Covalent capture: a new tool for the purification of synthetic and recombinant polypeptides.

BACKGROUND: Purification of polypeptides and proteins derived from recombinant DNA techniques and of long synthetic polypeptides often represents a challenge. Affinity methods exist, but generally require addition of a large recognition unit to the target protein and use of expensive purification media. Use of large units is dictated by the characteristics of non-covalent complexes, where the energy necessary to form the complex derives from the sum of multiple weak energy interactions. Covalent interactions in contrast are of high energy, even when only a few bonds are formed. We decided to explore the use of the reversible covalent bond formed between N-terminal cysteine and threonine residues with an aldehyde as a method of protein purification. RESULTS: A series of test peptides with N-terminal cysteine and threonine were captured by a polyethyleneglycol-polyacrylamide resin to which an aldehyde function had been grafted. Peptides with other amino acids at the N-terminus did not interact with the resin. A recombinant polypeptide with N-terminal cysteine was purified to 90% purity in one step. Polypeptides were eluted from the resin simply by adding a hydroxylamine derivative, which reacts with aldehyde functions to form an oxime. CONCLUSIONS: Polypeptides possessing N-terminal cysteine or threonine can be easily purified using this 'covalent capture' approach.

Rational engineering of a miniprotein that reproduces the core of the CD4 site interacting with HIV-1 envelope glycoprotein.

Protein-protein interacting surfaces are usually large and intricate, making the rational design of small mimetics of these interfaces a daunting problem. On the basis of a structural similarity between the CDR2-like loop of CD4 and the beta-hairpin region of a short scorpion toxin, scyllatoxin, we transferred the side chains of nine residues of CD4, central in the binding to HIV-1 envelope glycoprotein (gp120), to a structurally homologous region of the scorpion toxin scaffold. In competition experiments, the resulting 27-amino acid miniprotein inhibited binding of CD4 to gp120 with a 40 microM IC(50). Structural analysis by NMR showed that both the backbone of the chimeric beta-hairpin and the introduced side chains adopted conformations similar to those of the parent CD4. Systematic single mutations suggested that most CD4 residues from the CDR2-like loop were reproduced in the miniprotein, including the critical Phe-43. The structural and functional analysis performed suggested five additional mutations that, once incorporated in the miniprotein, increased its affinity for gp120 by 100-fold to an IC(50) of 0.1-1.0 microM, depending on viral strains. The resulting mini-CD4 inhibited infection of CD4(+) cells by different virus isolates. Thus, core regions of large protein-protein interfaces can be reproduced in miniprotein scaffolds, offering possibilities for the development of inhibitors of protein-protein interactions that may represent useful tools in biology and in drug discovery.

Novel miniproteins engineered by the transfer of active sites to small natural scaffolds.

Small multidisulfide-containing proteins are attractive structural templates to produce a biologically active conformation that mimics the binding surface of natural large proteins. In particular, the structural motif that is evolutionary conserved in all scorpion toxins has a small size (30-40 amino acid residues), a great structural stability, and high permissiveness for sequence mutation. This motif is composed of a beta-sheet and an alpha-helix bridged in the interior core by three disulfides. We have used this motif successfully to transfer within its beta-sheet new functional sites, including the curaremimetic loop of a snake neurotoxin and the CDR2-like site of human CD4. Accumulated evidence indicated that the two miniproteins produced, the curaremimetic miniprotein and the CD4 mimetic, contain the alpha/beta fold that is characteristic of the scaffold used and bind respectively to the acetylcholine receptor and to the envelope gp120 of HIV-1. Furthermore, the latter was shown to prevent viral infection of lymphocytes. These examples illustrate that, by the transfer of active sites to small and stable natural scaffolds, it is possible to engineer miniproteins reproducing, in part, the function of much larger proteins. Such miniproteins may be of great utility as tools in structure-function studies and as leads in drug design.

Synthetic full-length and truncated RANTES inhibit HIV-1 infection of primary macrophages.

OBJECTIVE: To determine the effect of beta-chemokines on HIV-1 infection of primary macrophages, and to search for chemokine derivatives devoid of biological effects but efficient at protecting CD4+ T lymphocytes and macrophages against HIV-1. DESIGN: Use of chemically synthesized molecules devoid of biological contaminants and monocyte-derived macrophages from healthy donors. METHODS: Full-length RANTES was chemically synthesized together with three derivatives, truncated of seven, eight and nine amino acids at the amino-terminus ([8-68]RANTES, [9-68]RANTES and [10-68]RANTES), which were tested for their biological activity and antiviral effects. RESULTS: Whereas full-length and truncated RANTES derivatives bound to beta-chemokine receptor CCR-5 with the same affinity as recombinant RANTES, the truncated forms were not chemotactic and acted as CCR-5 antagonists in this respect, although a partial agonist effect was noted on cell metabolism. Full-length RANTES and [8-68]RANTES protected T lymphocytes and macrophages from infection by HIV-1, although 10-fold higher concentrations of the truncated analogues were necessary to achieve the same effect as full-length RANTES. With regard to the effect of RANTES on HIV-1 infection of primary macrophages, our results contrast with most previously reported data. CONCLUSION: These data indicate that through binding to CCR-5, truncated RANTES derivatives that are devoid of detectable biological effects may represent candidates as drugs to protect both lymphocytes and macrophages from HIV- 1.

Consequence of the removal of evolutionary conserved disulfide bridges on the structure and function of charybdotoxin and evidence that particular cysteine spacings govern specific disulfide bond formation.

Scorpion toxins are miniglobular proteins containing a common structural motif formed by an alpha-helix on one face, an antiparallel beta-sheet on the opposite face, and three disulfide bonds making up most of its internal volume. We have investigated the role of these evolutionary conserved bonds by replacing each couple of bridged cysteine residues of the scorpion charybdotoxin by a pair of nonbridging L-alpha-aminobutyric acid (Aba) residues. Three analogues were obtained by solid-phase synthesis, Chab I, Chab II, and Chab III, containing the Aba residues in positions 7 and 28, 13 and 33, 17 and 35, respectively. Circular dichroism analysis showed that the purified Chab II acquired a conformation similar to that of charybdotoxin, while the Chab I and Chab III possess decreased nativelike characteristics. All analogues block single high-conductance Ca(2+)-activated K+ channels from rat skeletal muscle inserted into planar lipid bilayers, but with different potencies. Chab II is the most active analogue (KD = 8.0 x 10(-8) M), with a 9-fold lower affinity as compared to native charybdotoxin. Chab I and Chab III have, respectively, 180- and 580-fold lower affinity. Therefore, the removal of evolutionary conserved disulfide bridges does not prevent the toxin to adopt a functional and presumably nativelike structure. However, removal of one disulfide bond affects the yields of formation of correct pairing between the remaining cysteine residues, and only Chab I preserves the ability to form the native disulfide pairings with high efficiency. This is the only analogue to preserve particular spacings of three and one residue between the cysteines, which have been described to thermodynamically disfavor disulfide bond formation between the cysteines [Zhang R., and Snyder, G. H. (1989) J. Biol. Chem. 264, 18472-18479]. Therefore, we conclude that the position of the cysteine residues in the sequence of charybdotoxin, by disfavoring specific pairings and favoring others, may govern selective formation of specific disulfide bonds, thus, explaining the efficient folding properties of Chab I and of native charybdotoxin. The structural properties of the Chab analogues and the discovered role of the cysteine spacings have interesting implications in protein design and engineering.

Effect of RANTES on the infection of monocyte-derived primary macrophages by human immunodeficiency virus type 1 and type 2.

The effect of beta chemokines on human immunodeficiency virus type 1 (HIV-1) infection of primary macrophages is controversial, and their effect on HIV-2 infection of these cells has not yet been documented. We examined the effect of synthetic and recombinant regulated-on-activation, normal T cell-expressed and -secreted (RANTES) on HIV-1 and HIV-2 infection of primary monocyte-derived-macrophages (MDM) that were obtained as the adherent cells of 5-day cultures of blood mononuclear cells (PBMC), followed by 2-day culture without peripheral blood mononuclear cells (PBMCs) nor added cytokines. These MDM expressed CD4, CCR5 and CXCR4, the major coreceptors for HIV macrophage- and T cell-tropic isolates, respectively. Infection of MDM from different donors with HIV-1 or HIV-2 macrophage-tropic strains was reproducibly inhibited by RANTES. This inhibition depended on RANTES continuous presence in culture during and after infection. Treatment of MDM with RANTES just before or during, but not after, exposure to virus did not protect MDM from infection. When RANTES was added after MDM had been infected, and was continuously maintained in culture thereafter, no inhibition occurred and limited enhancement of infection could be observed. These data indicate that RANTES inhibits HIV-1 as well as HIV-2 infection of MDM, likely at a post-binding step, and support the role of CCR5 as the major coreceptor for HIV-1 and HIV-2 entry into primary macrophages.

Characterization of new polyclonal antibodies specific for 40 and 42 amino acid-long amyloid beta peptides: their use to examine the cell biology of presenilins and the immunohistochemistry of sporadic Alzheimer's disease and cerebral amyloid angiopathy cases.

BACKGROUND: In Alzheimer's disease (AD), the main histological lesion is a proteinaceous deposit, the senile plaque, which is mainly composed of a peptide called A beta. The aggregation process is thought to occur through enhanced concentration of A beta 40 or increased production of the more readily aggregating 42 amino acid-long A beta 42 species. MATERIALS AND METHODS: Specificity of the antibodies was assessed by dot blot, Western blot, ELISA, and immunoprecipitation procedures on synthetic and endogenous A beta produced by secreted HK293 cells. A beta and p3 production by wild-type and mutated presenilin 1-expressing cells transiently transfected with beta APP751 was monitored after metabolic labeling and immunoprecipitation procedures. Immunohistochemical analysis was performed on brains of sporadic and typical cerebrovascular amyloid angiopathy (CAA) cases. RESULTS: Dot and Western blot analyses indicate that IgG-purified fractions of antisera recognize native and denaturated A beta s. FCA3340 and FCA 3542 display full specificity for A beta 40 and A beta 42, respectively. Antibodies immunoprecipitate their respective synthetic A beta species but also A beta s and their related p3 counterparts endogenously secreted by transfected human kidney 293 cells. This allowed us to show that mutations on presenilin 1 triggered similar increased ratios of A beta 42 and its p 342 counterpart over total A beta and p3. ELISA assays allow detection of about 25-50 pg/ml of A beta s and remain linear up to 750 to 1500 pg/ml without any cross-reactivity. FCA18 and FCA3542 label diffuse and mature plaques of a sporadic AD case whereas FCA3340 only reveals the mature lesions and particularly labels their central dense core. In a CAA case, FCA18 and FCA3340 reveal leptomeningeal and cortical arterioles whereas FCA3542 only faintly labels such structures. CONCLUSIONS: Polyclonal antibodies exclusively recognizing A beta 40 (FCA 3340) or A beta 42 (FCA3542) were obtained. These demonstrated that FAD-linked presenilins similarly affect both p342 and A beta 42, suggesting that these mutations misroute the beta APP to a compartment where gamma-secretase, but not alpha-secretase, cleavages are modified. Overall, these antibodies should prove useful for fundamental and diagnostic approaches, as suggested by their usefulness for biochemical, cell biological, and immunohistochemical techniques.

Cathepsin D displays in vitro beta-secretase-like specificity.

The formation of A beta and A beta-containing fragments is likely a key event in the process of neural degeneration in Alzheimer's disease. The N-terminal residue (Asp-1) of A beta and its C-terminally extended sequences is liberated from the beta-amyloid precursor protein (beta APP) by beta-secretase(s). This activity appears highly increased by the presence (N-terminally to Asp-1) of a double-mutation (KM-->NL) found in several Swedish families affected by early onset Alzheimer's disease. By means of synthetic peptides encompassing the "normal' (N peptide) and mutated (delta NL peptide) sequences targeted by beta-secretase(s), we have detected a human brain protease displaying preferred efficiency for the delta NL peptide than for the non-mutated analog. This activity is sensitive to pepstatin, maximally active at acidic pH and hydrolyses the two peptides at the expected M/D or L/D cleavage sites. Such acidic activity is also detected in rat brain, PC12 cells and primary cultured astrocytes. The pepstatin sensitivity and pH maximum of the brain activity that appeared reminiscent of those displayed by the acidic protease cathepsin D led us to examine this enzyme as a putative beta-secretase-like candidate. Purified cathepsin D displays higher catalytic parameters for the delta NL peptide than for the non-mutated peptide, cleaves these two substrates at the expected M/D or L/D sites, and is maximally active at acidic pH. However, cathepsin D does not cleave peptides bearing mutations that were previously shown to drastically lower or fully block A beta secretion by transfected cells. Furthermore, cathepsin D hydrolyses recombinant baculoviral delta NL beta APP751 at a 6-fold higher rate than beta APP751 and gives rise to a 12-kDa C-terminal product that is recognized by antibodies fully specific of the N-terminus of A beta. Altogether, our study indicates that cathepsin D displays several in vitro beta-secretase-like properties that suggests that this protease could fulfill such a role, at least in the Swedish genetic form of Alzheimer's disease.