HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Full Text (PDF) All Versions of this Article: 19/6/653 04-2900fjev1    most recent Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Lahiri, D. K. Articles by Maloney, B. Search for Related Content PubMed PubMed Citation Articles by Lahiri, D. K. Articles by Maloney, B.

Characterization of the APP proximal promoter and 5'-untranslated regions: identification of cell type-specific domains and implications in APP gene expression and Alzheimer’s disease

Debomoy K. Lahiri^*,,1, Yuan-Wen Ge^* and Bryan Maloney^*

Departments of
^* Psychiatry and
Medical and Molecular Genetics, Institute of Psychiatric Research, Indiana University School of Medicine, Indianapolis, Indiana, USA

¹Correspondence: E-mail: dlahiri{at}iupui.edu

SPECIFIC AIMS

Neuritic plaque, mostly derived from the amyloid ß peptide (Aß), is one of the main characteristics of Alzheimer’s disease (AD). Our goal is to functionally characterize the proximal promoter region (PPR) and the 5'-untranslated region (5'-UTR) of the gene encoding the Aß precursor protein (APP). We characterized these regions by DNA elecrophoretic mobility shift assay (EMSA) and reporter gene assay. We studied the role of a recently discovered CAGA box that is unique to the 5'-UTR of APP mRNA sequences of plaque-forming mammalian species. We define the PPR as running from –46/–1 and the UTR as running from +1/+144 with +1 as the transcription start site.

PRINCIPAL FINDINGS

1. Distinct DNA-protein interaction of PPR and 5'-UTR fragments of the APP gene in nuclear extracts
We probed six regions containing the proximal promoter region and/or 5'-UTR in human neuroblastoma (NB) and HeLa nuclear extracts: –46/144, –46/100, –46/54, 54/144, 54/100, and 100/144 (Fig. 1 A). Two differences appeared among PPR/UTR fragments in terms of DNA-protein complex formation (Fig. 1B ). Band I (lanes 7–22) appeared with the –46/144, –46/100, –46/54, and 54/144 fragments. (Minor variations in migration of DNA-protein complexes corresponded closely to migration of unbound probe, indicating that pattern is due to probe length, not alterations in protein binding.) Breadth of this band suggests it is from a DNA-multiprotein complex rather than DNA and one protein. Band II (lanes 11, 15, 16, 23) is likely to represent one protein binding specifically to the 54/100 fragment. This protein may take part in interactions seen for the –46/144, –46/100, and 54/144 fragments.

View larger version (64K): [in this window] [in a new window]   Figure 1. Generation of APP gene PPR/UTR fragments and EMSA to detect nuclear protein(s) interacting with different regions of the APP PPR and 5'-UTR. A) Diagram of all PPR/UTR (–46/144) fragment with pertinent restriction enzyme sites indicated. Thick bar represents the amyloid CAGA box. Numbers shown are bp, TSS being +1. PPR/UTR-based expression clones derived from the respective fragments are indicated according to length and direction. These fragments were subsequently cloned into a reporter gene (chloramphenicol acetyltransferase: CAT) vector. B) EMSA was performed with 6 different PPR/5'-UTR probes and analyzed in a 5% polyacrylamide gel as indicated. EMSA was carried out with each radiolabeled probe in the presence of no nuclear extract (lanes 1–6), nuclear extracts from NB (lanes 7–12), or from HeLa (lanes 13–24) cell lines, in the presence (lanes 19–24) or absence (lanes 1–18) of TPA. Arrows indicate DNA-protein complex bands. Arrows that share the same Roman numeral indicate presumably identical proteins from each extract.

2. Role of AP2 transcription factor with APP 5'-UTR
In HeLa cell nuclear extracts, strong interaction (Fig. 1 , lanes 13, 14) appeared to correspond to band I of NB interactions (lanes 7–10). A weaker band (lanes 15, 16) more than likely corresponds to this band. There may be an additional interaction (lanes 13, 14, band III) absent in NB extracts. Induction with TPA changed HeLa binding profile, including band intensities, to more closely resemble that in NB nuclear extract. The complexes for –46/144 and –46/100 (lanes 19, 20) even more closely resemble band I of NB extracts (lanes 7, 8). Bands from the –46/54 and 54/144 fragments (lanes 21, 22) were stronger and migrated in accord with the corresponding fragments incubated with NB extracts (lanes 9, 10). A faint but distinct band appeared (lane 23) with the 54/100 fragment at a position analogous to band II in NB (lane 11). Transcription factor database search revealed 100% homology to the activator protein 2 (AP2) binding sequence (CCCGCGC) at 43/49 (reversed), 76/82 (reversed), and 132/138. Therefore, we propose AP2 as a likely candidate for this pathway.

3. Functional activity of the PPR/UTR fragments is position and cell type specific
We transfected NB and PC12 cell cultures with different PPR/UTR CAT reporter fusion constructs (Fig. 1A ) and the pSV₂CAT vector. CAT reporter protein level was measured by ELISA and normalized to total protein. The PC12 (Fig. 2 A) and NB (Fig. 2B ) transfections displayed differences between cell lines and among specific PPR/UTR fragments. In PC12 cells, the –46/144 fragment enhanced CAT expression to nearly 6-fold that of the base vector. In NB cells, very little enhancement activity was seen. The "basal" (–46/54) fragment, by contrast, functioned as an enhancer in PC12 and NB cells, driving 10- and 2.5-fold CAT expression, respectively.

View larger version (25K): [in this window] [in a new window]   Figure 2. Expression of PPR/UTR deletion series–CAT fusion clones in PC12 and NB cells. PC12 (A) or NB (B) cells were transfected, respectively, with all deletion clones generated and with unaltered pSV2CAT. Reporter CAT protein expression of equal protein samples from transfected cells was measured by ELISA. Results were analyzed via SAS. Letters indicate Waller-Duncan groups. Samples sharing the same letter are not significantly different from each other.

The remainder of the 5'–UTR (54/144) consists of two cell type-specific (CTS) regions that function differently in PC12 and NB cells. CTS-I (54/100) weakly enhanced expression (2.5-fold empty vector) on its own in PC12 cells (not statistically significant). It suppressed expression when combined with the basal domain. Its activity in NB cells was somewhat different, suppressing expression to below pSV₂CAT on its own and suppressing activity of the basal domain. CTS-II (100/144) has no apparent activity on its own in the forward orientation in PC12 cells, but acts as a suppressor in NB cells. In combination with the basal and CTS-I domains (–46/100), CTS-II has no specific activity in PC12 but suppresses activity of these two domains in NB.

CTS-I and CTS-II combined (54/144), without the basal domain, enhance reporter expression in PC12 and NB cells. CTS-I + II in reverse orientation (144/54) has no significant enhancement activity in PC12 cells and suppresses activity in NB cells. CTS-II in reverse orientation (144/100), enhanced reporter gene expression in PC12 cells but not in NB cells. Figure 3 summarizes the domain locations.

View larger version (25K): [in this window] [in a new window]   Figure 3. Functional domains of the APP gene PPR/5'-UTR. Scale schematic of the functional domains of PPR/5'-UTR as determined by EMSA and reporter gene expression assays. Numbers shown are bp, TSS being +1. The basal (–46/54), including possible Bsup (–46/–1) and Benh (+1/54), CTS-I (54/100), and CTS-II (100/144) domains are indicated by brackets above the sequence. Position of a bipartite iron responsive element (IRE) is indicated by brackets beneath the domain diagram. Other sites of interest are indicated by position on dotted lines corresponding in length to the –46/144 PPR/UTR fragment. Solid black thick bar: amyloid CAGA box. Falling diagonal: AGAC sequence CAGA box. Square crosshatch: TCTG CAGA box. Diagonal crosshatch: putative SP1 binding sites. Vertical bars: IL-1 acute box. Check pattern: consensus AP2 binding sequences. Dotted box: TGF-ß binding site. Diagonal check: potential stem loop structure near upstream of the TSS.

4. Interaction of the 54/144 fragment is cell type specific and cytokine dependent and appears in mouse brain extract
We investigated cell line specificity of nuclear protein interactions with the combined CTS-I/CTS-II domains (54/144) of the APP 5'-UTR. The formation of distinct DNA-protein complexes was observed with radiolabeled 54/144 fragment in nuclear extracts from human astrocyte (U373) and NB cells, and mouse brain in a cell type-specific manner. NB and mouse brain bands were confirmed by competition EMSA. This band is fairly strong in U373 extracts. Another band is faint in mouse fibroblast-like (NIH-3T3) nuclear extracts but appears distinctly when nuclear extracts are derived from the same cells induced with transforming growth factor-ß (TGF-ß).

5. Mutation of CAGA box "amyloid" alters DNA-protein interaction and reporter gene clone expression
To investigate whether the amyloid CAGA (83/86) box within the CTS-I domain (54/100) has specific effects on DNA-protein interaction, complementary oligomer pairs were synthesized to produce a battery of CAGA box amyloid mutants (with 8–12 flanking nt on either side) and tested in PC12 and NB cell nuclear extracts.

There is a distinct DNA-protein interaction for wild-type probe and PC12 and a much weaker interaction between wild-type probe and NB nuclear extracts. This is notable in light of PC12 vs. NB cell line specific differential functional activity of the CTS-I domain (Figs. 2 , 3) . EMSA revealed dramatic differences among mutants and less visible difference between cell nuclear extracts. Most mutations increased apparent DNA-protein interaction for NB and PC12 nuclear extracts when compared to wild-type.

Functional assay revealed that certain CAGA mutants alter expression from the wild-type clone, which supports a role for the CAGA box in APP gene expression. This variation also depended on each clone’s orientation.

CONCLUSIONS AND SIGNIFICANCE

We have characterized the PPR and 5'-UTR of the APP gene by DNA-nuclear protein and functional assays. EMSA results show specific DNA-protein interaction in most of these regions. Specific binding with nuclear proteins was lost with only the 100/144 fragment. TPA induction of HeLa cells altered nuclear extract binding pattern to more closely resemble neuronal cells. This is consistent with the presence of three putative AP2 sites in the 5'-UTR sequence. As these AP2 sites overlap with SP1 sites, the CTS I/II regions may be a site of SP1 and AP2 coregulation.

Compared with control vector, we observed significantly higher levels of reporter gene activity in PC12 and NB cell lines with all constructs that contained the PPR with the exception of –46/144 in NB cells. The –46/54 sequence resulted in the greatest expression in both cell lines and has thus been named basal. This is not to be construed to mean that no possibility of inducible activity resides in this region, especially since a potentially active AP2 site (43/49) was found.

One of the constructs that lacked PPR sequence, 54/144, displayed high expression in both cell lines similar to the fragment that contained the PPR to the end of an IL-1 acute box (–46/100). Its expression was equal to or significantly higher than expression driven by the full-length (–46/144) fragment. The 54/100 fragment had different expression patterns depending on cell line. We have designated it CTS-I to reflect this cell line specificity. However, further work is necessary to confirm its role.

The 100/144 (CTS II) fragment showed no significantly different reporter activity from the vector in PC12 cells, but suppressed activity in NB cells. It modified the activity of CTS-I in NB and PC12 cells and suppressed basal domain activity for NB but not PC12 cells.

We propose a model of functional domains of the APP PPR/UTR. The minimal promoter is within a basal domain (–46/54). The +1/104 sequence of the rhesus monkey APP 5'-UTR drove expression in undifferentiated PC12 cells (12-fold over unmodified vector); the –47/104 sequence in that reporter system drove expression at 75% of +1/104, permitting dividing basal further into basal suppressor (Bsup) at –46/–1 and basal enhancer (Benh) at +1/54. However, the rhesus sequence has 12 substitutions and a 3-base deletion, making assignment of subdomains to the human PPR/UTR tentative.

We investigated potential activity of CAGA box amyloid (83/86) in CTS-I. Altering the CAGA box or adjacent sequence could alter specific DNA-protein interaction and alter reporter gene clone expression. This finding points to a defined potential target to investigate for drug regulation of APP expression.

These results lead us to propose that the 5'-UTR of the APP gene functions as the core of a UTRosome, in which a gene’s 5'-UTR and PPR may interact as a unit to regulate expression at the transcriptional and post-transcriptional levels.

FOOTNOTES

To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.04-2900fje;

This article has been cited by other articles:

				J. Hardy and A. Myers Genetic variability in expression of proteins and the risk of sporadic neurologic diseases Neurology, February 27, 2007; 68(9): 632 - 633. [Full Text] [PDF]

				S. Bandyopadhyay, J. Ni, A. Ruggiero, K. Walshe, M. S. Rogers, N. Chattopadhyay, M. A. Glicksman, and J. T. Rogers A High-Throughput Drug Screen Targeted to the 5'Untranslated Region of Alzheimer Amyloid Precursor Protein mRNA J Biomol Screen, August 1, 2006; 11(5): 469 - 480. [Abstract] [PDF]

This Article Full Text (PDF) All Versions of this Article: 19/6/653 04-2900fjev1    most recent Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Lahiri, D. K. Articles by Maloney, B. Search for Related Content PubMed PubMed Citation Articles by Lahiri, D. K. Articles by Maloney, B.