Development of a Bispecific IgG1 Antibody Targeting BCMA and PDL1.

We designed, produced, and purified a novel IgG1-like, bispecific antibody (bsAb) directed against B-cell maturation antigen (BCMA), expressed by multiple myeloma (MM) cells, and an immune checkpoint inhibitor (ICI), PDL1, expressed in the MM microenvironment. The BCMA×PDL1 bsAb was fully characterized in vitro. BCMA×PDL1 bound specifically and simultaneously, with nM affinity, to both native membrane-bound antigens and to the recombinant soluble antigen fragments, as shown by immunophenotyping analyses and surface plasmon resonance (SPR), respectively. The binding affinity of bsAb for PDL1 and BCMA was similar to each other, but PDL1 affinity was about 10-fold lower in the bsAb compared to parent mAb, probably due to the steric hindrance associated with the more internal anti-PDL1 Fab. The bsAb was also able to functionally block both antigen targets with IC50 in the nM range. The bsAb Fc was functional, inducing human-complement-dependent cytotoxicity as well as ADCC by NK cells in 24 h killing assays. Finally, BCMA×PDL1 was effective in 7-day killing assays with peripheral blood mononuclear cells as effectors, inducing up to 75% of target MM cell line killing at a physiologically attainable, 6 nM, concentration. These data provide the necessary basis for future optimization and in vivo testing of this novel bsAb.

A Comprehensive Technology Platform for the Rapid Discovery of Peptide Inhibitors against SARS-CoV-2 Pseudovirus Infection.

We developed and validated a technology platform for designing and testing peptides inhibiting the infectivity of SARS-CoV-2 spike protein-based pseudoviruses. This platform integrates target evaluation, in silico inhibitor design, peptide synthesis, and efficacy screening. We generated a cyclic peptide library derived from the receptor-binding domain (RBD) of the SARS-CoV-2 spike protein and the angiotensin-converting enzyme 2 (ACE2) receptor. The cell-free validation process by ELISA competition assays and Surface Plasmon Resonance (SPR) studies revealed that the cyclic peptide c9_05, but not its linear form, binds well to ACE2. Moreover, it effectively inhibited the transduction in HEK293, stably expressing the human ACE2 receptor of pseudovirus particles displaying the SARS-CoV-2 spike in the Wuhan or UK variants. However, the inhibitory efficacy of c9_05 was negligible against the Omicron variant, and it failed to impede the entry of pseudoviruses carrying the B.1.351 (South African) spike. These variants contain three or more mutations known to increase affinity to ACE2. This suggests further refinement is needed for potential SARS-CoV-2 inhibition. Our study hints at a promising approach to develop inhibitors targeting viral infectivity receptors, including SARS-CoV-2's. This platform also promises swift identification and evaluation of inhibitors for other emergent viruses.

Aß1-6A2V(D) peptide, effective on Aß aggregation, inhibits tau misfolding and protects the brain after traumatic brain injury.

Alzheimer's disease (AD), the leading cause of dementia in older adults, is a double proteinopathy characterized by amyloid-ß (Aß) and tau pathology. Despite enormous efforts that have been spent in the last decades to find effective therapies, late pharmacological interventions along the course of the disease, inaccurate clinical methodologies in the enrollment of patients, and inadequate biomarkers for evaluating drug efficacy have not allowed the development of an effective therapeutic strategy. The approaches followed so far for developing drugs or antibodies focused solely on targeting Aß or tau protein. This paper explores the potential therapeutic capacity of an all-D-isomer synthetic peptide limited to the first six amino acids of the N-terminal sequence of the A2V-mutated Aß, Aß1-6A2V(D), that was developed following the observation of a clinical case that provided the background for its development. We first performed an in-depth biochemical characterization documenting the capacity of Aß1-6A2V(D) to interfere with the aggregation and stability of tau protein. To tackle Aß1-6A2V(D) in vivo effects against a neurological decline in genetically predisposed or acquired high AD risk mice, we tested its effects in triple transgenic animals harboring human PS1(M146 V), APP(SW), and MAPT(P301L) transgenes and aged wild-type mice exposed to experimental traumatic brain injury (TBI), a recognized risk factor for AD. We found that Aß1-6A2V(D) treatment in TBI mice improved neurological outcomes and reduced blood markers of axonal damage. Exploiting the C. elegans model as a biosensor of amyloidogenic proteins' toxicity, we observed a rescue of locomotor defects in nematodes exposed to the brain homogenates from TBI mice treated with Aß1-6A2V(D) compared to TBI controls. By this integrated approach, we demonstrate that Aß1-6A2V(D) not only impedes tau aggregation but also favors its degradation by tissue proteases, confirming that this peptide interferes with both Aß and tau aggregation propensity and proteotoxicity.

Amyposomes, a nanotechnological chaperone with anti-amyloidogenic activity.

AIM: The effect of liposomes bi-functionalized with phosphatidic acid and with a synthetic peptide derived from human apolipoprotein E has been evaluated on the aggregation features of different amyloidogenic proteins: human Amyloid ß1-40 (Aß1-40), transthyretin (TTR) variant S52P, human ß2microglobulin (ß2m) variants ΔN6 and D76N, Serum Amyloid A (SAA). METHODS: The formation of fibrillar aggregates of the proteins was investigated by ThioflavinT fluorescence assay and validated by Atomic Force Microscopy. RESULTS: The results show that liposomes are preventing the transition of non-aggregated forms to the fibrillar state, with stronger effects on Aß1-40, ß2m ΔN6 and SAA. Liposomes also induce disaggregation of the amyloid aggregates of all the proteins investigated, with stronger effects on Aß1-40, ß2 D76N and TTR.SPR assays show that liposomes bind Aß1-40 and SAA aggregates with high affinity (KD in the nanomolar range) whereas binding to TTR aggregates showed a lower affinity (KD in the micromolar range). Aggregates of ß2m variants showed both high and low affinity binding sites. Computed Structural analysis of protein fibrillar aggregates and considerations on the multidentate features of liposomes allow to speculate a common mechanism of action, based on binding the ß-stranded peptide regions responsible for the amyloid formation. CONCLUSION: Thus, multifunctional liposomes perform as pharmacological chaperones with anti-amyloidogenic activity, with a promising potential for the treatment of a number of protein-misfolding diseases.Key messageAmyloidosis is a group of diseases, each due to a specific protein misfolding.Anti-amyloidogenic nanoparticles have been gaining the utmost importance as a potential treatment for protein misfolding disorders.Liposomes bi-functionalized with phosphatidic acid and with a synthetic peptide derived from human apolipoprotein E showed anti-amyloidogenic activity.

Rational Design of a Peptidomimetic Inhibitor of Gelsolin Amyloid Aggregation.

Gelsolin amyloidosis (AGel) is characterized by multiple systemic and ophthalmic features resulting from pathological tissue deposition of the gelsolin (GSN) protein. To date, no cure is available for the treatment of any form of AGel. More than ten single-point substitutions in the GSN gene are responsible for the occurrence of the disease and, among them, D187N/Y is the most widespread variant. These substitutions undergo an aberrant proteolytic cascade, producing aggregation-prone peptides of 5 and 8 kDa, containing the Gelsolin Amyloidogenic Core, spanning residues 182-192 (GAC182-192). Following a structure-based approach, we designed and synthesized three novel sequence-specific peptidomimetics (LB-5, LB-6, and LB-7) built on a piperidine-pyrrolidine unnatural amino acid. LB-5 and LB-6, but not LB-7, efficiently inhibit the aggregation of the GAC182-192 amyloidogenic peptides at sub-stoichiometric concentrations. These peptidomimetics resulted also effective in vivo, in a C. elegans-based assay, in counteracting the proteotoxicity of aggregated GAC182-192. These data pave the way to a novel pharmacological strategy against AGel and also validate a toolbox exploitable in other amyloidogenic diseases.

Characterization of raloxifene as a potential pharmacological agent against SARS-CoV-2 and its variants.

The new coronavirus SARS-CoV-2 is the causative agent of the COVID-19 pandemic, which so far has caused over 6 million deaths in 2 years, despite new vaccines and antiviral medications. Drug repurposing, an approach for the potential application of existing pharmaceutical products to new therapeutic indications, could be an effective strategy to obtain quick answers to medical emergencies. Following a virtual screening campaign on the most relevant viral proteins, we identified the drug raloxifene, a known Selective Estrogen Receptor Modulator (SERM), as a new potential agent to treat mild-to-moderate COVID-19 patients. In this paper we report a comprehensive pharmacological characterization of raloxifene in relevant in vitro models of COVID-19, specifically in Vero E6 and Calu-3 cell lines infected with SARS-CoV-2. A large panel of the most common SARS-CoV-2 variants isolated in Europe, United Kingdom, Brazil, South Africa and India was tested to demonstrate the drug's ability in contrasting the viral cytopathic effect (CPE). Literature data support a beneficial effect by raloxifene against the viral infection due to its ability to interact with viral proteins and activate protective estrogen receptor-mediated mechanisms in the host cells. Mechanistic studies here reported confirm the significant affinity of raloxifene for the Spike protein, as predicted by in silico studies, and show that the drug treatment does not directly affect Spike/ACE2 interaction or viral internalization in infected cell lines. Interestingly, raloxifene can counteract Spike-mediated ADAM17 activation in human pulmonary cells, thus providing new insights on its mechanism of action. A clinical study in mild to moderate COVID-19 patients (NCT05172050) has been recently completed. Our contribution to evaluate raloxifene results on SARS-CoV-2 variants, and the interpretation of the mechanisms of action will be key elements to better understand the trial results, and to design new clinical studies aiming to evaluate the potential development of raloxifene in this indication.

A Surface Plasmon Resonance-Based Assay for Simultaneous Measurement of Serum Concentrations of Therapeutic Antibodies and Anti-Drug Antibodies.

The monitoring of the blood levels of therapeutic antibodies and their immune responses is proposed to guide and optimize therapy with these expensive drugs. We describe a novel Surface Plasmon Resonance (SPR)-based assay suitable for the simultaneous determination of serum concentrations of infliximab and anti-infliximab antibodies. The real-time detection by SPR avoids the incubation/washing steps of commonly used methods, thus allowing faster and more reliable measurements, in particular for low-affinity anti-drug antibodies. This method proved to be highly reproducible and may be well applied to other biotherapeutics.

A novel hotspot of gelsolin instability triggers an alternative mechanism of amyloid aggregation.

Gelsolin comprises six homologous domains, named G1 to G6. Single point substitutions in this protein are responsible for AGel amyloidosis, a hereditary disease causing progressive corneal lattice dystrophy, cutis laxa, and polyneuropathy. Although several different amyloidogenic variants of gelsolin have been identified, only the most common mutants present in the G2 domain have been thoroughly characterized, leading to clarification of the functional mechanism. The molecular events underlying the pathological aggregation of 3 recently identified mutations, namely A551P, E553K and M517R, all localized at the interface between G4 and G5, are here explored for the first time. Structural studies point to destabilization of the interface between G4 and G5 due to three structural determinants: ß-strand breaking, steric hindrance and/or charge repulsion, all implying impairment of interdomain contacts. Such rearrangements decrease the temperature and pressure stability of gelsolin but do not alter its susceptibility to furin cleavage, the first event in the canonical aggregation pathway. These variants also have a greater tendency to aggregate in the unproteolysed forms and exhibit higher proteotoxicity in a C. elegans-based assay. Our data suggest that aggregation of G4G5 variants follows an alternative, likely proteolysis-independent, pathway.

Doxycycline Inhibition of a Pseudotyped Virus Transduction Does Not Translate to Inhibition of SARS-CoV-2 Infectivity.

The rapid spread of the pandemic caused by the SARS-CoV-2 virus has created an unusual situation, with rapid searches for compounds to interfere with the biological processes exploited by the virus. Doxycycline, with its pleiotropic effects, including anti-viral activity, has been proposed as a therapeutic candidate for COVID-19 and about twenty clinical trials have started since the beginning of the pandemic. To gain information on the activity of doxycycline against SARS-CoV-2 infection and clarify some of the conflicting clinical data published, we designed in vitro binding tests and infection studies with a pseudotyped virus expressing the spike protein, as well as a clinically isolated SARS-CoV-2 strain. Doxycycline inhibited the transduction of the pseudotyped virus in Vero E6 and HEK-293 T cells stably expressing human receptor angiotensin-converting enzyme 2 but did not affect the entry and replication of SARS-CoV-2. Although this conclusion is apparently disappointing, it is paradigmatic of an experimental approach aimed at developing an integrated multidisciplinary platform which can shed light on the mechanisms of action of potential anti-COVID-19 compounds. To avoid wasting precious time and resources, we believe very stringent experimental criteria are needed in the preclinical phase, including infectivity studies with clinically isolated SARS-CoV-2, before moving on to (futile) clinical trials.

Can Antiviral Activity of Licorice Help Fight COVID-19 Infection?

The phytotherapeutic properties of Glycyrrhiza glabra (licorice) extract are mainly attributed to glycyrrhizin (GR) and glycyrrhetinic acid (GA). Among their possible pharmacological actions, the ability to act against viruses belonging to different families, including SARS coronavirus, is particularly important. With the COVID-19 emergency and the urgent need for compounds to counteract the pandemic, the antiviral properties of GR and GA, as pure substances or as components of licorice extract, attracted attention in the last year and supported the launch of two clinical trials. In silico docking studies reported that GR and GA may directly interact with the key players in viral internalization and replication such as angiotensin-converting enzyme 2 (ACE2), spike protein, the host transmembrane serine protease 2, and 3-chymotrypsin-like cysteine protease. In vitro data indicated that GR can interfere with virus entry by directly interacting with ACE2 and spike, with a nonspecific effect on cell and viral membranes. Additional anti-inflammatory and antioxidant effects of GR cannot be excluded. These multiple activities of GR and licorice extract are critically re-assessed in this review, and their possible role against the spread of the SARS-CoV-2 and the features of COVID-19 disease is discussed.

Surface plasmon resonance unveils important pitfalls of enzyme-linked immunoassay for the detection of anti-infliximab antibodies in patients' sera.

Measurements of serum concentrations of therapeutic antibodies and anti-drug antibodies (ADA) can support clinical decisions for the management of non-responders, optimizing the therapy. In the present study we compared the results obtained by classical ELISA and a recently proposed surface plasmon resonance (SPR)-based immunoassay, in 76 patients receiving infliximab for inflammatory bowel diseases. The two methods indicated very similar serum concentrations of the drug, but there were striking differences as regards ADA. All the sera showing ADA by ELISA (14) also showed ADA by SPR, but the absolute amounts were different, being 7-490 times higher with SPR, with no correlation. Eight patients showed ADA only with SPR, and these ADA had significantly faster dissociation rate constants than those detectable by both SPR and ELISA. The underestimation, or the lack of detection, of ADA by ELISA is likely to reflect the long incubation steps which favor dissociation of the patient's low-affinity ADA, while the commercial, high-affinity anti-infliximab antibodies used for the calibration curve do not dissociate. This problem is less important with SPR, which monitors binding in real time. The possibility offered by SPR to detect ADA in patients otherwise considered ADA-negative by ELISA could have important implications for clinicians.

Nonphosphorylated tau slows down Aß1-42 aggregation, binds to Aß1-42 oligomers, and reduces Aß1-42 toxicity.

The formation of neurofibrillary tangles and amyloid plaques accompanies the progression of Alzheimer's disease. Tangles are made of fibrillar aggregates formed by the microtubule-associated protein tau, whereas plaques comprise fibrillar forms of amyloid-beta (Aß). Both form toxic oligomers during aggregation and are thought to interact synergistically to each promote the accumulation of the other. Recent in vitro studies have suggested that the monomeric nonphosphorylated full-length tau protein hinders the aggregation of Aß1-40 peptide, but whether the same is true for the more aggregation-prone Aß1-42 was not determined. We used in vitro and in vivo techniques to explore this question. We have monitored the aggregation kinetics of Aß1-42 by thioflavine T fluorescence in the presence or the absence of different concentrations of nonphosphorylated tau. We observed that elongation of Aß1-42 fibrils was inhibited by tau in a dose-dependent manner. Interestingly, the fibrils were structurally different in the presence of tau but did not incorporate tau. Surface plasmon resonance indicated that tau monomers bound to Aß1-42 oligomers (but not monomers) and hindered their interaction with the anti-Aß antibody 4G8, suggesting that tau binds to the hydrophobic central core of Aß recognized by 4G8. Tau monomers also antagonized the toxic effects of Aß oligomers in Caenorhabditis elegans. This suggests that nonphosphorylated tau might have a neuroprotective effect by binding Aß1-42 oligomers formed during the aggregation and shielding their hydrophobic patches.

Characterization of the neutralizing anti-emicizumab antibody in a patient with hemophilia A and inhibitor.

BACKGROUND: The genetically engineered, humanized, bispecific monoclonal antibody emicizumab (Hemlibra) that mimics the cofactor activity of activated factor VIII (FVIII) has been approved for treatment of hemophilia A patients with and without inhibitor. In the pivotal premarketing clinical trials, emicizumab prophylaxis significantly reduced bleeding rates compared with previous treatments and was well tolerated. However, a consequence of this novel therapy may be the host immune response to a foreign protein. OBJECTIVE: Characterization of the neutralizing anti-emicizumab antibody associated with the loss of treatment efficacy. PATIENT: A pediatric hemophilia A patient with inhibitor enrolled in the HAVEN2 (Study of Emicizumab Administered Subcutaneously (SC) in Pediatric Participants With Hemophilia A and Factor VIII (FVIII) Inhibitors) clinical trial. METHODS: The anti-emicizumab antibody has been characterized with Western blot and enzyme-linked immunosorbent assay (ELISA). The antibody was affinity purified and sequenced. Binding affinity to full-length and papain-digested emicizumab was analyzed using surface plasmon resonance and byo-layer interferometry. RESULTS: The neutralizing anti-emicizumab antibody was highly polyclonal with high-affinity binding mainly to the Fab portion of emicizumab with a small amount of binding to the Fc portion. Molecular interaction experiments between emicizumab and the purified antibody indicated the presence of at least two components with similar affinities. CONCLUSIONS: Although the incidence of neutralizing anti-emicizumab antibody is rare, this study highlights the importance of a close monitoring and the need of a simple laboratory assay to promptly detect these antibodies in patients with a history of poor drug efficacy.

A portable optical-fibre-based surface plasmon resonance biosensor for the detection of therapeutic antibodies in human serum.

Different lines of evidence indicate that monitoring the blood levels of therapeutic antibodies, characterized by high inter-individual variability, can help to optimize clinical decision making, improving patient outcomes and reducing costs with these expensive treatments. A surface plasmon resonance (SPR)-based immunoassay has recently been shown to allow highly reliable and robust monitoring of serum concentrations of infliximab, with significant advantages over classical ELISA. The next level of advancement would be the availability of compact and transportable SPR devices suitable for easy, fast and cheap point-of-care analysis. Here we report the data obtained with recently developed, cost-effective, optical-fibre-based SPR sensors (SPR-POF), which allow the construction of a compact miniaturized system for remote sensing. We carried out an extensive characterization of infliximab binding to an anti-infliximab antibody immobilized on the SPR-POF sensor surface. The present proof-of-principle studies demonstrate the feasibility of the proposed SPR-POF platform for the specific detection of infliximab, in both buffer and human serum, and pave the way for further technological improvements.

The Anti-Amyloidogenic Action of Doxycycline: A Molecular Dynamics Study on the Interaction with Aß42.

The pathological aggregation of amyloidogenic proteins is a hallmark of many neurological diseases, including Alzheimer's disease and prion diseases. We have shown both in vitro and in vivo that doxycycline can inhibit the aggregation of Aß42 amyloid fibrils and disassemble mature amyloid fibrils. However, the molecular mechanisms of the drug's anti-amyloidogenic property are not understood. In this study, a series of molecular dynamics simulations were performed to explain the molecular mechanism of the destabilization of Aß42 fibrils by doxycycline and to compare the action of doxycycline with those of iododoxorubicin (a toxic structural homolog of tetracyclines), curcumin (known to have anti-amyloidogenic activity) and gentamicin (an antibiotic with no experimental evidence of anti-amyloidogenic properties). We found that doxycycline tightly binds the exposed hydrophobic amino acids of the Aß42 amyloid fibrils, partly leading to destabilization of the fibrillar structure. Clarifying the molecular determinants of doxycycline binding to Aß42 may help devise further strategies for structure-based drug design for Alzheimer's disease.

A Surface Plasmon Resonance-based assay to measure serum concentrations of therapeutic antibodies and anti-drug antibodies.

Therapeutic drug and immunogenicity monitoring (TDIM) is increasingly proposed to guide therapy with biologics, characterised by high inter-individual variability of their blood levels, to permit objective decisions for the management of non-responders and reduce unnecessary interventions with these expensive treatments. However, TDIM has not yet entered clinical practice partly because of uncertainties regarding the accuracy and precision of enzyme-linked immunosorbent assays (ELISA). Here we report the characterisation of a novel surface plasmon resonance (SPR)-based TDIM, applied to the measurement of serum concentrations of infliximab, an antibody against tumour necrosis factor α (anti-TNFα), and anti-infliximab antibodies. SPR has the obvious advantages of directly detecting and measuring serum antibodies in minutes, avoiding the long incubation/separation/washing/detection steps of the methods proposed so far, reducing complexity and variability. Moreover, drug and anti-drug antibodies can be measured simultaneously. This new method was validated for sensitivity and reproducibility, and showed cost-effectiveness over commercial ELISA kits. This method may be applied to other biotherapeutics. These data pave the way for the development of SPR-based point-of-care devices for rapid on-site analysis.

Cellular prion protein neither binds to alpha-synuclein oligomers nor mediates their detrimental effects.

α-Synuclein oligomers are crucial players in the pathogenesis of Parkinson's disease. Some mechanisms involved in α-synuclein oligomer detrimental effects include membrane damage, neuroinflammation and protein-protein interactions. Recently, the cellular prion protein (PrPC) emerged as an interactor of α-synuclein oligomers, apparently mediating their detrimental activities. Through direct in vivo and in vitro approaches we herein investigated the existence of a direct cross-talk between α-synuclein oligomers and PrPC. In vitro, we assessed α-synuclein oligomer toxicity by comparing the effect in Prnp+/+ versus PrPC knockout (Prnp0/0) hippocampal neurons. Through an in vivo acute mouse model, where α-synuclein oligomers injected intracerebroventricularly induce memory impairment and neuroinflammation, we verified whether these detrimental effects were preserved in Prnp0/0 mice. In addition, PrPC-α-synuclein oligomer direct binding was investigated through surface plasmon resonance. We found that PrPC was not mandatory to mediate α-synuclein oligomer detrimental effects in vitro or in vivo. Indeed, α-synuclein oligomer toxicity was comparable in Prnp+/+ and Prnp0/0 neurons and both Prnp+/+ and Prnp0/0 mice injected with α-synuclein oligomers displayed memory deficit and hippocampal gliosis. Moreover, surface plasmon resonance analyses ruled out PrPC-α-synuclein oligomer binding. Our findings indicate that PrPC neither binds α-synuclein oligomers nor mediates their detrimental actions. Therefore, it is likely that PrPC-dependent and PrPC-independent pathways co-exist in Parkinson's disease.

Doxycycline counteracts neuroinflammation restoring memory in Alzheimer's disease mouse models.

ß-Amyloid oligomers (AßOs) and neuroinflammation are 2 main culprits to counteract in Alzheimer's disease (AD). Doxycycline (DOXY) is a second generation antibiotic of the tetracycline class that are promising drugs tested in many clinical trials for a number of different pathologies. DOXY is endowed with antiamyloidogenic properties and better crosses the blood-brain barrier, but its efficacy has never been tested in AD mice. We herein show that 15- to 16-month-old APP/PS1dE9 (APP/PS1) AD mice receiving DOXY under different treatment regimens recovered their memory without plaque reduction. An acute DOXY treatment was, also, sufficient to improve APP/PS1 mouse memory, suggesting an action against soluble AßOs. This was confirmed in an AßO-induced mouse model, where the AßO-mediated memory impairment was abolished by a DOXY pretreatment. Although AßOs induce memory impairment through glial activation, assessing the anti-inflammatory action of DOXY, we found that in both the AßO-treated and APP/PS1 mice, the memory recovery was associated with a lower neuroinflammation. Our data promote DOXY as a hopeful repositioned drug counteracting crucial neuropathological AD targets.

QSAR model for blood-brain barrier permeation.

BACKGROUND AND OBJECTIVE: Predicting blood-brain barrier permeability for novel compounds is an important goal for neurotherapeutics-focused drug discovery. It is impossible to determine experimentally the blood-brain barrier partitioning of all possible candidates. Consequently, alternative evaluation methods based on computational models are desirable or even necessary. The CORAL software (http://www.insilico.eu/coral) has been checked up as a tool to build up quantitative structure - activity relationships for blood-brain barrier permeation. METHODS: The Monte Carlo technique gives possibility to build up predictive model of an endpoint by means of selection of so-called correlation weights of various molecular features. Descriptors calculated with these weights are basis for correlations "structure-endpoint". RESULTS: The approach gives good models for three random splits into the training and validation sets. The best model characterized by the following statistics for the external validation set: the number of compounds is 41, determination coefficient is equal to 0.896, root mean squared error is equal to 0.175. CONCLUSIONS: The suggested approach can be applied as a tool for prediction of blood-brain barrier permeation.