The mouse APLP2 gene. Chromosomal localization and promoter characterization.

Senile plaques are primarily comprised of deposits of the β-amyloid protein derived from larger amyloid precursor proteins (APPs). APP is a member of a gene family, including amyloid precursor-like proteins APLP1 and APLP2. Using interspecific mouse backcross mapping, we localized the mouse APLP2 gene to the proximal region of mouse chromosome 9, syntenic with a region of human 11q. We cloned an 1.2-kilobase mouse genomic fragment containing the APLP2 gene promoter. The APLP2 promoter lacks a typical TATA box, is GC-rich, and contains several sequences for transcription factor binding. S1 nuclease protection analysis revealed the presence of multiple transcription start sites. The lack of a TATA box, the presence of a high GC content, and multiple transcription start sites place the APLP2 promoter in the class of promoters of “housekeeping genes.” Regulatory regions within the promoter were assayed by transfection of mouse N2a and Ltk cells with constructs containing progressive 5′-deletions of the APLP2 promoter fused to the bacterial chloramphenicol acetyl transferase (CAT) reporter gene. A minimal region that includes sequences 99 bp upstream of the predominant transcription start site of the APLP2 promoter was sufficient to direct high levels of CAT expression.

Notably, APLP2 shares considerable sequence homology with APP with the exception of the ␤-amyloid domain (5,7,8). In earlier studies, we demonstrated that APLP2 matures through the same unusual secretory/cleavage pathway as APP. Furthermore, APLP2 pre-mRNAs are alternatively spliced to generate at least four alternatively spliced transcripts (9,10). Using in situ hybridization and reverse transcriptase-polymerase chain reaction (RT-PCR) approaches, we and others have demonstrated that in most adult tissues, APLP2 and APP mRNAs were expressed at similar, if not identical, levels. There are several exceptions; notably, in liver APP mRNA is essentially undetectable, but APLP2 mRNA is fairly abundant (5,7,9,11). In recent studies, we have also demonstrated that specific alternatively spliced APLP2 mRNAs are differentially expressed in the olfactory epithelium (12). Moreover, APLP2 is highly enriched in olfactory sensory axons and axon terminals in glomeruli. On the other hand, APP is expressed, albeit at lower levels, in olfactory sensory neurons and to a lesser extent in sensory axons. This suggests that APLP2 and APP are regulated differentially in selected neuronal populations.
In order to assess whether the differential levels of APLP2 and APP expression may be a reflection of differences in sequence elements contained within respective promoters, we cloned and characterized an ϳ1.2-kb fragment of the mouse APLP2 gene promoter. The mouse APP promoter has been characterized previously (13). We show that the mouse APLP2 gene promoter contains several features characteristic of promoters of "housekeeping genes"; these include the lack of a typical TATA box, the presence of a high GC content, and multiple transcription start sites. These latter features of the APLP2 promoter are similar to features described for mouse, rat, and human APP promoter regions (13)(14)(15)(16)(17). We assessed whether the APLP2 promoter contained positive or negative regulatory elements by transfecting mouse neuroblastoma (N2a) cells and mouse fibroblast (Ltk Ϫ ) cells with constructs containing progressive 5Ј-truncated promoter fragments of the APLP2 gene fused with the reporter gene chloramphenicol acetyl transferase (CAT). We demonstrate that CAT expression remains fairly constant across different deletion constructs in both N2a and Ltk Ϫ cells and that a fragment representing just 99 bp upstream of the predominant transcription start site is sufficient to direct high levels of transgene expression in both cell lines. Interestingly, 5Ј-deletion studies of the human, mouse, and rat promoters also revealed that ϳ100 bp of the respective promoters can drive high levels of expression of reporter genes (13,15,18).
A description of the probes and restriction fragment length polymorphisms for the loci linked to APLP2 including low density lipoprotein receptor (Ldlr), preproenkephalin (Penk), and E26 avian leukemia oncogene (Ets1) has been reported previously (21). Recombination distances were calculated as described (22) using the computer program SPRETUS MADNESS. Gene order was determined by minimizing the number of recombination events required to explain the allele distribution patterns.
Library Screening-To isolate the promoter of mouse APLP2, a 67-bp fragment of the 5Ј-untranslated region of mouse APLP2, position Ϫ64 to ϩ3 with respect to the translation start codon, was generated by PCR and labeled with [␣-32 P]dATP by random primer-initiated synthesis. This probe was used to screen a genomic DNA library, prepared from 129 SV mouse embryonic stem cells cloned into EMBL3. Hybridization and wash conditions were 50% formamide, 6 ϫ SSC at 42°C for 16 h, and 2 ϫ SSC at 50°C for 15 min, followed by 0.2 ϫ SSC at 50°C for 15 min, respectively.
One positive phage contained a 14-kb SalI-SalI fragment, which included 2.8 kb of sequence upstream of the translation start codon. A 2.84-kb SalI-HindIII restriction fragment from this phage was subsequently subcloned into Bluescript KS ϩ (Stratagene) to generate plasmid pAPLP2P and partially sequenced with Sequenase (U.S. Biochemical Corp.). Sequences were analyzed for putative transcription factor binding sites using a MacVector version 4.1 software package.
RNA Isolation-Total RNA was isolated by homogenization of mouse thymus, heart, brain, liver, kidney, lung, testes, and spleen in 4 M guanidine thiocyanate and centrifugation of the lysate over a 5.7 M cesium chloride cushion (23). Poly(A) ϩ RNA from Chinese hamster ovary (CHO) cells was prepared similarly with the addition of fractionation on an oligo(dT)-Sepharose column (24). Total cytoplasmic RNA, from confluent dishes of mouse neuroblastoma (N2a) cells and mouse fibroblast (Ltk Ϫ ) cells, was isolated as described (25). S1 Nuclease Analysis-A 534-bp KpnI-HindIII fragment, extending from 494 bp upstream of the translation start site to 40 bp into exon 1, was liberated from pAPLP2P and subcloned into KpnI-HindIII-digested Bluescript KS ϩ (Stratagene), to generate plasmid pAPLP2S1. This plasmid was linearized with HindIII, which lies 40 bp 3Ј to the translation start codon, and the 5Ј ends were dephosphorylated with calf alkaline phosphatase. S1 nuclease probe was prepared by 5Ј end-labeling with [␥-32 P]ATP. For S1 nuclease analysis (25) 0.02 pmol of 32 Pend-labeled double-stranded DNA probe was mixed with either 20 g of total RNA or with 1 g of poly(A) ϩ RNA and hybridized in a solution containing 80% formamide, 0.4 M NaCl, 40 mM PIPES, pH 6.4, and 1 mM EDTA for 12-16 h at 57°C. Samples were then diluted 15-fold with ice-cold S1 nuclease buffer to yield a final concentration of 1 ϫ S1 buffer (0.2 M NaCl, 30 mM NaOAc, pH 4.5, 5 mM ZnCl 2 , and 0.05 g/l salmon sperm DNA) and treated with 100 units of S1 nuclease at 25°C for 1 h. S1-resistant hybrids were fractionated by electrophoresis on 4% acrylamide, 9 M urea-containing gels, and the protected probe was visualized by autoradiography.
RT-PCR-To determine the endogenous levels of APLP2 and APP mRNA in mouse N2a and mouse Ltk Ϫ cells, 1 g of total cytoplasmic RNA was reverse-transcribed in the presence of reverse transcriptase and random hexamer primers (Pharmacia Biotech Inc.). The first strand cDNA obtained from reverse-transcribed RNA was then subjected to PCR with degenerate primers, APP/APLP2S and APP/ APLP2AS (5). Primer APP/APLP2S is GAGCAYGCCCRYTTCCA-GAARGC, where Y ϭ C ϩ T and R ϭ A ϩ G, and encodes APLP2-751 residues 386 -392 or APP-751 residues 368 -374. Primer APP/APLP2AS is GGAGGTGTGTCATMACCTGGGA, where M ϭ A ϩ C, and is com-plementary to sequences that encode APLP2-751 residues 527-532 or APP-751 residues 509 -514 (5). PCR was performed at an annealing temperature of 58°C for 20 cycles. PCR generated 444-bp products consisting of a mixture of APP and APLP2 cDNAs, which were subsequently digested with XhoI to specifically cleave the APP-related species. Digested PCR products were fractionated on 2% agarose gels and stained with ethidium bromide. PCR products generated from plasmids encoding mouse APLP2 and mouse APP templates were used as controls.
Construction of Deletion Plasmids-A ϳ2.8-kb SalI-BamHI fragment, extending from ϳ2.7 kb upstream of the transcription start codon to 62 bp of exon 1, was isolated by PCR using a sense primer EMBL (GCTTCTCATAGAGTCTTGCAGACAAACTGCGCAAC, located in the left arm of EMBL3 polylinker; Ref. 26) and an antisense primer BamHIϩ62 (CCGGGATCCCTCTCCCCGTCTCTCGCACAGCCAGGC-TACAG, located from ϩ62 to ϩ31 with respect to the transcription start codon), in the presence of -APLP2 DNA and subcloned into SalI-BamHI-digested pBLCAT3 (27), to generate plasmid pAPLP2PCAT. pAPLP2PCAT was digested with PstI and religated to generate plasmid pAPLP2PCAT-380. During this digestion, a 590-bp PstI-PstI fragment was isolated and cloned in the sense orientation into PstI-digested pAPLP2PCAT-380 to generate plasmid pAPLP2PCAT-971. APLP2 promoter fragments range from Ϫ380 to ϩ62 in pAPLP2PCAT-380 and from Ϫ971 to ϩ62 in pAPLP2PCAT-971 (with respect to the transcription start site).
Cell Culture, Transfection, and CAT Assay-Mouse N2a cells were grown in Dulbecco's modified Eagle's medium and reduced serummodified Eagle's medium with 10% fetal bovine serum. Cells were plated 22-26 h prior to transfection at a density of 0.25 ϫ 10 6 cells/well in a 6-well dish. N2a cells were transiently transfected with 2 g of double CsCl-purified DNA using a calcium phosphate co-precipitation procedure (30). 0.12 g of pBLCAT3, or equivalent molar amounts of APLP2-CAT constructs containing various 5Ј-deletions of APLP2 promoter were adjusted to 2 g with empty vector DNA. DNA was incubated with 62.5 mol of CaCl 2 and 1 ϫ BES-buffered saline (pH 6.97) at 25°C for 20 min, and the mixture was added dropwise to each well. Cells were incubated at 3% CO 2 for 16 -18 h, after which time the precipitate was removed by washing cells two times with culture medium. The cells were subsequently returned to 5% CO 2 for 12-14 h, washed once with 1 ϫ phosphate-buffered saline and scraped in 200 l of 0.25 M Tris/HCl, pH 7.9.
To assay for CAT activity, 20 g of cell lysate was incubated in the presence of 1.1 mM acetyl CoA, 100 nCi of [ 14 C]chloramphenicol (60 mCi/mmol) in 0.22 M Tris/HCl, pH 7.7, at 37°C for 45 min. Acetylated and nonacetylated forms of chloramphenicol were extracted with 0.5 ml of ethyl acetate and separated by ascending silica gel thin-layer chromatography in chloroform:methanol (95:5) at room temperature. Thinlayer chromatography sheets were then air-dried, and acetylated and nonacetylated forms of chloramphenicol were quantified using a Phos-phorImager. The percentages of monoacetylated forms of chloramphenicol were plotted for each construct and normalized to the CAT activity of pRSVCAT. Each construct was tested in three separate transfections, and standard error of the mean was determined.
For transfections of mouse fibroblast Ltk Ϫ cells, cells were plated at a density of 0.2 ϫ 10 6 cells/well in a 6-well dish. Cells were transiently transfected with 4.26 g of pBLCAT3 or equivalent molar amounts of CAT plasmids containing various 5Ј-deletions of the APLP2 promoter adjusted to 7 g with empty vector DNA. 20 g of cell lysate was used for CAT assays.

RESULTS AND DISCUSSION
Recent studies have indicated that APP is a member of a larger gene family that includes APLP1 and APLP2. The phys-iological function(s) and regulation of the APP-related proteins is not well understood. In this study, we mapped the genomic location of APLP2 and have analyzed the APLP2 promoter for the presence of potential regulatory sequences that may be involved in transcriptional activity of the APLP2 gene.
Chromosomal Localization of APLP2-The chromosomal location of the mouse APLP2 gene was determined by interspecific backcross analysis using progeny derived from matings of ((C57BL/6J ϫ M. spretus)F 1 ϫ C57BL/6J) mice. This interspecific backcross mapping panel has been typed for over 1800 loci that are well distributed among all of the autosomes as well as the X chromosome (19). C57BL/6J and M. spretus DNAs were digested with several enzymes and analyzed by Southern blot hybridization for informative restriction fragment length polymorphisms using a mouse cDNA APLP2 probe. The 6.6-, 4.2-, and 2.7-kb M. spretus restriction fragment length polymorphisms (see "Materials and Methods") were used to follow the segregation of the APLP2 locus in backcross mice. The mapping results indicated that APLP2 is located in the proximal region of mouse chromosome 9 linked to Ldlr, Penk, and Ets1. Although 152 mice were analyzed for every marker and are shown in the segregation analysis (Fig. 1), up to 185 mice were typed for some pairs of markers. Each locus was analyzed in pairwise combinations for recombination frequencies using the additional data. The ratios of the total number of mice exhibiting recombinant chromosomes to the total number of mice analyzed for each pair of loci and the most likely gene order are: centromere, Ldlr (3/162) Penk (13/185) APLP2 (5/158) Ets1. The recombination frequencies (expressed as genetic distances in centimorgans Ϯ the standard error) are as follows: Ldlr (1.9 Ϯ 1.1) Penk (7.0 Ϯ 1.9) APLP2 (3.2 Ϯ 1.4) Ets1.
We have compared our interspecific map of chromosome 9 with a composite mouse linkage map that reports the map location of many uncloned mouse mutations (provided from the Mouse Genome data base, a computerized data base maintained at The Jackson Laboratory, Bar Harbor, ME). APLP2 mapped in a region of the composite map that lacks mouse mutations with a phenotype that might be expected for an alteration in this locus (data not shown).
The proximal region of mouse chromosome 9 shares regions of homology with human chromosomes 19p, 8q, and 11q (summarized in Fig. 1). The recent assignment of APLP2 to 11q23-q25 (31) confirms and extends the synteny between mouse chromosome 9 and human 11q.
Transcription Initiation Site of Mouse APLP2 mRNA-RNA prepared from CHO cells and several mouse tissues was subjected to S1 nuclease protection analysis using a doublestranded DNA probe 5Ј end-labeled 40 bp downstream of the translation start codon ( Fig. 2A). Although the predominant start site is located ϳ89 bp upstream of the translation start codon, there appeared to be variable levels of alternatively initiated APLP2 transcripts in mRNA isolated from CHO cells or from mouse thymus, heart, brain, liver, kidney, lung, testes, and spleen (Fig. 2B, lanes 1 and 2-9, respectively). Primer extension analysis also revealed the presence of multiple transcription start sites in mouse tissues (data not shown). Multiple transcription start sites have been identified for human (14) and rat (15) APP mRNA, with the predominant start sites located 146 and 156 bp upstream of the translation start codons, respectively. However, the transcription start site of mouse APP has not yet been reported.
Isolation of Genomic Sequences Containing the Mouse APLP2 Promoter-We screened ϳ800,000 independent phage-containing genomic DNA from a 129 SV embryonic stem cell library with a 67-bp fragment of the 5Ј-untranslated region of APLP2 (position Ϫ64 to ϩ3 with respect to the translation start codon). We obtained two overlapping phage with the longest insert containing 2.8 kb of sequence upstream of the translation start codon. ϳ1.2 kb of this promoter region was sequenced (Fig. 3).
The DNA sequence upstream of the predominant transcription start site contains a CAAT box (Ϫ135 in antisense orientation) but lacks a typical TATA box (Fig. 3). The promoter has a high GC content, specifically between positions Ϫ1 and Ϫ300 (68%) and Ϫ500 and Ϫ700 (69%). Multiple consensus sequences for transcription factor binding sites are present in the entire region, including one AP-1, two AP-2s, five GC boxes, one GC element, two GC factors, and seven SP-1 sites. Similar putative transcription factor binding sites are found in the APP promoter, however, at different locations with respect to the transcription start site (13,14,16,17). Furthermore, the APP promoter contains sites for transcription factors not present in the APLP2 promoter, including a potential heat shock element and an overlapping AP-1/AP-4 site (14,16,18), suggesting that the transcriptional regulation of APLP2 and APP genes may be dissimilar. The presence of multiple transcription start sites, the absence of a typical TATA box, the high GC content, and the presence of GC-rich boxes places the APLP2 promoter in the class of promoters of housekeeping genes; these include the human, rat, and mouse APP genes (13,14,16,17), the adenosine deaminase gene (32), the dihydrofolate reductase gene (33), and the hamster prion gene (34).
Recently, the upstream AP-1 site (position Ϫ350 with respect to the predominant transcription start site) in the APP promoter has been implicated in protein kinase C mediated upregulation of APP gene expression (35). The AP-1 binding activity is thought to be composed of Jun-Jun homodimers. Interleukin-1, nerve growth factor, and retinoic acid, agents known to increase APP gene expression, have been shown to induce c-jun and c-fos expression and cause transcriptional activation of target genes through AP-1 sites (36 -39). Furthermore, interleukin-1 effects are thought to involve protein kinase C activation (40). It remains to be determined if APLP2 gene expression is also regulated by interleukin-1, nerve growth factor, and retinoic acid, particularly in view of the presence of a potential AP-1 site located at position Ϫ982.
99 bp of the Mouse APLP2 Promoter Is Sufficient to Direct High Levels of CAT Expression in N2a and Ltk Ϫ Cells-To identify regulatory sequences responsible for the expression of the mouse APLP2 gene, we constructed plasmids containing progressive 5Ј-deletions of the APLP2 promoter fused upstream of the bacterial reporter gene CAT, as diagrammed in Fig. 4B. Equimolar amounts of each construct were transfected into mouse neuroblastoma (N2a) (Fig. 4C) and mouse fibroblast (Ltk Ϫ ) (Fig. 4D) cells. RT-PCR analysis of cytoplasmic RNA from mouse N2a and mouse Ltk Ϫ cells with degenerate primers which hybridize to both APLP2 and APP mRNA revealed that these two cell lines express moderate levels of endogenous APLP2 mRNA (Fig. 4A, lanes 1 and 2). Hence, we concluded that these cell lines would be appropriate for analysis of the APLP2 promoter.
Progressive 5Ј-deletions from position Ϫ971 to position Ϫ99, with respect to the predominant transcription start site, had no significant effect on promoter activity in either of the two cell lines tested. These findings suggest that in N2a and Ltk Ϫ cells, 99 bp of the APLP2 promoter are sufficient for directing high FIG. 2. Mapping of the 5 termini of APLP2 mRNA. A, diagram of S1 nuclease protection assay. A 533-bp KpnI-HindIII restriction fragment from the mouse APLP2 promoter was subcloned into Bluescript KS ϩ . The locations of restriction sites are with respect to the translation start site (ATG). S1 nuclease probe was prepared by 32 P end labeling at HindIII after linearizing the clone with HindIII. B, S1 nuclease protection assay of poly(A) ϩ RNA from CHO cells (lane 1), of total RNA from mouse thymus, heart, brain, liver, kidney, lung, testes, and spleen (lanes 2-9, respectively). tRNA served as negative control (lane levels of promoter activity. Similarly, studies that analyzed progressive 5Ј-deletions of the APP promoter from human, mouse, and rat have shown that reporter gene expression levels remained fairly constant up to approximately 100 bp upstream of the predominant transcription start site (13,15,18).
In summary, we have localized APLP2 to the proximal region of mouse chromosome 9, characterized ϳ1.2 kb of the APLP2 promoter, and shown it to contain features characteristic of promoters in the class of housekeeping genes. We further showed that 99 bp upstream of the predominant transcription start site are sufficient to direct high levels of promoter activity.
Given the similarities in overall structure of the APLP2 and APP promoters and the minimal sequence requirements for transcription initiation, it is highly likely that additional sequence elements distal to the regions analyzed here are responsible for differential expression of APLP2/APP in specific neuronal populations or systemic organs (i.e. liver). Further studies will be directed toward using transgenic strategies with larger genomic fragments to clarify these issues with the eventual goal of identifying transcription factors responsible for mediating basal level of APLP2 gene expression.