Leonora Niv-Spector

Development and characterization of high affinity leptins and leptin antagonists.

Leptin is a pleiotropic hormone acting both centrally and peripherally. It participates in a variety of biological processes, including energy metabolism, reproduction, and modulation of the immune response. So far, structural elements affecting leptin binding to its receptor remain unknown. We employed random mutagenesis of leptin, followed by selection of high affinity mutants by yeast surface display and discovered that replacing residue Asp-23 with a non-negatively charged amino acid leads to dramatically enhanced affinity of leptin for its soluble receptor. Rational mutagenesis of Asp-23 revealed the D23L substitution to be most effective. Coupling the Asp-23 mutation with alanine mutagenesis of three amino acids (L39A/D40A/F41A) previously reported to convert leptin into antagonist resulted in potent antagonistic activity. These novel superactive mouse and human leptin antagonists (D23L/L39A/D40A/F41A), termed SMLA and SHLA, respectively, exhibited over 60-fold increased binding to leptin receptor and 14-fold higher antagonistic activity in vitro relative to the L39A/D40A/F41A mutants. To prolong and enhance in vivo activity, SMLA and SHLA were monopegylated mainly at the N terminus. Administration of the pegylated SMLA to mice resulted in a remarkably rapid, significant, and reversible 27-fold more potent increase in body weight (as compared with pegylated mouse leptin antagonist), because of increased food consumption. Thus, recognition and mutagenesis of Asp-23 enabled construction of novel compounds that induce potent and reversible central and peripheral leptin deficiency. In addition to enhancing our understanding of leptin interactions with its receptor, these antagonists enable in vivo study of the role of leptin in metabolic and immune processes and hold potential for future therapeutic use in disease pathologies involving leptin.

Pegylated leptin antagonist is a potent orexigenic agent: preparation and mechanism of activity.

Leptin, a pleiotropic adipokine, is a central regulator of appetite and weight and a key immunomodulatory protein. Although inborn leptin deficiency causes weight gain, it is unclear whether induced leptin deficiency in adult wild-type animals would be orexigenic. Previous work with a potent competitive leptin antagonist did not induce a true metabolic state of leptin deficiency in mice because of a short circulating half-life. In this study, we increased the half-life of the leptin antagonist by pegylation, which resulted in significantly increased bioavailability and retaining of antagonistic activity. Mice administered the pegylated antagonist showed a rapid and dramatic increase in food intake with weight gain. Resulting fat was confined to the mesenteric region with no accumulation in the liver. Serum cholesterol, triglyceride, and hepatic aminotransferases remained unaffected. Weight changes were reversible on cessation of leptin antagonist treatment. The mechanism of severe central leptin deficiency was found to be primarily caused by blockade of transport of circulating leptin across the blood-brain barrier with antagonisms at the arcuate nucleus playing a more minor role. Altogether we introduce a novel compound that induces central and peripheral leptin deficiency. This compound should be useful in exploring the involvement of leptin in metabolic and immune processes and could serve as a therapeutic for the treatment of cachexia.

Mapping leptin-interacting sites in recombinant leptin-binding domain (LBD) subcloned from chicken leptin receptor

The binding domain of the chicken leptin receptor [chLBD (chicken leptin-binding domain)], subcloned from the full-size chicken leptin receptor and prepared in an Escherichia coli system, was subjected to site-directed mutagenesis to identify the amino acids involved in leptin binding. A total of 22 electrophoretically pure, >90% monomer-containing mutants were expressed, refolded and purified. The effects of the mutations were tested by the ability to form complexes with ovine leptin, and the kinetic parameters of interaction were determined by surface plasmon resonance. Six mutants were used to determine whether mutations of several amino acids that differ between chLBD and mammalian LBDs will affect affinity: none showed any such effect, except the mutant A105D (Ala(105)-->Asp), which exhibited some decrease in affinity. Surface plasmon resonance analysis identified six mutants in which binding activity was totally abolished (F73A, Y14A/F73A, V76A/F77A, L78A/L79A, V76A/F77A/L78A/L79A and A105D/D106V) and six mutants (Y14A, R41A, R41A/S42A/K43A, V103A, V135A/F136A and F136A) in which affinity for the hormone was reduced, mainly by increased dissociation rates. Gel-filtration experiments indicated the formation of a 1:1 ovine or human leptin-chLBD complex with a molecular mass of approx. 41 kDa. Gel-filtration experiments yielded 1:1 complexes with those mutants in which affinity had decreased, but not with the six mutants, which had totally lost their binding capacity. Modelling the leptin-chLBD complex indicated that the binding domain of the latter is located mainly in the L3 loop, which contributes nine amino acid residues interacting with leptin. Contact-surface analysis identified the residues having the highest contribution to the recognition site to be Phe73, Phe77 and Leu79.

Preparation of Leptin Antagonists by Site-Directed Mutagenesis of Human, Ovine, Rat, and Mouse Leptin's Site III

Abstract:  Six muteins of human, ovine, rat, and mouse leptins mutated to Ala in amino acids 39–41 or 39–42 were prepared by site‐directed mutagenesis of the putative site III, which does not affect binding but is necessary for receptor activation, then expressed, solubilized in 4.5 M urea, at pH 11.3 in presence of cysteine, refolded and purified to homogeneity by anion‐exchange chromatography on Q‐Sepharose or combination of anion‐exchange chromatography followed by gel filtration. The overall yields were 400–800 mg from 5 L of fermentation. All proteins were >98% pure as evidenced by SDS‐PAGE and contained at least 95% monomers as documented by gel‐filtration chromatography under nondenaturing conditions. Circular dichroism analysis revealed that all six muteins have identical secondary structure characteristic of nonmutated leptins, namely 52–63% of alpha helix content. All muteins formed a 1:1 complex with chicken leptin binding domain, (chLBD) and bound chLBD or membrane‐embedded leptin receptor with affinity identical to WT leptins. Muteins were devoid of any biological activity in several bioassays but were potent competitive antagonists. Some muteins were pegylated using 40 kDa PEG. Although pegylation decreased the in vitro activity, increasing circulation half‐life can recompensate this deficit, so pegylated antagonists are expected to be more potent in vivo.

Leptin signaling and apoptotic effects in human prostate cancer cell lines.

Prostate cancer (PCa) progression is often associated with transactivation of the androgen receptor (AR) by endogenous hormones/growth factors. One such factor affecting growth, proliferation, and apoptostis (pro‐/anti‐) in various cancers is the adipokine leptin. This research studied leptin‐induced signaling and apoptosis in androgen sensitive (LNCaP, PC3/AR) and insensitive (PC3, DU145) PCa cell lines.

Principles of strategic drug delivery to the brain (SDDB): Development of anorectic and orexigenic analogs of leptin

The blood-brain barrier (BBB) presents a tremendous challenge for the delivery of drugs to the central nervous system (CNS). This includes drugs that target brain receptors for the treatment of obesity and anorexia. Strategic drug delivery to brain (SDDB) is an approach that considers in depth the relations among the BBB, the candidate therapeutic, the CNS target, and the disease state to be treated. Here, we illustrate principles of SDDB with two different approaches to developing drugs based on leptin. In normal body weight humans and in non-obese rodents, leptin is readily transported across the BBB and into the CNS where it inhibits feeding and enhances thermogenesis. However, in obesity, the transport of leptin across the BBB is impaired, resulting in a resistance to leptin. As a result, it is difficult to treat obesity with leptin or its analogs that depend on the leptin transporter for access to the CNS. To treat obesity, we developed a leptin agonist modified by the addition of pluronic block copolymers (P85-leptin). P85-leptin retains biological activity and is capable of crossing the BBB by a mechanism that is not dependent on the leptin transporter. As such, P85-leptin is able to cross the BBB of obese mice at a rate similar to that of native leptin in lean mice. To treat anorexia, we developed a leptin antagonist modified by pegylation (PEG-MLA) that acts primarily by blocking the BBB transporter for endogenous, circulating leptin. This prevents blood-borne, endogenous leptin from entering the CNS, essentially mimicking the leptin resistance seen in obesity, and resulting in a significant increase in adiposity. These examples illustrate two strategies in which an understanding of the interactions among the BBB, CNS targets, and candidate therapeutics under physiologic and diseased conditions can be used to develop drugs effective for the treatment of brain disease.

Large-scale preparation and characterization of non-pegylated and pegylated superactive ovine leptin antagonist

Superactive ovine leptin antagonist (SOLA) was prepared by rational mutagenesis of the ovine leptin antagonist L39A/D40A/F41A mutant prepared previously in our lab by mutating wild type leptin to D23L/L39A/D40A/F41A. SOLA was expressed in Escherichia coli as insoluble inclusion bodies, refolded and purified to homogeneity (as evidenced by SDS-PAGE and analytical gel filtration) by ion-exchange chromatography. The purified protein was mono-pegylated at its N terminus by 20-kDa linear pegylation reagent. The D23L mutation resulted in ca. 5- to 6-fold increased affinity toward soluble human leptin binding domain and 6- to 8-fold increased inhibitory activity in two different in vitro bioassays. This increase was similar, though not identical, to our previous results with superactive mouse and human leptin antagonists. Pegylation decreased overall activity by 5- to 8-fold, but as shown previously for superactive mouse leptin antagonist, the prolonged half life in the circulation will likely result in higher activity in vivo. As amino acids 6-31 (VQDDTKTLIKTIVTRINDISHTQSVS), making up a main part of the first α-helix, are identical in human, mouse, rat, ovine, bovine and pig leptins, we anticipate that D23L mutations of the respective leptins will result in similar increases in affinity and consequent activity of other leptin antagonists.

Is leptin an important physiological regulator of CRP

To the editor: Interaction of human leptin (hLEP) with human C-reactive protein (hCRP) was recently suggested as one of the major reasons for leptin resistance1. This notion is interesting because, if correct, preparation of leptin mutants that do not interact with CRP but retain their ability to interact and activate leptin receptors (LEPRs) would give us a potent biopharmaceutical reagent that could overcome such resistance. To validate this notion, we used several standard methods of determining protein: protein interactions. In our first attempt to study the interaction between one recombinant and two native hCRPs (from three different sources) and hLEP (ref. 2), mixtures of the two proteins were prepared at various molar concentrations in 25 mM Tris-HCl; pH 8.0, or 10 mM HEPES buffer (pH 7.4), containing 200 mM NaCl and 2 mM CaCl2, and resolved on a Superdex 75 column. There was a clear separation between CRP and hLEP (Supplementary Fig. 1 online), and there was no binding between the two proteins, as evidenced by the fact that there was no shift in the CRP peak (see Supplementary Fig. 1). Moreover, the size of the hLEP peak was not reduced relative to that obtained with the same concentration of hLEP injected alone. This experiment was repeated several times using all three CRPs and gave identical results. All CRPs appeared as a single peak that corresponded to a molecular mass of ∼70 kDa and were eluted at a retention time of ∼13.8 min. As the anticipated molecular mass of the pentamer is 115 kDa, we checked whether this shift was owing to dissociation or retardation of the protein by the resin, the latter suggested by the fact that the peak was skewed to the right (Supplementary Fig. 1). Accordingly, fractions corresponding to the CRP peak were collected and subjected to SDS-PAGE. To avoid the disassociation of CRP from pentamers into monomers, we dissolved the CRP in a sample buffer containing reducing agent but avoided boiling it (Supplementary Fig. 1). Using this procedure, we compared commercial CRP to that eluted from the Superdex column and found that the eluate consisted of only pentamers (Supplementary Fig. 1), whereas we identified traces of monomer in the commercial CRP (Supplementary Fig. 1). Because the premixing concentrations of CRP and leptin (625 μg/ml and 80 μg/ml, respectively) exceeded the physiological concentrations of CRP, we concluded that the existence of leptin: CRP complexes should be detectable under our binding of CRP to the agarose matrix that was used. These results parallel the misidentification 30 years ago of SAP as ‘C1t’, a putative fourth subcomponent of complement protein C1. SAP had been detected during affinity chromatography isolation of C1 from serum on immobilized IgG, but, in fact, it was present solely as a consequence of its avid calcium-dependent binding to the agarose matrix8 and had no interaction with C1 or any role in its function9. Inclusion of appropriate controls and awareness of the ligand-binding properties of the pentraxins are essential in such studies. We also demonstrate here that a substantial excess of CRP does not inhibit the effects of leptin on food intake and body weight in wild-type mice. Neither our observations nor the 10,000-fold dynamic range of human plasma CRP (0.05–500 mg/liter)10 and its rapidity in the acute-phase response support the idea of an in vivo role for CRP in the modulation of leptin function.

The obese phenotype-inducing N82K mutation in human leptin disrupts receptor-binding and biological activity.

A novel homozygous mutation of the leptin gene was recently reported in an Egyptian child and his sister with severe early onset obesity. This mutation results from the substitution of asparagine (AAC) by lysine (AAA) at codon 103 of a non-mature (signal peptide-containing) leptin and corresponds to the N82K mutation in the mature protein. The patient had very low serum leptin levels, raising the question of whether the obese phenotype resulted from low leptin levels or from its lower intrinsic activity. To answer this question, we characterized the functional consequences of the N82K mutation. Wild-type (WT) human leptin was mutated accordingly, expressed in Escherichia coli at high yield, purified to homogeneity as a monomer and compared to WT human leptin prepared by the same methodology. Circular dichroism analysis of the mutated leptin indicated proper refolding and a secondary structure identical to that of the WT human leptin. In contrast to WT human leptin, the N82K mutant did not form a detectable complex with human leptin-binding domain (hLBD) and its binding capacity to hLBD assessed in a nonradioactive receptor-binding assay was at least 500-fold lower than that of WT human leptin. The biological activity of the N82K mutant, tested in two cell bioassays, was reduced by more than three orders of magnitude relative to WT human leptin. Therefore, though the present report does not explain the reason for the low circulating leptin levels it definitely documents that the reported obese phenotype originates not only from low serum leptin levels but also from the N82K mutants almost total lack of intrinsic leptin activity.

Preparation and characterization of mouse IL-22 and its four single-amino-acid muteins that act as IL-22 receptor-1 antagonists

Recombinant mouse interleukin 22 (mIL-22) and its variants encoding four muteins (Y51A, N54A, R55A and E117A) were expressed in Escherichia coli, refolded and purified to homogeneity as monomeric proteins by one-step ion-exchange chromatography. The binding of IL-22 and its four muteins to immobilized mIL-22 receptor α1 extracellular domain (mIL-22 Rα1-ECD) exhibited similar affinity, indicating that the single-amino-acid mutations do not affect its binding properties. Similarly, no differences were found in binding to IL-22 binding protein expressed on the surface of yeast cells, although the affinity of all five proteins to the binding protein was higher than that to IL-22 Rα1-ECD. In an in vitro bioassay, recombinant mIL-22 stimulated signal transducer and activator of transcription-3 phosphorylation in HepG2 cells, whereas the four muteins were completely (Y51A) or almost completely (N54A, R55A and E117A) devoid of this agonistic activity. Furthermore, the agonistic activity of mIL-22 could be inhibited in a dose-dependent manner by the four muteins with almost identical efficiency. mIL-22 and its Y51A mutein were pegylated by methoxy polyethylene glycol-propionylaldehyde-20 kDa, yielding a mixture of mono (75-80%) and double (20-25%) pegylated proteins. The pegylated proteins showed lower affinity (50 and 25%) toward immobilized mIL-22 Rα1-ECD than their non-pegylated analogs. Wild-type pegylated IL-22 exhibited 5- to 10-fold lower activity in the HepG2 bioassay than its non-pegylated counterpart. Preparation of recombinant mIL-22 antagonists provides new tools for the study of IL-22 activity and of eventual therapeutic means for attenuating its negative effects.