transporters Differential expression of putative ...

Differential expression of putative transbilayer amphipath transporters

MARGARET S. HALLECK, JOSEPH F. LAWLER, JR., SETH BLACKSHAW, LING GAO, PRIYA NAGARAJAN, COLEEN HACKER, SCOTT PYLE, JASON T. NEWMAN, YOSHINOBU NAKANISHI, HIROSHI ANDO, DANIEL WEINSTOCK, PATRICK WILLIAMSON and ROBERT A. SCHLEGEL Physiol. Genomics 1:139-150, 1999. ; You might find this additional info useful... This article has been cited by 16 other HighWire-hosted articles: http://physiolgenomics.physiology.org/content/1/3/139#cited-by Updated information and services including high resolution figures, can be found at: http://physiolgenomics.physiology.org/content/1/3/139.full

This information is current as of June 12, 2013.

Physiological Genomics publishes results of a wide variety of studies from human and from informative model systems with techniques linking genes and pathways to physiology, from prokaryotes to eukaryotes. It is published 24 times a year (twice monthly) by the American Physiological Society, 9650 Rockville Pike, Bethesda MD 20814-3991. Copyright © 1999 the American Physiological Society. ESSN: 1531-2267. Visit our website at http://www.the-aps.org/.

Downloaded from http://physiolgenomics.physiology.org/ by guest on June 12, 2013

Additional material and information about Physiological Genomics can be found at: http://www.the-aps.org/publications/physiolgenomics

Differential expression of putative transbilayer amphipath transporters MARGARET S. HALLECK,1 JOSEPH F. LAWLER, JR.,2 SETH BLACKSHAW,2 LING GAO,1 PRIYA NAGARAJAN,1 COLEEN HACKER,1 SCOTT PYLE,1 JASON T. NEWMAN,1 YOSHINOBU NAKANISHI,3 HIROSHI ANDO,3 DANIEL WEINSTOCK,4 PATRICK WILLIAMSON,5 AND ROBERT A. SCHLEGEL1 1Department of Biochemistry and Molecular Biology and 4Department of Veterinary Science, Penn State University, University Park, Pennsylvania 16802; 2Department of Neuroscience, Johns Hopkins University School of Medicine, Baltimore, Maryland 21205; 3Department of Pharmaceutical Science, Graduate School of Natural Science and Technology, Kanazawa University, Kanazawa, Ishikawa 920–0934, Japan; and 5Department of Biology, Amherst College, Amherst, Massachusetts 01002

in situ hybridization; P-type ATPase; aminophospholipid translocase; cholestasis; central nervous system

THE P-TYPE ADENOSINETRIPHOSPHATASES (P-type ATPases), named for the phosphorylated intermediate state of the enzyme, are a family of proteins that use the free energy of ATP hydrolysis to drive uphill transport of ions across membranes (22). Until recently, two subfamilies of the enzyme were generally recognized, one of which transports heavy metal ions such as Cu2⫹ or Cd2⫹ across a bilayer and a larger subfamily that transports non-heavy metal ions such as H⫹, Na⫹, K⫹, or Ca2⫹. Recently a third subfamily was defined on the basis of sequence similarity relations among members of the new subfamily, sequence motifs restricted to the sub-

Received 15 June 1999; accepted in final form 16 September 1999. Article published online before print. See web site for date of publication (http://physiolgenomics.physiology.org).

family, and transport of substrates other than metal ions (1, 5, 10, 27). Representatives of the subfamily are widespread and are found in yeasts, slime molds, nematodes, insects, mammals, plants, and protozoan parasites. The first cloned mammalian subfamily member and its yeast homolog transport aminophospholipids from the outer to the inner leaflet of the plasma membrane (27). The resulting asymmetric transbilayer distribution of the aminophospholipid, phosphatidylserine (PS), is a hallmark of normal cells. The generality of this PS asymmetry was convincingly demonstrated by Van den Eijnde et al. (28), who injected biotinylated annexin V (a PS-specific probe) intracardially into viable mouse embryos that were subsequently fixed, sectioned, and stained with horseradish peroxidaseavidin to identify cells in which PS was present on the cell surface in vivo. Nonapoptotic cells of all types were generally unlabeled, indicating that their PS was not exposed on the outer leaflet of the plasma membrane. In contrast, however, apoptotic cells derived from all germ layers were labeled in all tissues. These elegant studies firmly established that PS is sequestered to the inner leaflet in nearly all living cells and that loss of this asymmetry is a universal marker of cells undergoing apoptosis. Because PS expressed on the cell surface serves as a recognition signal for phagocytosis of apoptotic cells by macrophages (7, 14, 23, 24, 29), the physiological importance of the aminophospholipid transport activity in mammals is clear: it prevents inappropriate exposure of PS on the cell surface that could well lead to destruction of the cell presenting it. A second mammalian subfamily member was identified as the gene mutated in two types of familial inherited cholestasis. Cholestasis, or impaired bile flow, is the central manifestation of two clinically distinct, autosomal-recessive liver disorders, benign recurrent intrahepatic cholestasis (BRIC or Summerskill syndrome) and progressive familial intrahepatic cholestasis type 1 (PFIC1 or Byler disease) (4). By positional cloning, a single gene responsible for both of these diseases, FIC1, was identified; this gene is a member of the new subfamily of P-type ATPases, demonstrating that disrupting the function of one of the genes in this

1094-8341/99 $5.00 Copyright r 1999 the American Physiological Society

139


Halleck, Margaret S., Joseph F. Lawler, Jr., Seth Blackshaw, Ling Gao, Priya Nagarajan, Coleen Hacker, Scott Pyle, Jason T. Newman, Yoshinobu Nakanishi, Hiroshi Ando, Daniel Weinstock, Patrick Williamson, and Robert A. Schlegel. Differential expression of putative transbilayer amphipath transporters. Physiol. Genomics 1: 139–150, 1999.—The aminophospholipid translocase transports phosphatidylserine and phosphatidylethanolamine from one side of a bilayer to another. Cloning of the gene encoding the enzyme identified a new subfamily of P-type ATPases, proposed to be amphipath transporters. As reported here, mammals express as many as 17 different genes from this subfamily. Phylogenetic analysis reveals the genes to be grouped into several distinct classes and subclasses. To gain information on the functions represented by these groups, Northern analysis and in situ hybridization were used to examine the pattern of expression of a panel of subfamily members in the mouse. The genes are differentially expressed in the respiratory, digestive, and urogenital systems, endocrine organs, the eye, teeth, and thymus. With one exception, all of the genes are highly expressed in the central nervous system (CNS); however, the pattern of expression within the CNS differs substantially from gene to gene. These results suggest that the genes are expressed in a tissue-specific manner, are not simply redundant, and may represent isoforms that transport a variety of different amphipaths.

140

AMPHIPATH TRANSPORTER EXPRESSION

MATERIALS AND METHODS

Cloning of Subfamily Members Ib, Ih, IIa, IIb, and Va Ib. A 321-bp probe generated by PCR from a murine EST clone (W18129) containing the 38 end of the gene was used to screen a mouse brain cDNA library constructed in Lambda ZAPII (Stratagene). Three subsequent library screens using PCR probes generated from the most 58 sequence of each successive new clone resulted in three overlapping clones containing the entire coding sequence of Ib (GenBank AF156550). Ih. A 611-bp Stu I restriction fragment from a mouse EST clone (AA759581) containing the 58 end of the gene was used to screen the mouse brain library resulting in an overlapping clone containing 480 bp of new 38 sequence. Rescreening the mouse brain library with a 404-bp probe generated by PCR from the most 38 end of the new clone produced no new sequence. Using the same probe to screen a PCC4 mouse teratocarcinoma cDNA library constructed in Lambda ZapII (Stratagene) resulted in a clone containing 1,300 bp of new 38 sequence. A 484-bp PCR probe was generated from the most 38 sequence of this clone and was used to rescreen the PCC4 library, resulting in a clone containing 470 bp of new 38 sequence. Finally, a 437-bp PCR probe generated from the most 38 sequence of the previous clone was used to rescreen the PCC4 library, resulting in 518 bp of new sequence, including the 38 end of the gene. The sequence provided by these clones was submitted to GenBank (AF156551). IIa. An 858-bp probe generated by PCR from a rat EST clone (W75163) was used to screen the mouse brain library, resulting in a clone containing the 38 end of the gene. A 172-bp probe generated by PCR from the same EST clone was used to screen the library again, resulting in an overlapping clone that contained the remainder of the IIa gene (GenBank AF152243).

IIb. A 900-bp Spe I and Bsu36 I restriction fragment from a human EST clone (R51412) was used to screen the mouse brain library, resulting in a clone containing 600 bp of homologous murine sequence. A 343-bp PCR probe generated from the most 58 sequence of the clone was used to rescreen the library, resulting in three overlapping clones containing 1,553 bp of new sequence, including the 38 end of the gene. A 356-bp PCR probe generated from the most 58 end of the new sequence was used to rescreen the library, resulting in a clone containing 278 bp of new 58 sequence. A 364-bp PCR probe designed from the most 58 end of this new sequence provided 753 bp of new 58 sequence when the library was screened again. Finally, a 326-bp PCR probe generated from this sequence was used to rescreen the library, resulting in 104 bp of new sequence and the 58 end of the gene. The complete sequence provided by these clones was submitted to GenBank (AF155913). Va. A 240-bp probe generated by PCR from a murine expressed sequence tag (EST) clone (AA116479, AF011337) was used to screen the PCC4 library, resulting in two overlapping clones that contained the 58 end of the gene, as well as 1,347 bp of new 38 sequence overlapping another mouse EST clone (AA981599) containing 460 bp of further 38 sequence. A 385-bp probe generated by PCR from the EST clone was used to rescreen the PCC4 library, resulting in a clone containing 1,783 bp of new sequence and the 38 end of the gene. The sequence provided by these clones was submitted to GenBank (AF156549). Sequencing EST Clones EST clones AA437803 and AA754832 (Genome Systems) and AA639797 (Research Genetics) were sequenced in their entirety and reentered into the database (GenBank AF156546, AF156547, and AF156548, respectively). Sequence Comparisons and Tree Construction TBLASTN (2) was used to search the databases; alignments were generated using the CLUSTALW alignment available at the Baylor College of Medicine at http:// dot.imgen.bcm.tmc.edu:9331/multi-align/multi-align. html; sequence similarities were calculated using either the GAP program of the GCG package or BLAST. NeighborJoining trees were generated from CLUSTALW alignments using the MEGA program (15). Chromosomal Localizations Ia was previously localized on human chromosomes using a 765-bp fragment from human Ia cDNA as a probe for in situ hybridization (17). An 881-bp PCR product from Ib (1385– 2266) was used to screen a BAC human genomic library to obtain a clone that was used as a probe to map the Ib gene by standard fluorescence in situ hybridization techniques (11, 16). Ic (FIC1) was previously mapped using a genome search for shared segments in chromosomes from patients with familial intrahepatic cholestasis (12). Chromosomal localization for Ik was provided by the GenBank entry for AC004755 containing genomic DNA sequence from human chromosome 19. Localization of Ir to chromosome 3, interval D3S1553D3S1580, was determined by entering KIAA0956 (from the GenBank entry for AB023173) into a New Gene Map of the Human Genome from the International RH Mapping Consortium at http://www.ncbi.nlm.nih.gov/genemap/. The localization of Is to chromosome 13 was provided by GenBank entry AB028944. Localization of IIa to chromosome 20 was provided by Ishikawa et al. (13); localization to interval

http://physiolgenomics.physiology.org


new subfamily has clinical consequences. However, it is not clear whether the product of this gene transports aminophospholipids or whether it transports some other amphipathic molecule required for bile secretion, such as bile acids. In the case of the non-heavy metal ion transporters, the enzymes fall into several classes that differ in the ion transported and/or the membrane across which transport occurs. For example, one class transports Ca2⫹ across the plasma membrane, another transports Ca2⫹ across intracellular membranes such as the sarcoplasmic reticulum, and another transports Na⫹/K⫹ across the plasma membrane. Genes in the new subfamily of amphipath transporters also fall into several classes based on sequence similarities (10). If the functional basis for these different classes of enzymes is the same as for the non-heavy metal ion transporters, all of the genes in the class that contains the original aminophospholipid translocase gene may code for plasma membrane aminophospholipid transporters, but each may be expressed in a different tissue or at a different time in development. The other classes might then represent aminophospholipid transporters expressed in the membranes of other organelles or might transport amphipathic molecules other than aminophospholipids. To gain information that might allow distinction of these possibilities, we have examined the spatial and temporal expression pattern of a panel of subfamily members in the mouse.


141

D20S196-D20S183 on the chromosome 20 map was determined by entering KIAA00611 (13) into the Gene Map as described above. The chromosomal localization for IIb was previously described (10). Vb was localized by Nagase et al. (19) to chromosome 5 and to the interval D5S1955-D5S496 on the chromosome 5 map by entering KIAA0715 (19) into the Gene Map as described above. Localization of Vc to chromosome 15 was from Nagase et al. (18); localization to the interval D15S122-D15S156 on the chromosome 15 map was by entering KIAA0566 (18) into the Gene Map as described above; localization to 15q11-13 was described in GenBank entry Y09954. Northern Blot Analysis A Northern blot containing poly(A)⫹-enriched RNA from murine heart, brain, spleen, lung, liver, skeletal muscle,

kidney, and testis was purchased from Clontech. Probes were a 300-bp PCR product generated from Ib (2708–3016) or a 1.5-kb Ib clone (549–2094) obtained in screening, which produced the same results; a 611-bp Stu I restriction fragment of Ih (205–816); and a 911-bp Sau 3A-Ava I restriction fragment from Va (256–1168). Random primed oligonucleotide probes were labeled with 32P by the method of Feinberg and Vogelstein (8). Each probe was hybridized (6) separately to the stripped Northern blot. In situ Hybridization The same hybridization probes used for the Northern analysis of Ib, Ih, and Va, plus a 1.4-kb EcoR V-BamH I restriction fragment of Ia (U75321, 614–2000), an 835-bp EcoR I-Pvu II restriction fragment from a Ic EST (AA242626, 1–835), an 828-bp Kpn I restriction fragment from IIa



Fig. 1. Mammalian subfamily members. Annoted diagrams represent proteins translated from mammalian cDNAs and genomic DNA (Ik). Genes whose sequences are incomplete are displayed with jagged ends; blunt ends denote known termini. Numbers (#) mark lightly shaded P-type ATPase consensus sequences, letters mark subfamily-specific and class-specific motifs within and outside these consensus sequences (10), and heavily shaded regions represent the 10 transmembrane domains characteristic of P-type ATPases. Roman numerals at left of sequences represent class designation, and nos. at right indicate human chromosomal localization. Dashed lines denote portion of encoded sequence covered by cDNA probes used for in situ hybridizations and Northern analysis. Ia, the aminophospholipid translocase (APLT) (U75321); Ib (AF156550); Ic, FIC1 (AF038007); Id, resequenced EST AA437803 (AF156546); Ik, assembled from AC004755 using Genefinder and the EST AA827939; Im, EST AA896217; In, EST AA982924; Ih (AF156551); If, resequenced EST AA639797 (AF156548); Ig, resequenced EST AA754832 (AF156547); Iq, EST AI371849; Ir, KIAA0956 (AB023173); Is, KIAA1021 (AB028944); IIa (AF152243); IIb (AF155913); Va (AF156549); Vb, KIAA0715 (AB018258); Vc, KIAA0566 (AB011138).

142


(AF152243, 378–1206), and a 620-bp restriction fragment from IIb (AF155913, 2482–3006) were subcloned into pBluescript. Linearized plasmids were extracted four times with phenol-chloroform and twice with chloroform alone. Digoxigenin-labeled transcripts of both sense and antisense cRNA were generated using either T3 or T7 RNA polymerase (Boehringer Mannheim labeling kit, catalog no. 1277073), and integrity was confirmed via denaturing formaldehydeagarose gel electrophoresis (26). Seven-day-old mouse pups and 17-day mouse embryos were embedded in Tissue-tek (O.C.T. compound from Miles), and 15- to 20-µm sections were cut onto Superfrost Plus slides. Freshly cut sections were air dried and subsequently stored at ⫺80°C until use. Before hybridization, tissue was fixed in 4% paraformaldehyde, permeabilized in 0.1% Triton X-100, and then acetylated in triethanolamine and acetic anhydride for 10 min at room temperature. After prehybridization [50% formamide, 5⫻ standard saline citrate (SSC), 5⫻ Denhardt’s solution, and 250 µg/ml yeast tRNA] for 2 h at room temperature in a humidified chamber, labeled probe was applied to sections at a concentration of 200–300 ng/ml in hybridization buffer, and hybridization was carried out at 65–75°C for 18 h. Slides were washed in 5⫻ SSC, and digoxigenin-labeled probes were detected with anti-digoxigenin Fab alkaline phosphatase (Boehringer Mannheim) as per manufacturer’s instructions. A combination of colorimetric substrates [5-bromo-4-chloro-3-indolyl-phosphate, 4-toluidine salt (BCIP) and nitroblue tetrazolium chloride (NBT)] was used at a ratio of 1:1 (Boehringer Mannheim, catalog nos. 1383221 and 1383213, respectively). After washing with 5⫻ SSC, the slides were stained with DAPI (1 µg/ml) in 0.2⫻ SSC. Slides were then mounted with aqueous mounting medium and photographed. For in situ analysis of mouse testis, animals were perfused transcardially, first with saline and then with 4% paraformaldehyde in PBS. The fixed testes were dissected out, further treated with the same fixative for 6–12 h at 4°C, rinsed three times with PBS for 5 min, and immersed in 20% sucrose in PBS for 12–24 h at 4°C (20). The testes were then embedded in O.C.T. compound, frozen at ⫺80°C, and cut into sections of 10-µm thickness. Further treatment of the sections and in situ hybridization were carried out as previously described

(31). In brief, the sections were treated successively with Triton X-100 (0.3%), proteinase K (1 µg/ml), paraformaldehyde (4%), and deionized formamide (50%); RNA probes dissolved at 1 (Ib) or 0.5 (V) µg/ml in a hybridization solution consisting of 20 mM Tris-HCl (pH 8.0), 2.5 mM EDTA, 0.3 M NaCl, 10% dextran sulfate, 1⫻ Denhardt’s solution, 1 mg/ml yeast tRNA, and 50% formamide were added; and samples were incubated at 90°C for 15 min followed by quick chilling in ice. Hybridization was performed at 60°C for 16 h, and the sections were washed successively with wash solution A (0.3 M NaCl, 30 mM sodium citrate, 50% formamide) at 45°C (for the Ib probe) or 60°C (for the V probe) for 1 h, twice with wash solution B (10 mM Tris-HCl, pH 8.5, 0.5 M NaCl) at room temperature for 5 min, with wash solution B containing RNase A (20 µg/ml) at room temperature for 30 min, with wash solution A at 45° C for 1 h, with wash solution C (0.15 M NaCl, 15 mM sodium citrate, 50% formamide) at 45° C for 1 h, and finally with wash solution C at room temperature for 30 min. The sections were then treated with an alkaline phosphatase-conjugated anti-digoxigenin antibody (Boehringer Mannheim), and hybridization signals were visualized as described above using NBT and BCIP. Testis samples were counterstained with methyl green, and the spermatogenic stage of each cross section of the seminiferous tubules was determined based on the size and morphology of spermatogenic cells and their nuclei, as described previously (20, 21, 25). RESULTS

The yeast genome contains five putative transbilayer amphipath transporters that fall into four different classes based on sequence similarity and class-specific sequence motifs (10). Although mammals harbor a much larger number of these genes, they include representatives of only yeast classes I and II, while also having genes found in both mammals and nematodes, which define a new class (V) not seen in fungi (10). Figure 1 presents all the mammalian genes identified thus far. Genes whose sequences are complete are demarcated by blunt ends; genes identified from the



Fig. 2. Sequence alignments of several subfamily members. Portions of the CLUSTALW alignment of the 11 sequences used to create the phylogenetic tree in Fig. 3, top, are shown, demonstrating subfamily-specific consensus sequences (boxed) within P-type ATPase consensus sequences (black letters on gray background) or elsewhere; classspecific consensus sequences D, E, G; and the subclass-specific tripeptide. Class I is shown as white letters on black background, class V as black letters on white background, and class II as white letters on gray background.


143

EST database are demarcated by jagged ends, with one blunt end if the terminus has been identified. Numbers mark lightly shaded P-type ATPase consensus sequences, letters mark subfamily-specific and classspecific motifs within and outside these consensus sequences (see also Fig. 2), and solid regions represent

the 10 transmembrane domains characteristic of Ptype ATPases. In mammals, class I includes the aminophospholipid translocase (APLT) gene, the FIC1 gene, and as many as 10 other genes. All of the class I genes share greater than 35% sequence identity, as well as class I-specific



Fig. 3. Phylogenetic trees of the ATPase subfamily. Top: relationship between amino acid sequences of mammalian subfamily members and plasma membrane Ca2⫹ ATPase expressed as a Neighbor-Joining tree calculated for sequences between consensus regions D and J (see Figs. 1 and 2) using P-distance, complete deletion, and Felsenstein’s bootstrap test (9) (1,000 replications, confidence levels shown as percentages at axils). Bottom: relationship between all known complete subfamily sequences from diverse eukaryotic organisms, expressed as a Neighbor-Joining tree, using P-distance, pairwise deletion, and Felsenstein’s bootstrap test (1,000 replications). Numbered sequences are as follows (GenBank accession no., see Fig 1 for mammalian): 1, plasma membrane Ca2⫹ ATPase (PMCA2; Q01814); 2, Y73C8C (AF101318) and F02C9.3 (U80025); 3, F36H2 (Z81078) and VF36H2L (AL021466); 4, IIb; 5, IIa; 6, CAA93618, SPAC6C3 (Z69731.1); 7, ATC7_YEAST YIL048W (P40527); 8, PFMAL3P6.27 (Z98551); 9, ATC8_ YEAST, YM8520.11c (Q12674); 10, PFU16955 (U16955); 11, Ih; 12, T24H7.5 (U28940); 13, W09D10.2, CAB07859.1 (Z93785); 14, Va; 15, AB005245 (complement, join 69835 – 70053..70147 – 70608..70687 – 70906..71017–71132..71229–71889..72001– 72306..72407 – 73894) assembled using BLAST comparisons; 16, SPAC4F10.16c (Z98980.1, CAB11719); 17, ATCX_SCHPO (Q09891); 18, ATC4_YEAST, YDR093W, YD8557.01 (Q12675); 19, ATC5_YEAST, YER166w (P32660, AAB64693); 20, Y17G9 (AC006719, assembled using Genefinder and ESTs); 21, Ic, FIC1; 22, F20D21.10, AAD25608.1 (AC005287); 23, SPBC887.12; 24, DRS2, YAL026C, ATC3_YEAST (AL033388.1); 25, Y49E10.12 (assembled using Genefinder and ESTs); 26, AC005986, AI517398 (assembled using Genefinder); 27, Ib; 28, Ia.

144


Fig. 4. Northern analysis of the expression of mammalian subfamily members in mouse tissues. Analyses are for Ib (A), Ih (B), and Va (C). D: summary of Northern analyses for 7 subfamily members. Ia, IIa, and IIb are from Ref. 10; Ic is from Ref. 4. H, heart; B, brain; Sp, spleen; Lg, lung; Lv, liver; M, muscle; K, kidney; Ts, testis.

other class I sequences than it is to class V sequences, class V sequences are more similar to other class I sequences than they are to the Ih sequence, suggesting that Ih could represent a separate class. However, because the bootstrap value for the placement of the 1h interior branch is low (88%), these relationships cannot be supported with confidence. To validate the relationship among the mammalian sequences depicted in Fig. 3, top, a second dendrogram was generated using all the mammalian and nonmammalian genes for which complete sequences are known (Fig. 3, bottom). This analysis reveals that the several different classes of enzymes evolved early, before the divergence of plants, fungi, protozoa, and animals from each other. The position of the Ih sequence (no. 11) in the dendrogram would suggest that the I-⍀ subclass is more closely related to classes III and V than to the rest of class I. But again the confidence probability is low (39%, not shown) for the actual position of the class I branch relative to the other classes, leaving it unclear at present whether Ih and the other members of I-⍀ form a distantly related subclass of I or actually represent another, different class; these alternatives should become distinguishable as more complete sequences become known. However, the existence of an EST clone from zebrafish (Danio rerio) (AI558738) with 68% sequence identity to Ih suggests that this group of genes is also present in other vertebrates. The human chromosomal localizations of the APLT gene (17) and the FIC1 gene (4) have recently been reported. The localization of Ib is reported here. In addition, a number of other genes in the subfamily can be mapped from the databases. Each of these localizations is shown in Fig. 1. Of the 10 genes mapped, two pairs are located on the same chromosome, the first pair (FIC1 and IIb) being localized within the same



consensus sequences in regions D and E (Fig. 2). These genes can be further subdivided into three subclasses by three criteria: 1) members within a subclass share greater than 45% sequence identity, 2) the overall length of the genes and the positioning of consensus and diagnostic sequences are similar, and 3) members of a subclass share a specific tripeptide motif at the end of the large cytosolic domain (Fig. 2). The subclass with a GAW motif, designated subclass I-⌺, contains the APLT gene and one other known gene. The subclass with a GRW motif, designated subclass I-⌽, contains the FIC1 gene and possibly four other genes. The subclass with a GHF motif, designated I-⍀, has five possible members. At present, two class II genes have been identified that, besides the previously reported class II-specific sequences (Ref. 10; D, E, and G in Fig. 2), have a GRN motif at the end of the large cytosolic domain. Currently, three class V genes have been identified, all of which share a GHW motif. The phylogenetic relationship of the genes is presented in Fig. 3. The dendrogram in Fig. 3, top, compares partial sequences from 10 mammalian members of the subfamily, as well as a partial sequence of PMCA2 (the human plasma membrane Ca2⫹ transporter) over a core region covering most of the the large cytoplasmic domain of the proteins (sequence D to sequence J, see Figs. 1 and 2). With gaps in the alignment removed, the length of the sequences compared was 779 amino acids. This comparison clearly shows that the subfamily differs distinctly from PMCA2, and thereby other members of that branch of the P-type ATPases, and confirms that members previously assigned to the same class or subclass are the most related. Calculation of the pairwise distances along the branches of the tree reveals, unexpectedly, that although the Ih sequence is more closely related to the

145


Table 1. Expression of mammalian subfamily members in mouse tissues as shown by in situ hybridization Probe

Stage

Ia

7-Day

⫹⫹⫹ olfactory epithelium ⫹ lung

E17

⫹⫹ lung epithelium,

Ib

7-Day E17

Ic

7-Day

Eye

Respiratory

⫹ olfactory epithelium

Ih

⫹⫹ bronchioles

7-Day

E17 IIa

7-Day

E17

IIb

⫹ liver, salivary gland ⫹⫹⫹ intestinal epithelium ⫹⫹⫹ colonic crypts ⫹ stomach epithelium ⫹ salivary gland, liver ⫹⫹⫹ intestinal epithelium ⫹⫹ salivary gland, liver ⫹ intestinal villi (not crypts) ⫹ salivary gland ⫹⫹ salivary gl., intestinal epith. ⫹ intestinal crypts

⫹ lungs

⫹ kidney

⫹ kidney cortex

⫹⫹ adrenal medulla

Other

⫹⫹⫹ teeth (molars) ⫹⫹ thymus ⫹ brown fat, fat cells ⫹⫹ brown fat, ⫹ thymus ⫹ thymus

⫹⫹ adrenal

⫹⫹⫹ uterine epithelium ⫹ kidney ⫹⫹⫹ kidney medulla

⫹ liver, salivary gland ⫹ intestinal epithelium ⫹ liver, salivary gland ⫹⫹ intestinal epithelium, ⫹ liver

⫹ lungs

7-Day E17

Endocrine

⫹⫹ teeth

⫹⫹⫹ teeth ⫹ adrenal

⫹ thymus medulla

⫹⫹⫹ thyroid ⫹⫹ adrenal

⫹ thymus

⫹ adrenal

⫹ thymus

⫹ salivary gland

7-Day E17

Va

⫹⫹⫹ ganglion cell layer ⫹⫹⫹ nuclear layers ⫹⫹⫹ some amicrine cells

⫹⫹ olfactory epithelium ⫹⫹⫹ airway epithelium

⫹⫹⫹ intestinal epithelium ⫹⫹⫹ colonic crypts ⫹⫹ salivary gland, ⫹ liver ⫹⫹ liver, intestinal epithelium

Urogenital

⫹ bronchioles

⫹ thymus ⫹ kidney cortex ⫹ kidney cortex

⫹ sinus epithelium

⫹⫹ entire adrenal

⫹⫹⫹ thymus

⫹⫹⫹ adrenal cortex isol. to zona glomerulosa

⫹ thymus

7-Day, 7-day-old mouse pups; E17, 17-day mouse embryos.

general region of chromosome 18 and the second pair (Ib and Is) on chromosome 13. Whereas genomic sequences in the database have identified pseudogenes from the subfamily, genes represented by EST clones are by definition expressed in the organism. EST clones representing the mammalian genes depicted in Fig. 1 are derived from tissues ranging from brain, lung, thymus, kidney, heart, mammary glands, testis, and uterus, and tumors of these organs, as well as from embryos and teratocarcinoma cell lines. Northern analyses of two class I genes [APLT (10); FICI (4)] and two class II genes (10) indicated that each of the genes is expressed as a single mRNA species of characteristic size, with the exception of Ia, which is expressed as a closely spaced doublet of RNA species (10). Analysis of two other class I genes (Ib and Ih) and one class V gene (Va) are shown in Fig. 4, A–C,

respectively. Each gene appears to be expressed as a single predominant mRNA species: Ib at 4.4 kb in heart, brain, and testis; Ih at ⬃8 kb in heart, brain, liver, and muscle; and Va at ⬃6 kb in testis. The band seen at ⬃4.4 kb in testis using the Ih probe (Fig. 4B) is most likely due to cross hybridization with mRNA from a different subfamily member. Because none of the other six probes produced a Northern profile with a band at 4.4 kb exclusively in testis, the Ih probe is likely not hybridizing to transcripts from any of the six genes. Rather, it may be hybridizing to the transcript from a closely related member of the subfamily whose expression pattern has not yet been examined. One likely candidate is the Is gene (Fig. 1), which encodes a protein 91% identical in amino acid sequence to the Ih protein. Figure 4D summarizes the expression of these seven genes in tissues of the mouse, determined by



E17

Digestive

146


Fig. 5. In situ hybridization analysis of the expression of mammalian subfamily members in intestine, lung, and the adrenal gland. Ic (FIC1) is expressed in colonic crypts (Cr; A), Ih in intestinal epithelium and villi (V) (B) and in lung airways (C), and Va in the entire adrenal gland (Ad) of 7-day-old pups (D) but exclusively in the zona glomerulosa (arrow) of the adrenal cortex in 17-day embryos (inset). Br, bronchioles; K, kidney. Bar, 50 µm (in A and B) or 300 µm (in C and D).

1. Although expressed in very few tissues, Ic (FIC1) is highly expressed in the epithelium of the stomach, intestines (most impressively in colonic crypts, see Fig. 5A), and liver of both 7-day pups and embryos. Ia expression has a similar distribution within the intestinal epithelium and colonic crypts. To a lesser extent, IIa is expressed in the intestinal crypts of 7-day pups and IIb is expressed in the intestinal epithelium of embryos. In contrast, Ih appears to be expressed in intestinal villi, rather than the crypts (Fig. 5B). Within the respiratory system, Ia and Ih are expressed in the olfactory and lung epithelia; Ih particularly in the bronchioles (Fig. 5C). All of the probes except Ib hybridize with adrenal glands. Va is expressed in the entire

Fig. 6. In situ hybridization analysis of the expression of mammlian subfamily members in retina, teeth, thymus, and uterine epithelium of 7-day-old pups. A: IIa expression in outer nuclear layer (ONL), inner nuclear layer (INL), some amacrine cells (Am), and the ganglion cell layer (GCL) of the retina. PE, pigmented epithelium; Ph, photoreceptors; OPL, outer plexiform layer; IPL, inner plexiform layer. B: Ia expression in teeth (T). Tg, tongue. C: Va expression in the thymus (Thy). Aur, auricle of heart. D: Ih expression in the uterine epithelium (UE). LI, large intestine. Bar, 50 µm (in A), 100 µm (in B), 200 µm (in C ), and 30 µm (in D).



Northern analysis using the probes indicated by the dashed lines in Fig. 1. The differences in transcript sizes combined with the distinctive patterns of tissue expression of each of the genes indicates that the probes used in these analyses were specific for the genes they represented and that cross-hybridization with transcripts from other genes in the subfamily was negligible. To define more precisely the regions and cell types within tissues expressing the genes, in situ hybridization was performed on sections of mouse embryos and pups using the same probes used for the Northern analyses. The expression of genes in tissues other than the nervous system and testis is summarized in Table

147


Table 2. Expression of mammalian subfamily members in mouse nervous system as shown by in situ hybridization Stage

CNS

Ia

7-Day

Ib

E17 7-Day

⫹⫹⫹ basal forebrain, amygdala, subthalmic nucleus, substantia nigra pars compacta ⫹⫹ hippocampus (all areas), layers 2, 4, and 6 of cerebral cortex ⫹⫹ lateral geniculate nuclei, cerebellar granule cells, superior colliculus ⫹⫹ other brain stem nuclei, subventricular zone, peduncular pontine nucleus ⫹ striatum, thalmus, olfactory bulb (mitral cells), mammilary bodies, glia ⫹ hypothalmus, nucleus of CN5 ⫹⫹ post mitotic neurons, dev. motor neurons ⫹⫹⫹⫹ mesencephalic nucleus of CN 5, ⫹⫹⫹ mitral cells of olfactory bulb ⫹⫹ deep layers of cerebral cortex, ⫹ hippocampus, multiple brain stem nuclei ⫹⫹⫹ mature neurons, ⫹ basal forebrain, spinal cord none none ⫹⫹⫹ choroid plexus, mesencephalic nucleus of CN 5, ⫹⫹ hippocampus ⫹ cerebellar granule and purkinjie cells, pons, medulla ⫹⫹ choroid plexus ⫹⫹ all neurons (except olfactory granule cells and most of thalamus) ⫹⫹ lateral dorsal nucleus, scattered glia, ⫹ cerebellar granule cells ⫹⫹⫹⫹ all spinal cord neurons ⫹⫹⫹ all postmitotic neurons but not in proliferating neurons of ventricular zone ⫹⫹⫹ all areas of hippocampus and subiculum, ventromedial nucleus of hypothalamus ⫹⫹⫹ layers 2, 4, 6 of cerebral cortex (strongest in 4), superior olive of the brainstem ⫹⫹ olfactory bulb, olfactory granule cells, basal forebrain ⫹⫹ anteriodorsal thalamic nuclei, ⫹ dorsolateral thalamic nuclei ⫹ scattered cells in sup and inf colliculi, nuclei of CN 5 and 7 ⫹⫹ all spinal cord neurons (motor ⬎ other), ⫹ pontine nuclei ⫹⫹ subiculum ⫹ developing cerebellar granule cells, hippocampus, ⫹ cells adjacent to corpus collosum, olfactory bulb mitral cells ⫹⫹ dorsal horns of the developing spinal cord, developing hypothalamus ⫹⫹ clumped distribution in ventricular zone surrounding lateral ventricles

Ih

E17 7-Day E17 7-Day

IIa

E17 7-Day

Ic

E17 IIb

Va

7-Day

E17 7-Day E17

PNS

⫹⫹⫹ peripheral ganglia, enteric neurons

⫹⫹ dorsal root ganglia, enteric neurons ⫹⫹⫹ dorsal root and cranial ganglia ⫹⫹ semilunar ganglion ⫹⫹ dorsal root ganglia, ⫹⫹⫹ trigeminal ganglion none none ⫹⫹⫹ semilunar ganglion ⫹⫹⫹ trigeminal ganglion ⫹⫹⫹ myenteric plexus

⫹⫹⫹⫹ dorsal root ganglia, mesenteric plexus ⫹⫹⫹⫹ myenteric ganglia, trigeminal ganglion ⫹⫹ spinal trigeminal nucleus ⫹ dorsal root ganglia ⫹ spinal vestibular nucleus ⫹ dorsal root ganglia, enteric neurons

CNS, central nervous system; PNS, peripheral nervous system; 7-Day, 7-day-old mouse pups; E17, 17-day mouse embryos.

adrenal in the 7-day pups but only in a very specific layer of the cortex (zona glomerulosa) in embryos (Fig. 5D), whereas Ia is expressed mostly in the adrenal medulla in embryos. The only significant expression in the eye is seen with IIa, which is found in the ganglion cell layer, inner and outer nuclear layers, and some amacrine cells in the retina (Fig. 6A). Ia, Ic, and Ih are expressed in the enamel layer of teeth (Fig. 6B). Ia, Ib, Ih, IIa, IIb, and Va are all expressed in the thymus (Fig. 6C). Ih is highly expressed in uterine epithelium (Fig. 6D) and the medulla of the kidney. In addition to these evidences for regionalization of expression at the organ level, the genes are expressed in a particularly complex pattern in brain and testis, organs in which many of the genes in the subfamily are expressed. Most of the genes were expressed in the nervous system of both 7-day-old pups and 17-day

embryos, particularly in the brain (Table 2), with the exception of Ic (FIC1), which Northern analysis had already indicated was not expressed in the brain. Ia, IIa, and IIb demonstrate a mostly neuronal expression both in pups and in embryos (Table 2) and intense staining of the hippocampus (Fig. 7, A and B). Other genes, Ib and Ih, demonstrate areas of exquisitely specific expression, such as in the mesencephalic nucleus of cranial nerve V (Fig. 7C). Ih is also highly expressed in the choroid plexus, where no other subfamily gene is expressed (see Fig. 7C). A more precise description of the expression of the genes in the testis is provided by in situ hybridizations carried out on mouse testis sections. As shown in Fig. 8 and summarized in Table 3, Ib and Va each hybridized exclusively to cells in early stages of spermatid development (round spermatids, steps 1–8) as opposed to elongated spermatids, spermatogonia, spermatocytes,



Probe

148


Leydig cells, or Sertoli cells, whereas Va is expressed slightly earlier in spermiogenesis than is Ib. DISCUSSION

PS is restricted to the inner leaflet of the plasma membrane bilayer in nearly all cells of the mouse embryo (28). However, during embryogenesis cells of various lineages undergoing apoptosis in a wide variety of tissues expose PS on their surface (28). In the specific cases of lymphocytes, neutrophils, and vascular smooth



Fig. 7. In situ hybridization analysis of the expression of mammalian subfamily members in mouse brain. A: Ia expression in the hippocampus, dentate gyrus (DG) and the CA1/CA3 subdivisions of Ammon’s Horn. B: Ia expression (at a higher magnification of CA1, indicated by box in A) in the cytoplasm (arrow) of hippocampal neurons. C: Ih expression in the choroid plexus (CP), the cerebellar granule cell layer (GCL), the Purkinje cell layer (P), and the mesencephalic nucleus of cranial nerve 5 (MC5). Bar, 250 µm (in A and C) and 25 µm (in B).

muscle cells, this exposed PS marks apoptotic cells for phagocytosis (7, 3, 23), thus preventing the inflammation associated with cell lysis. Because the APLT is responsible for transporting and restricting PS to the inner leaflet of the bilayer, it would be expected that all cells should express the enzyme. However, this simple picture has become complicated by the discovery of as many as 17 different genes within the subfamily of genes which the APLT first identified. With the discovery of several classes of genes within the APLT subfamily, it was suggested, by analogy with the other subfamilies of P-type ATPases, that each class of enzymes might transport a different substrate (10). In this view, the various genes within class I might all transport aminophospholipids, but at different times or in different tissues. The present studies were designed partly to test that possibility. As is evident, genes a, b, c, and h in class I all display distinct patterns of tissue expression, although expression of some genes overlaps (discussed below), consistent with each gene encoding an APLT. It is important to note that the pattern of expression of those genes that are most similar to each other (Ia and Ib, 66% identical; or IIa and IIb, 76% identical) were not the same, strongly discounting the possibility of cross hybridization. On the other hand, probes for two genes with little similarity to one another (Ib and Ih, 38% identical) and unlikely to cross hybridize both strongly hybridized to cranial nerve V ganglia and nuclei, indicating specific coexpression of at least two subfamily genes. Cloning of the Ia APLT gene began with purification of the protein from secretory vesicles, namely the adrenal medulla chromaffin granules. It is therefore not surprising that Ia is expressed in the adrenal medulla of embryos (the adrenal gland was not present in the sagittal sections of pups probed with Ia). The presence of the enzyme in these membranes suggests an association with the secretory process. For example, concentration of PS in the cytosolic leaflet of the granule membrane, in apposition to the PS on the cytosolic leaflet of the plasma membrane, may play a role in the vesicle-plasma membrane fusion event of secretion; speculative mechanisms have been proposed (30). If high levels of the APLT are required in cells specialized for secretion, such a pattern of expression should be seen here. In fact, expression is seen in tissues such as the enamel layer of teeth, intestinal crypt cells, and liver, all of which are sites of high levels of secretion. However, many endocrine organs, such as the thyroid, are not particularly strong expressors of the Ia gene, and tissues such as the thymus or brown fat, which do express the gene, are not actively secretory. These results argue that elevated expression of the Ia gene may serve some more subtle function than secretion in the cells where it is expressed. This view is supported by the highly idiosyncratic distribution of Ia expression (as well as the other class I genes, with the exception of Ic) in the brain.


149

The FIC1 (Ic) gene exhibits a particularly restricted pattern of tissue-specific expression. Because the gene was identified by its role in bile acid secretion, it might be expected that it would be heavily expressed in liver. Although expression in the liver is observed, the level of expression is greatest in the intestine and particularly in the colonic crypts, suggesting that expression of the FIC1 gene product in crypt cells may play a role in recycling of bile acids from the intestine to the liver. Further resolution of the role played by FIC1 in bile acid secretion will await determination of the substrate transported by the FIC1 enzyme, as well as of the subcellular location of the enzyme. For instance, should the enzyme transport aminophospholipids and be located in internal vesicles, a general, but cell typespecific, role in vesicular transport that supports bile acid transport would be implied. However, should the Table 3. Expression of mammalian subfamily members in mouse testis as shown by in situ hybridization Cell Type

Ib

Va

Spermatogonia Spermatocytes Spermatids Steps 1–3 Steps 4–5 Steps 6–8 Steps 9–12 Steps 13–16 Leydig cells Sertoli cells

⫺ ⫺

⫺ ⫺

⫹ ⫹⫹ ⫹⫹ ⫹ ⫺ ⫺ ⫺

⫹⫹ ⫹⫹ ⫹ ⫺ ⫺ ⫺ ⫺

enzyme be located in the plasma membrane and transport bile acids, a very direct role in bile acid transport would be evident. Similarly, the substrate transported by the class II and V genes remains unknown. In the absence of other evidence, the analogy with similar groupings of genes in the larger subfamily of P-type ATPases suggests that the products of class II and V genes might transport molecules distinct from those transported by class I proteins. In the case of class II, the two genes currently known have a highly overlapping distribution of expression in somatic tissues, which might suggest that they are a redundant pair. Their distribution in the brain, on the other hand, is overlapping but quite distinct, arguing that the genes support different functions. However, in the case of class V, temporal expression of Va overlaps with, but is distinct from, temporal expression of Ib in the early stages of spermatid development, suggesting the two genes may in fact have similar functions but be expressed at slightly different times during maturation. The fact that disabling the FIC1 gene produces a readily discernable pathological phenotype forcefully demonstrates that at least in this instance the multiplicity of genes in the subfamily does not simply reflect redundancy of function, because other genes in the subfamily are apparently unable to compensate for the function provided by FIC1. This result calls attention to the importance that knockouts of other members of the subfamily in mice will play in pointing the way to the roles each of the different enzymes in the subfamily play in normal physiological processes.



Fig. 8. In situ hybridization analysis of the expression of mammalian subfamily members in mouse testis. Ib (A and B) and Va (C and D) expression is found in the cytoplasm (arrows) of round spermatids (r) but not elongated spermatids (e) within tubules of seminiferous epithelium. Stages of maturation are designated by Roman numerals (Ref. 21). Bar, 50 µm.

150


The authors thank Andrew Clark for advice concerning the phylogenetic analysis. This work was supported by a Medical Scientist Training Grant (to J. F. Lawler) and by National Institutes of Health Grant GM-55862. Address for reprint requests and other correspondence: R. A. Schlegel, Dept. of Biochemistry and Molecular Biology, 428 South Frear Laboratory, Penn State University, University Park, PA 16802.

17.

REFERENCES 18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.



1. Andre, B. An overview of membrane transport proteins in Saccharomyces cerevisiae. Yeast 11: 1575–1611, 1995. 2. Altschul, S. F., W. Gish, W. Miller, E. W. Myers, and D. J Lipman. Basic local alignment search tool. J. Mol. Biol. 215: 403–410, 1990. 3. Bennett, M. R, D. F. Gibson, S. M. Schwartz, and J. F. Tait. Binding and phagocytosis of apoptotic vascular smooth muscle cells is mediated in part by exposure of phosphatidylserine. Circ. Res. 77: 1136–1142, 1995. 4. Bull, L. N., M. J. van Eijk, L. Pawlikowska, J. A. DeYoung, J. A. Juijn, M. Liao, L. W. Klomp, N. Lomri, R. Berger, B. F. Scharschmidt, A. S. Knisely, R. H. Houwen, and N. B. Freimer. A gene encoding a P-type ATPase mutated in two forms of hereditary cholestasis. Nature Genet. 18: 219–224, 1998. 5. Catty, P., A. de Kerchove d’Exaerde, and A. Goffeau. The complete inventory of the yeast Saccharomyces cerevisiae P-type transport ATPases. FEBS Lett. 409: 325–332, 1997. 6. Church, G. M., and W. Gilbert. Genomic sequencing. Proc. Natl. Acad. Sci. USA 81: 1991–1995, 1984. 7. Fadok, V. A., D. R. Voelker, P. A. Campbell, J. J. Cohen, D. L. Bratton, and P. M. Henson. Exposure of phosphatidylserine on the surface of apoptotic lymphocytes triggers specific recognition and removal by macrophages. J. Immunol. 148: 2207–2216, 1992. 8. Feinberg, A. P., and B. Vogelstein. A technique for radiolabeling DNA restriction endonuclease fragments to high specificity. Anal. Biochem. 132: 6–13, 1983. 9. Felsenstein, J. Confidence limits on phylogenies: an approach using the bootstrap. Evolution 39: 783–791, 1985. 10. Halleck, M. S., D. Pradhan, C. Blackman, C. Berkes, P. Williamson, and R. A. Schlegel. Multiple members of a third subfamily of P-type ATPases identified by genomic sequences and ESTs. Genome Res. 8: 354–361, 1998. 11. Heng, H., and L.-C. Tsui. Modes of DAPI banding and simultaneous in situ hybridization. Chromosoma 102: 325–332, 1993. 12. Houwen R. H., S. Baharloo, K. Blankenship, P. Raeymaekers, J. Juyn, L. A. Sandkuijl, and N. B. Freimer. Genome screening by searching for shared segments: mapping a gene for benign recurrent intrahepatic cholestasis. Nature Genet. 8: 380– 386, 1994. 13. Ishikawa K, T. Nagase, M. Suyama, N. Miyajima, A. Tanaka, H. Kotani, N. Nomura, and O. Ohara. Prediction of the coding sequences of unidentified human genes. X. The complete sequences of 100 new cDNA clones from brain which can code for large proteins in vitro. DNA Res. 5: 169–176, 1998. 14. Krahling, S., M. K. Callahan, P. Williamson, and R. A. Schlegel. Exposure of phosphatidyl-serine is a general feature in the phagocytosis of apoptotic lymphocytes by macrophages. Cell Death Differ. 6: 183–189, 1999. 15. Kumar, S., K. Tamura, and M. Nei. MEGA: Molecular Evolutionary Genetics Analysis (version 1.01). University Park, PA: University of Pennsylvania, 1993. 16. Lichter, P., C. J. Tang, K. Call, G. Hermanson, G. A. Evans, D. Housman, and D. C. Ward. High-resolution mapping of

human chromosome 11 by in situ hybridization with cosmid clones. Science 247: 64–69, 1990. Mouro, I., M. S. Halleck, R. A. Schlegel, M. G. Mattei, P. Williamson, A. Zachowski, P. Devaux, J.-P. Cartron, and Y. Colin. Cloning, expression and chromosomal mapping of a human ATPase II gene, member of the third subfamily of P-type ATPases and orthologous to the presumed bovine and murine aminophospholipid translocase. Biochem. Biophys. Res. Comm. 257: 333–339, 1999. Nagase T, K. Ishikawa, N. Miyajima, A. Tanaka, H. Kotani, N. Nomura, and O. Ohara. Prediction of the coding sequences of unidentified human genes. IX. The complete sequences of 100 new cDNA clones from brain which can code for large proteins in vitro. DNA Res. 5: 31–39, 1998. Nagase, T., K. Ishikawa, M. Suyama, R. Kikuno, N. Miyajima, A. Tanaka, H. Kotani, N. Nomura, and O. Ohara. Prediction of the coding sequences of unidentified human genes. XI. The complete sequences of 100 new cDNA clones from brain which code for large proteins in vitro. DNA Res. 5: 277–286, 1998. Numata M., T. Ono, and S. Iseki. Expression and localization of the mRNA for DNA (cytosine-5)-methyltransferase in mouse seminiferous tubules. J. Histochem. Cytochem. 42: 1271–1276, 1994. Oakberg, E. F. A description of spermiogenesis in the mouse and its use in analysis of the cycle of the seminiferous epithelium and germ cell renewal. Am. J. Anat. 99: 391–409, 1956. Pedersen, P., and E. Carafoli. Ion motive ATPases. I. Ubiquity, properties, and significance to cell function. Trends Biochem. Sci. 12: 146–150, 1987. Pradhan, D., S. Krahling, P. Williamson, and R. A. Schlegel. Multiple systems for recognition of apoptotic lymphocytes by macrophages. Mol. Biol. Cell 8: 767–778, 1997. Pradhan, D., P. Williamson, and R. A. Schlegel. Phosphatidylserine vesicles inhibit phagocytosis of erythrocytes with a symmetric transbilayer distribution of phospholipids. Mol. Membr. Biol. 11: 181–188, 1994. Russell L. D., R. A. Ettlin, A. P. Sinha Hikim, and E. D. Clegg. Histological and Histopathological Evaluation of the Testis. Clearwater, FL: Cache River Press, 1990, p. 119–161. Sambrook, J., E. F. Fritsch, and T. Maniatis. Molecular Cloning: A Laboratory Manual. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press, 1989, p. 7.43–7.45. Tang, X., M. S. Halleck, R. A. Schlegel, and P. Williamson. A subfamily of P-type ATPases with aminophospholipid transporting activity. Science 272: 1495–1497, 1996. Van den Eijnde, S. M., L. Boshart, C. P. M. Reutelingsperger, C. I. De Zeeuw, and C. Vermeij-Keers. Phosphatidylserine plasma membrane asymmetry in vivo: a pancellular phenomenon which alters during apoptosis. Cell Death Differ. 4: 311–316, 1997. Verhoven, B, S. Krahling, R. A. Schlegel, and P. Williamson. Regulation of phosphatidylserine exposure and phagocytosis of apoptotic T lymphocytes. Cell Death Differ. 6: 262–270, 1999. Williamson, P., and R. A. Schlegel. Back and forth: the regulation and function of transbilayer phospholipid movement in eukaryotic cells. Mol. Memb. Biol. 11: 199–216, 1994. Yokouchi Y., K. Ohsugi, H. Sasaki, and A. Kuroiwa. Chicken homeobox gene Msx-1: structure, expression in limb buds and effect of retinoic acid. Development 113: 431–444, 1991.