INTRODUCTION

Top Abstract Introduction Procedures Results & Discussion References

Amyloid -peptide (A)¹ of 39-42 amino acid residues comprises the major proteinaceous component of amyloid deposits in the brains of patients with Alzheimer's disease (1-6). The deposition of A aggregates (fibrils) is believed to be an early and critical event in the pathogenesis of Alzheimer's disease. The mechanisms by which A aggregates exert their detrimental effects are not well understood, but may involve effects through interactions with specific cell-surface receptors or binding proteins. Several receptors and binding proteins have been reported to interact with A, but none appears to be able to discriminate A monomers from A aggregates (7-9).

Recently, we have identified a putative TGF- active-site motif (WSXD) in TGF- isoforms (TGF-₁ and TGF-₂) (10). Synthetic peptides containing this motif in the middle of the peptide exhibit TGF- antagonist activity. Multiple conjugation of these peptides to carrier proteins not only enhances TGF- antagonist activity but also confers partial TGF- agonist activity (10). Since A contains a motif (FAED) that is similar to the putative TGF- active-site motif (WSXD) and since A aggregates would provide multivalencies with many copies of the putative active-site motif (11), we hypothesize that the A monomer and A aggregates may function as TGF- antagonist and partial TGF- agonist, analogous to previously described monovalent and multivalent TGF- peptide antagonist/partial agonist, respectively (10). To test this hypothesis, we investigated the TGF- antagonist/agonist activity of the A-(1-40) monomer and A-(1-40)-bovine serum albumin conjugate (A-(1-40)-BSA) which contains ~5-10 A-(1-40) peptides per molecule of protein and mimics A aggregates in multivalencies (11). In this communication, we demonstrate that A-(1-40) monomers inhibited ¹²⁵I-labeled TGF-₁ binding to cell-surface TGF- receptors in mink lung epithelial cells (Mv1Lu cells). We also show that A-(1-40)-BSA exhibited a potent cytotoxicity toward bovine cerebral endothelial (BCE) cells and rat post-mitotic differentiated hippocampal neuronal cells (H19-7 cells), and strongly inhibited ¹²⁵I-TGF- binding to cell-surface TGF- receptors and TGF--induced expression of plasminogen activator inhibitor 1 (PAI-1). A-(1-40)-BSA but not A-(1-40) monomers inhibited cellular proliferation of Mv1Lu cells.

	EXPERIMENTAL PROCEDURES

Top Abstract Introduction Procedures Results & Discussion References

Materials-- Na¹²⁵I (17 Ci/mg) and [methyl-³H]thymidine (67 Ci/mmol) were purchased from ICN Radiochemicals (Irvine, CA). High molecular mass protein standards (myosin, 205 kDa; -galactosidase, 116 kDa; phosphorylase, 97 kDa; bovine serum albumin, 66 kDa), A-(1-40), A-(1-16), and A-(1-20) were obtained from Sigma. A-(1-40), A-(25-35), and A-(12-28) were obtained from Bachem Bioscience, Inc. (King of Prussia, PA). Disuccinimidyl suberate was obtained from Pierce. TGF-₁ was purchased from Austral Biologicals (San Ramon, CA). ₁²⁵-(41-65), a synthetic pentacosapeptide with an amino acid sequence corresponding to the 41st to 65th amino acid residues of TGF-₁, was synthesized as described previously (10). Mv1Lu cells were grown in 10% fetal calf serum in Dulbecco's modified Eagle's medium.

Preparation of BSA Conjugates of A-(1-40) and A Fragments-- About 0.25 µmol of A-(1-40), A-(1-16), A-(12-28), or A-(25-35) dissolved in 167 µl of H₂O were mixed with 133 µl of 0.1 M NaHCO₃ and 150 µl of 0.1 M NaHCO₃ containing 0.25 mg of BSA. After adjusting the pH to 7.8-8.0, 10 µl of 27 mM disuccinimidyl suberate (a bifunctional cross-linking agent) in Me₂SO was added into the solution. After mixing at 4 °C for 16 h, 50 µl of 1 M ethanolamine were added, and the reaction mixture was mixed at room temperature for 2 h, then dialyzed (with dialysis tubing, M_r cutoff, 25,000) against 2 liters of 0.1 M NaHCO₃ (the pH was adjusted at 8.0). The dialysates were changed four times. The BSA conjugates were kept at 4 °C prior to use and were determined to contain ~5-10 peptides per molecule of protein based on analyses of amino acid composition and SDS-polyacrylamide gel electrophoresis (10).

Specific Binding of ¹²⁵I-TGF-₁ to Mv1Lu Cells-- ¹²⁵I-TGF-₁ was prepared by iodination of TGF-₁ with Na¹²⁵I and chloramine T according to our published procedure (10, 12, 13). ¹²⁵I-TGF-₁ binding to cell-surface TGF- receptors was assayed by incubating Mv1Lu cells with 0.1 nM ¹²⁵I-TGF-₁ in the presence of various concentrations of A-(1-40), A fragments, and their BSA conjugates at 0 °C for 2.5 h. The specific binding of ¹²⁵I-TGF-₁ to cell-surface TGF- receptors was estimated as described previously (10, 12, 13).

¹²⁵I-TGF-₁ Affinity Labeling of Cell-surface TGF- Receptors in Mv1Lu Cells-- The ¹²⁵I-TGF-₁ affinity labeling of cell-surface TGF- receptors in Mv1Lu cells was performed as described previously (10, 12, 13). After affinity labeling, the labeled TGF- receptors were analyzed by 5% SDS-polyacrylamide gel electrophoresis and autoradiography.

[methyl-³H]Thymidine Incorporation and Northern Blot Analysis-- The [methyl-³H]thymidine incorporation into cellular DNA and Northern blot analysis of PAI-1 and glyceraldehyde-3-phosphate dehydrogenase were performed as described previously (10, 13). The relative intensity of transcript on the autoradiogram was quantitated by a PhosphorImager.

Cytotoxicity Assay Using BCE Cells and Rat Postmitotic Differentiated H19-7 Cells-- BCE cells were prepared from bovine brain as described previously (14) and cultured in Dulbecco's modified Eagle's medium containing 10% fetal calf serum, heparin (0.5 mg/ml), and endothelial growth supplements (75 µg/ml). Rat hippocampal progenitor cells with neuronal lineage (H19-7 cells) were immortalized with a temperature-sensitive SV40 large T antigen provided by Drs. Eva M. Eves and Marsha R. Rosner, Ben May Institute for Cancer Research, University of Chicago (15). H19-7 cells were cultured at 33 °C in Dulbecco's modified Eagle's medium containing 10% fetal calf serum, 50 µg/ml streptomycin, 50 units/ml penicillin, and 200 µg/ml G418. H19-7 cells grown under this condition were defined as mitotic progenitor cells with neuronal lineage. To achieve a postmitotic differentiated state, N2 supplement and basic fibroblast growth factor (10 ng/ml) were added to the medium, and the temperature shifted to the nonpermissive range for a temperature-sensitive SV40 large T antigen (39 °C) for 24 h to allow differentiation (15) before cytotoxicity assay.

	RESULTS AND DISCUSSION

Top Abstract Introduction Procedures Results & Discussion References

A and its fragments have been found to exert cytotoxic and trophic effects on cells in culture. A-(25-35) mimics the cytotoxic activity of A (5), whereas A-(1-28), A-(25-28), and A-(1-40) exhibit acetylcholine release inhibitory activity (4). The sequence of VFF (residues 18-20) has been implicated in mediating the amnestic activity of A fragments (17). These cytotoxic and trophic effects of A appear to be mediated by different domains, e.g. the sequences of residues 25-35, 25-28, and 18-20 (4, 5, 17). We noted that, in addition to these domains, A possesses a motif (FAED, 20th to 23rd amino acid residues), which is similar to the WSXD putative TGF- active-site motif. Known TGF- peptide antagonists ₁²⁵-(41-65) and ₂²⁵-(41-65) that contain this motif are synthetic peptides with amino acid sequences corresponding to the 41st to 65th residues of TGF-1 and TGF-2, respectively (Fig. 1) (10). Replacement of the tryptophan residue in the motif by a phenylalanine residue does not affect the antagonist activity of ₁²⁵-(41-65).² Thus, the FAED in the A monomer may be a functional TGF- active-site motif. To test this possibility, we determined the effects of A-(1-40) monomers and A fragments possessing and lacking the FAED motif on ¹²⁵I-TGF-₁ binding to cell-surface TGF- receptors in Mv1Lu cells, a standard model system for investigating TGF- receptor types and TGF--induced cellular responses (12, 13). As shown in Fig. 2, A-(1-40), and A-(12-28), both of which contain the motif, exhibited ¹²⁵I-TGF-₁ receptor binding inhibitory activities with IC₅₀ of ~3 and ~30 µM, respectively. A-(25-35), A-(1-20), and A-(1-16), all of which lack the motif, failed to show ¹²⁵I-TGF-₁ binding inhibitory activity at any concentration up to 30 µM. These results indicate that A-(1-40) possesses a functional TGF- active-site motif.

View larger version (16K): [in this window] [in a new window]   Fig. 1.   Amino acid sequences of TGF- peptide antagonists (125-(41-65) and 225-(41-65)), A-(1-40), and A fragments. The amino acid residues underlined are the putative TGF- active-site motifs. Identical amino acid residues are boxed with a solid line, whereas functionally homologous residues are boxed with a broken line. Dutch, Dutch-type Alzheimer's disease.

View larger version (17K): [in this window] [in a new window]   Fig. 2.   Effects of A-(1-40) and A fragments on 125I-TGF-1 binding in Mv1Lu cells. Cells were incubated with 0.1 nM 125I-TGF-1 and various concentrations of A-(1-40), A-(12-28), A-(25-35), A-(1-20), and A-(1-16) at 0 °C for 2.5 h. The specific binding of 125I-TGF- without A-(1-40) and A fragments was taken as 0% inhibition (5,429 ± 780 cpm/well). The error bars are means ± S.D. of triplicate cell cultures. The data are representative of seven experiments that gave comparable results.

It has been reported that the aggregation of A monomers is not easily controlled in vitro, as it is strongly affected by A concentration, pH, ionic strength, and incubation time (18-22). In order to produce a multivalent stable inhibitor, we prepared A-(1-40)-bovine serum albumin conjugate (A-(1-40)-BSA, M_r ~90,000-100,000) containing ~5-10 A-(1-40) per molecule of BSA according to the procedure of Huang, et al. (10). Unlike the rather unstable A-(1-40) aggregates, A-(1-40)-BSA is stable (at 4 °C) for at least a few weeks and has a consistent valence due to the covalent nature of the attachment of A-(1-40) to BSA. This conjugate is meant to mimic A-(1-40) aggregates by possessing multiple valences of A-(1-40) per molecule and cytotoxicity toward BCE cells and rat postmitotic differentiated H19-7 cells. As shown in Fig. 3, A-(1-40)-BSA exhibited cytotoxicity toward BCE cells and H19-7 cells that was ~670 times more potent than that of the A-(1-40) monomer (Fig. 3, A and B). A-(1-40)-BSA at 75 nM was as potent as 50 µM A-(1-40) in causing cell death of BCE cells and H19-7 cells as determined by the 3-(4,5-dimethyl thiazol-2-yl)-2,5-diphenyl tetrazolium bromide assay (16) and trypan blue exclusion assay (14). Similar cytotoxic effects of A-(1-40) and A-(1-40)-BSA in BCE cells and H19-7 cells were also noted using lactate dehydrogenase assay (14) and 3-(4,5-dimethyl thiazol-2-yl)-2,5-diphenyl tetrazolium bromide assay, respectively (data not shown).

View larger version (37K): [in this window] [in a new window]   Fig. 3.   Cytotoxic effects of A-(1-40) and A-(1-40)-BSA in BCE cells and H19-7 cells. A, BCE cells were exposed to A-(1-40) or A-(1-40)-BSA at concentrations indicated for 48 h. The extent of cell survival (mean ± S.D.) was determined in quadruplicate cell cultures by the MTT assay. The asterisk denotes a p value of <0.05. The data are representative of two experiments that gave comparable results. B, H19-7 cells were exposed to A-(1-40) or A-(1-40)-BSA at concentrations indicated for 24 h. The extent of cell survival (mean ± S.D.) was determined in quadruplicate cell cultures by the trypan blue exclusion assay. The asterisk denotes a p value of <0.05. The data are representative of two experiments that gave comparable results.

To determine the TGF- antagonist activity of A-(1-40)-BSA, we examined the ¹²⁵I-TGF-₁ receptor binding inhibitory activity of A-(1-40)-BSA in Mv1Lu cells. As shown in Fig. 4A, A-(1-40)-BSA inhibited ¹²⁵I-TGF-₁ binding to cell-surface TGF- receptors with an IC₅₀ of ~30 nM. This is 100-fold more potent than A-(1-40) monomers. The BSA conjugate of A-(12-28), which contains the FAED motif, showed somewhat weaker ¹²⁵I-TGF-₁ binding inhibitory activity (IC₅₀ ~ 0.3 µM). The BSA conjugates of the A-(1-16) and A-(25-35) fragments, which lack the motif, did not inhibit ¹²⁵I-TGF-₁ binding. In control experiments, BSA subjected to conjugation conditions without peptide also had no inhibitory activity (10). ¹²⁵I-TGF-₁ affinity labeling analysis revealed that A-(1-40)-BSA conjugate almost completely inhibited ¹²⁵I-TGF-₁ binding to type I, type II, type III, and type V TGF- receptors (TR -I, TR-II, TR-III, and TR-V) at 0.1 µM, whereas A-(1-40) only partially inhibited ¹²⁵I-TGF-₁ binding to its receptors at 5 µM (Fig. 4B). These results indicate that the multiple valencies enhance the ¹²⁵I-TGF-₁ binding-inhibitory activity of A-(1-40).

View larger version (23K): [in this window] [in a new window]   Fig. 4.   Effects of the BSA conjugates of A-(1-40) and its fragments on 125I-TGF-1 binding (A) and 125I-TGF-1-affinity labeling (B) in Mv1Lu cells. A, cells were incubated with various concentrations of the BSA conjugates of A-(1-40), A-(12-28), A-(25-35), or A-(1-16) at 0 °C for 25 h. The specific binding of 125I-TGF-1 without the BSA conjugates was taken as 0% inhibition. The error bars are means ± S.D. of triplicate cell cultures. The data are representative of five experiments that gave comparable results. B, after 125I-TGF-1 binding, in the absence (control) and presence of 10 µM 125(41-65), 0.1 µM A-(1-40)-BSA, or 5 µM A-(1-40), the 125I-TGF-1 affinity labeling of cell-surface TGF- receptors was performed. The 125I-TGF-1 affinity-labeled receptors were analyzed by 5% SDS-polyacrylamide gel electrophoresis and autoradiography. The brackets indicate the locations of the types I, II, III, and V TGF- receptors (TR-I, TR-II, TR-III, and TR-V).

Dimerization is known to be required for TGF- activity (23) and multivalent TGF- peptide antagonists have been shown to exhibit partial TGF- agonist activity as assayed by growth inhibition (10). We therefore examined the TGF- agonist activity of A-(1-40)-BSA by measuring its inhibition of DNA synthesis using Mv1Lu cells. As shown in Fig. 5A, 0.35 µM of multivalent A-(1-40)-BSA produced ~35% inhibition of [methyl-³H]thymidine incorporation into DNA of Mv1Lu cells. Neither A-(1-40) monomers (0.35 µM) nor BSA conjugated in the absence of peptide (at any concentration up to 10 µM) affected DNA synthesis in this system. The DNA synthesis inhibition induced by 0.35 µM A-(1-40)-BSA was blocked in the presence of 10 µM ₁²⁵-(41-65), a specific TGF- peptide antagonist (data not shown). These results suggest that multiple valencies of A-(1-40) confer TGF- agonist activity, i.e. inhibit cellular proliferation as measured by DNA synthesis. To support this suggestion, we determined the effect of A-(1-40)-BSA on DNA synthesis of type I TGF- receptor-defective mutant and wild-type mink lung epithelial cells (R1B and Mv1Lu cells) (12, 24). If the DNA synthesis inhibition by A-(1-40)-BSA is mediated by cell surface TGF- receptors, R1B cells, which lack expression of the functional type I TGF- receptor (12, 24), should respond very little if any to A-(1-40)-BSA DNA synthesis inhibition. As shown in Table I, A-(1-40)-BSA did not significantly affect DNA synthesis of R1B cells. This result is consistent with the suggestion that the A-(1-40)-BSA exhibits TGF- agonist activity in growth inhibition.

View larger version (29K): [in this window] [in a new window]   Fig. 5.   Effects of BSA conjugates of A-(1-40) and A fragments on DNA synthesis (A) and TGF-1-induced PAI-1 expression (B) in Mv1Lu cells. A, cells were incubated with various concentrations of A-(1-40) and BSA conjugates of A-(1-40), A-(12-28), A-(25-35), and A-(1-20) at 37 °C for 16 h. DNA synthesis was determined by measuring [methyl-3H]thymidine incorporation into cellular DNA. The [methyl-3H]thymidine incorporations in the presence and absence of 10 pM TGF-1 were taken as 100 and 0% inhibition (2,242 ± 679 and 25,493 ± 1,200 cpm/well, respectively). The error bars are means ± S.D. of triplicate cell cultures. The data are representative of six experiments which gave comparable results. B, cells were treated with 0.5 pM TGF-1 in the presence of various concentrations of A-(1-40)-BSA at 37 °C for 2 h. PAI-1 expression was determined by Northern blot analysis. The expression of glyceraldehyde-3-phosphate dehydrogenase (G3PDH) was used as a control. The relative intensities of transcripts on the autoradiograms were estimated by a PhosphorImager.

View this table: [in this window] [in a new window]   Table I Effect of A-(1-40)-BSA and A-(1-40) on DNA synthesis of type I TGF- receptor-defective and wild-type mink lung epithelial cells (R1B and Mv1Lu cells)

The transcriptional expression of PAI-1 is known to be stimulated by TGF-₁ (25-27). To determine whether A-(1-40)-BSA interferes with this effect, Mv1Lu cells were treated with 0.5 pM TGF-₁ plus various concentrations of A-(1-40)-BSA for 2 h at 37 °C. Northern blot analysis was then performed. As shown in Fig. 5B, A-(1-40)-BSA did not stimulate the expression of PAI-1 (lane 2 versus lane 1). However, A-(1-40)-BSA diminished the PAI-1 expression stimulated by 0.5 pM TGF- in a dose-dependent manner (lanes 4-6). This suggests that A-(1-40)-BSA can bind to TGF- receptors and function as an antagonist for TGF- as assayed by transcriptional activation.

In summary, A-(1-40)-BSA is a stable multivalent analogue of naturally occurring A aggregates seen in Alzheimer's disease lesions and is more potent than A-(1-40) as a TGF- antagonist that blocks TGF- binding to TGF- receptors. The cytotoxicity of A-(1-40)-BSA toward BCE cells and H19-7 cells is ~670 times more potent than that of A-(1-40). Furthermore, A-(1-40)-BSA, which has multiple A-(1-40) peptides per BSA molecule, possesses partial TGF- agonist activity (growth inhibition). These results suggest that A monomers and A aggregates may participate in the pathogenesis of neuronal death in Alzheimer's disease patients through their TGF- antagonist and agonist activities. TGF- has been shown to protect neurons from cell death (28-32). Since TGF- expression has been detected in Alzheimer's disease lesions (28, 33-35), we hypothesize that the TGF- antagonist activity (TGF- receptor binding inhibitory activity) of A-(1-40) monomers and aggregates may counteract this neuroprotective effect of TGF-. As both glial cells and monocytes have been shown to express TGF- (35) and to respond to TGF- stimulation (28), the partial TGF- agonist activity (growth inhibition) of A aggregates may also play an important role in the chemotaxis and activation of astrocytes and microglia that are associated with Alzheimer's disease.

The familial Alzheimer's disease (FAD) (36) and Dutch-type Alzheimer's disease (37) patients may provide some clues to the structure/function relationship of the putative TGF- active-site motif (FAED) in A, since these patients have mutations within this motif (Fig. 1). The mutations in both FAD (A692G) and Dutch-type (E693Q) patients may provide a TGF- active-site motif with particularly robust function on the basis of studies of various motifs in synthetic TGF- peptide antagonists² (10). If the 2nd and 3rd amino acid residues in the motif are amino acids with small side chains (Gly, Ser, Cys, and Ala residues) and noncharged amino acids, respectively, in the TGF- peptide antagonist motif (WXXD), the potency of TGF- antagonism is enhanced. Determining the TGF- activities of FAD and Dutch-type A mutant peptides would test the hypothesis that the TGF- activities of these peptides are important in the mechanism of A in the neuronal degeneration of Alzheimer's disease.