cytochem.
40
:
1491
32
Perou, C. M., K. J. Moore, D. L. Nagle, D. J. Misumi, E. A. Woolf, S. H. McGrail, L. Holmgren, T. H. Brody, B. J. Dussault, Jr, C. A. Monroe, et al
1996
. Identification of the murine beige gene by YAC complementation and positional cloning.
Nat. Genet.
13
:
303
33
Kingsmore, S. F., M. D. Barbosa, V. T. Tchernev, J. C. Detter, A. C. Lossie, M. F. Seldin, R. F. Holcombe.
1996
. Positional cloning of the Chediak-Higashi syndrome gene: genetic mapping of the beige locus on mouse chromosome 13.
J. Invest. Med.
44
:
454
34
Adam-Klages, S., D. Adam, K. Wiegmann, S. Struve, W. Kolanus, J. Schneider-Mergener, M. Kronke.
1996
. FAN, a novel WD-repeat protein, couples the p55 TNF-receptor to neutral sphingomyelinase.
Cell
86
:
937
35
Kwak, E., N. Gerald, D. A. Larochelle, K. K. Vithalani, M. L. Niswonger, M. Maready, A. De Lozanne.
1999
. LvsA, a protein related to the mouse beige protein, is required for cytokinesis in Dictyostelium.
Mol. Biol. Cell
10
:
4429
36
Pawson, T., J. D. Scott.
1997
. Signaling through scaffold, anchoring, and adaptor proteins.
Science
278
:
2075
37
Patschinsky, T., T. Hunter, F. S. Esch, J. A. Cooper, B. M. Sefton.
1982
. Analysis of the sequence of amino acids surrounding sites of tyrosine phosphorylation.
Proc. Natl. Acad. Sci. USA
79
:
973
38
Hunter, T..
1982
. Synthetic peptide substrates for a tyrosine protein kinase.
J. Biol. Chem.
257
:
4843
39
Cooper, J. A., F. S. Esch, S. S. Taylor, T. Hunter.
1984
. Phosphorylation sites in enolase and lactate dehydrogenase utilized by tyrosine protein kinases in vivo and in vitro.
J. Biol. Chem.
259
:
7835
40
Certain, S., F. Barrat, E. Pastural, F. L. Deist, J. Goyo-Rivas, N. Jabado, M. Benkerrou, R. Seger, E. Vilmer, G. Beullier, et al
2000
. Protein truncation test of LYST reveals heterogenous mutations in patients with Chediak-Higashi syndrome.
Blood
95
:
979
41
Jantsch-Plunger, V., M. Glotzer.
1999
. Depletion of syntaxins in the early Caenorhabditis elegans embryo reveals a role for membrane fusion events in cytokinesis.
Curr. Biol.
9
:
738
42
Heese, M., U. Mayer, G. Jurgens.
1998
. Cytokinesis in flowering plants: cellular process and developmental integration.
Curr. Opin. Plant Biol.
1
:
486
43
Kreder, D., O. Krut, S. Adam-Klages, K. Wiegmann, G. Scherer, T. Plitz, J. M. Jensen, E. Proksch, J. Steinmann, K. Pfeffer, M. Kronke.
1999
. Impaired neutral sphingomyelinase activation and cutaneous barrier repair in FAN-deficient mice.
EMBO J.
18
:
2472
44
Elias, P. M..
1983
. Epidermal lipids, barrier function, and desquamation.
J. Invest. Dermatol.
80
:
44s
45
Perou, C. M., J. D. Leslie, W. Green, L. Li, D. M. Ward, J. Kaplan.
1997
. The Beige/Chediak-Higashi syndrome gene encodes a widely expressed cytosolic protein.
J. Biol. Chem.
272
:
29790
46
Skowyra, D., K. L. Craig, M. Tyers, S. J. Elledge, J. W. Harper.
1997
. F-box proteins are receptors that recruit phosphorylated substrates to the SCF ubiquitin-ligase complex.
Cell
91
:
209
47
Wei, S., A. M. Gamero, J. H. Liu, A. A. Daulton, N. I. Valkov, J. A. Trapani, A. C. Larner, M. J. Weber, J. Y. Djeu.
1998
. Control of lytic function by mitogen-activated protein kinase/extracellular regulatory kinase 2 (ERK2) in a human natural killer cell line: identification of perforin and granzyme B mobilization by functional ERK2.
J. Exp. Med.
187
:
1753
48
Stinchcombe, J. C., G. M. Griffiths.
1999
. Regulated secretion from hemopoietic cells.
J. Cell Biol.
147
:
1
49
Barak, Y., E. Nir.
1987
. Chediak-Higashi syndrome.
Am. J. Pediatr. Hematol.-Oncol.
9
:
42
50
Cervero, C., B. Heinrichs, J. Villarrubia, J. L. Velasco, M. T. Ferro, J. Lopez, L. Escribano.
1994
. [Chediak-Higashi-like inclusions in acute myeloblastic leukemia: ultrastructural study].
Sangre
39
:
135
51
Barbosa, M. D., Q. A. Nguyen, V. T. Tchernev, J. A. Ashley, J. C. Detter, S. M. Blaydes, S. J. Brandt, D. Chotai, C. Hodgman, R. C. Solari, M. Lovett, and S. F. Kingsmore. 1996. Identification of the homologous beige and Chediak-Higashi syndrome genes. [Published erratum appears in 1997 Nature 385:97.] Nature382:262.
52
Menard, M., K. M. Meyers.
1988
. Storage pool deficiency in cattle with the Chediak-Higashi syndrome results from an absence of dense granule precursors in their megakaryocytes.
Blood
72
:
1726