A Systematic Review of Stem Cell Differentiation into Keratinocytes for Regenerative Applications.

To improve wound healing or treatment of other skin diseases, and provide model cells for skin biology studies, in vitro differentiation of stem cells into keratinocyte-like cells (KLCs) is very desirable in regenerative medicine. This study examined the most recent advancements in in vitro differentiation of stem cells into KLCs, the effect of biofactors, procedures, and preparation for upcoming clinical cases. A range of stem cells with different origins could be differentiated into KLCs under appropriate conditions. The most effective ways of stem cell differentiation into keratinocytes were found to include the co-culture with primary epithelial cells and keratinocytes, and a cocktail of growth factors, cytokines, and small molecules. KLCs should also be supported by biomaterials for the extracellular matrix (ECM), which replicate the composition and functionality of the in vivo extracellular matrix (ECM) and, thus, support their phenotypic and functional characteristics. The detailed efficient characterization of different factors, and their combinations, could make it possible to find the significant inducers for stem cell differentiation into epidermal lineage. Moreover, it allows the development of chemically known media for directing multi-step differentiation procedures.In conclusion, the differentiation of stem cells to KLCs is feasible and KLCs were used in experimental, preclinical, and clinical trials. However, the translation of KLCs from in vitro investigational system to clinically valuable cells is challenging and extremely slow.

Development of a Composite Hydrogel Containing Statistically Optimized PDGF-Loaded Polymeric Nanospheres for Skin Regeneration: In Vitro Evaluation and Stem Cell Differentiation Studies.

Platelet-derived growth factor-BB (PDGF-BB) is a polypeptide growth factor generated by platelet granules faced to cytokines. It plays a role in forming and remodeling various tissue types, including epithelial tissue, through interaction with cell-surface receptors on most mesenchymal origin cells. However, it breaks down quickly in biological fluids, emphasizing the importance of preserving them from biodegradation. To address this challenge, we formulated and evaluated PDGF-encapsulated nanospheres (PD@PCEC) using polycaprolactone-polyethylene glycol-polycaprolactone. PD@PCECs were fabricated through the triple emulsion methodology and optimized by using the Box-Behnken design. The encapsulation efficiency (EE) of nanoencapsulated PDGF-BB was investigated concerning four variables: stirring rate (X1), stirring duration (X2), poly(vinyl alcohol) concentration (X3), and PDGF-BB concentration (X4). The selected optimized nanospheres were integrated into a gelatin-collagen scaffold (PD@PCEC@GC) and assessed for morphology, biocompatibility, in vitro release, and differentiation-inducing activity in human adipose-derived stem cells (hADSCs). The optimized PD@PCEC nanospheres exhibited a particle size of 177.9 ± 91 nm, a zeta potential of 5.2 mV, and an EE of 87.7 ± 0.44%. The release profile demonstrated approximately 85% of loaded PDGF-BB released during the first 360 h, with a sustained release over the entire 504 h period, maintaining bioactivity of 87.3%. The study also included an evaluation of the physicochemical properties of the scaffolds and an assessment of hADSC adhesion to the scaffold's surface. Additionally, hADSCs cultivated within the scaffold effectively differentiated into keratinocyte-like cells (KLCs) over 21 days, evidenced by morphological changes and upregulation of keratinocyte-specific genes, including cytokeratin 18, cytokeratin 19, and involucrin, at both transcriptional and protein levels.

Recent advancements in biosensor designs toward the detection of intestine cancer miRNA biomarkers.

Cancer diagnosis and treatment have been of broad interest among scientists in the last decades due to the high death rate, widespread occurrence, and recurrence after treatment. The survival rate of cancer patients depends greatly on early detection and appropriate treatments. Therefore developing new technologies applicable to sensitive and specific methods of cancer detection is an inevitable task for cancer researchers. Abnormal miRNA expression is contributed to severe diseases such as cancers and since their expression level and type differ strictly during carcinogenesis and later metastasis and treatments, the improved detection accuracy of these miRNAs would undoubtedly lead to early diagnosis, prognosis, and targeted therapy. Biosensors are accurate and straightforward analytical devices that have had practical applications especially in the last decade. Their domain is still growing through a combination of attractive nanomaterials and amplification methods, leading to innovative biosensing platforms for the efficient detection of miRNAs as diagnostic and prognostic biomarkers. In this review, we will provide the recent developments in biosensors to detect intestine cancer miRNA biomarkers and also discuss the challenges and outcomings of this field.

Mode of binding, kinetic and thermodynamic properties of a lipid-like drug (Fingolimod) interacting with Human Serum Albumin.

Introduction: Fingolimod is a drug that is used to treat multiple sclerosis (MS). It has pH-dependent solubility and low solubility when buffering agents are present. Multi-spectroscopic and molecular modeling methods were used to investigate the molecular mechanism of Fingolimod interaction with human serum albumin (HSA), and the resulting data were fitted to the appropriate models to investigate the molecular mechanism of interaction, binding constant, and thermodynamic properties. Methods: The interaction of Fingolimod with HSA was investigated in a NaCl aqueous solution (0.1 mM). The working solutions had a pH of 6.5. Data was collected using UV-vis, fluorescence quenching titrations, FTIR, and molecular modeling methods. Results: According to the results of the fluorescence quenching titrations, the quenching mechanism is static. The apparent binding constant value (KA = 4.26×103) showed that Fingolimod is a moderate HSA binder. The reduction of the KA at higher temperatures could be a result of protein unfolding. Hydrogen bonding and van der Waals interactions are the main contributors to Fingolimod-HSA complex formation. FTIR and CD characterizations suggested a slight decrease in the α-helix and ß-sheets of the secondary structure of HSA due to Fingolimod binding. Fingolimod binds to the binding site II, while a smaller tendency to the binding site I was observed as well. The results of the site marker competitive experiment and the thermodynamic studies agreed with the results of the molecular docking. Conclusion: The pharmacokinetic properties of fingolimod can be influenced by its HSA binding. In addition, considering its mild interaction, site II binding drugs are likely to compete. The methodology described here may be used to investigate the molecular mechanism of HSA interaction with lipid-like drugs with low aqueous solubility or pH-dependent solubility.

Long and Short-term Metformin Consumption as a Potential Therapy to Prevent Complications of COVID-19.

Purpose: The aim of the study is to evaluate the effect of metformin in complication improvement of hospitalized patients with COVID-19. Methods: This was a randomized clinical trial that involved 189 patients with confirmed COVID-19 infection. Patients in the intervention group received metformin-500 mg twice daily. Patients who received metformin before admission were excluded from the control group. Patients who were discharged before taking at least 2000 mg of metformin were excluded from the study. Primary outcomes were vital signs, need for ICU admission, need for intubation, and mortality. Results: Data showed that patients with diabetes with previous metformin in their regimen had lower percentages of ICU admission and death in comparison with patients without diabetes (11.3% vs. 26.1% (P=0.014) and 4.9% vs. 23.9% (P≤0.001), respectively). Admission time characteristics were the same for both groups except for diabetes and hyperlipidemia, which were significantly different between the two groups. Observations of naproxen consumption on endpoints, duration of hospitalization, and the levels of spO2 did not show any significant differences between the intervention and the control group. The adjusted OR for intubation in the intervention group versus the control group was 0.21 [95% CI, 0.04-0.99 (P=0.047)]. Conclusion: In this trial, metformin consumption had no effect on mortality and ICU admission rates in non-diabetic patients. However, metformin improved COVID-19 complications in diabetic patients who had been receiving metformin prior to COVID-19 infection, and it significantly lowered the intubation rates.

Investigations of the molecular mechanism of diltiazem binding to human serum albumin in the presence of metal ions, glucose and urea.

The molecular mechanism and thermodynamic properties of the interaction between diltiazem (DTZ) and human serum albumin (HSA), has been studied in vitro using spectroscopic techniques (UV-Vis, fluorescence, FTIR), and molecular docking methods. The effect of acidic and basic pH, glucose, urea, and metal ions on the DTZ-HSA binding has been investigated as well. According to the results, there is a 1:1 interaction between DTZ and HSA, while the quenching mechanism is static up to 313 K. The apparent binding constant was 2.09 × 106 M-1 that indicates a strong binding between DTZ and HSA. DTZ binding was increased in acidic pH while its binding was slowly decreased in the presence of glucose, urea, and metal ions. Thermodynamic studies showed that DTZ binds to HSA via an exothermic and spontaneous reaction via hydrogen bonding and electrostatic interactions. The conformational alteration of HSA is obvious according to the FTIR study. The site marker competitive study confirmed the binding of DTZ to the warfarin binding site. Molecular docking studies showed that DTZ binds to subdomain IB (-9.22 kcal mol-1) and subdomain IIIA (-9.03 kcal mol-1) with a higher tendency. Also, the results showed that the oxygen and nitrogen atoms of hydroxyl and amino functional groups of DTZ facilitate hydrogen bond formation. HighlightsStrong binding of diltiazem to HSA was studied and confirmed by fluorescence quenching titrations.Diltiazem binding to HSA reduces in the presence of metal ions, glucose, urea and alkaline pH.Diltiazem binding to HSA is exothermic and spontaneous.

Molecular mechanism and thermodynamic study of Rosuvastatin interaction with human serum albumin using a surface plasmon resonance method combined with a multi-spectroscopic, and molecular modeling approach.

Rosuvastatin (ROS) is an anti-cholesterol drug belonging to statin drugs. A multi-spectroscopic approach combined with a molecular modeling technique was used to assess ROS association with human serum albumin (HSA). Besides, an HSA immobilized surface plasmon resonance (SPR) chip was used to obtain kinetic parameters (ka, kd, and KD). Fluorescence quenching titrations revealed that ROS interacts with HSA via a dynamic, exothermic, enthalpy-driven mechanism. Hydrogen bonds and van der Waals interactions as the most prevalent bonding forces contribute to ROS-HSA complex formation. ROS binding to HSA alters HSA conformation. The SPR results indicated that ROS and HSA have a strong interaction possessing an equilibrium constant (KD) of 1.55 × 10-8 M at 298 K. A competitive analysis of site markers showed that ROS has a higher tendency to bind to the warfarin binding site (site IIA), which may explain why warfarin has a higher anticoagulant effect in ROS users. FRET analysis indicated that non-radiation energy transfer occurred between ROS and HSA. According to molecular docking studies, ROS prefers binding sites IB and IIA while the ROS-HSA complex stabilizes due to the hydrogen bond and π-π interaction. The presence of hydrogen-bond donors and acceptors, as well as aromatic ROS moieties, facilitates such interactions.

Development of an ethanol-free salbutamol sulfate metered-dose inhaler: Application of molecular dynamic simulation-based prediction of intermolecular interaction.

INTRODUCTION: More than fifty years after the commercialization of the Ventolin metered-dose inhaler (MDI), its constituent active ingredient, salbutamol sulfate (SS), remains the most prescribed short-acting beta agonist for the first-line treatment of acute asthma attacks and the metered-dose inhaler remains its primary dosage form. The first generation of Ventolin MDI was developed at a time when environmental and regulatory concerns were less stringent than today. The MDI industry is now on the verge of a second major reformulation effort in response to environmental concerns. This paper serves to illustrate how modern computational modeling of molecular interactions can aid the reformulation process. By way of a case study, computational modeling was performed to compare poly(ethylene glycol) 400 (PEG400) and, separately, isopropyl myristate (IPM) as substitutes for the ethanol used in some generic salbutamol sulfate suspension-based hydrofluoroalkane MDIs. METHODS: PEG400 and IPM were investigated as potential alternative cosolvents to ethanol in HFA134a-based SS suspension MDI formulations. Density functional theory (DFT) molecular dynamics simulations were used to evaluate the compatibility of the candidate cosolvents with the formulation's components. Corresponding physical formulations were filled into polyethylene terephthalate (PET) and, separately, aluminium canisters. In-vitro pharmaceutical product performance and macroscopic visual appearance were assessed and compared to the results of the simulation studies. RESULTS: The simulation studies indicated that PEG400 would be a good candidate as a replacement for ethanol whereas IPM would not. The in-vitro and visual assessments support the predicted outcome of the simulation studies. CONCLUSION: This work suggests that molecular dynamics simulations may provide a useful tool to aid the selection of compatible excipients when reformulating MDI suspension-based products, thereby reducing the time and cost associated with manufacturing and testing of physical samples.

Prediction of blood-brain distribution: effect of ionization.

Two simple multiple linear regression models were proposed to calculate the logarithm of the blood to brain concentration ratio (log BB) of drugs or drug-like compounds. The drugs were classified into two groups according to their ionization state in blood, and the significant parameters were selected using the train sets for each group. For un-ionizable compounds, the logarithm of distribution coefficient in octanol-water in pH 7.4 (log D(7.4)) and molecular weight are the significant parameters, whereas for ionizable compounds, log D(7.4) and number of hydrogen bond acceptor are significant parameters. The developed models were validated and their prediction capabilities checked using an external dataset of 25 compounds. In addition to the acceptable prediction errors, comparison of the external data analysis results with previously proposed models confirmed superior prediction capability of newly developed models.

A comprehensive review of in silico approaches for the prediction and modulation of aldehyde oxidase-mediated drug metabolism: The current features, challenges and future perspectives.

The importance of aldehyde oxidase (AOX) in drug metabolism necessitates the development and application of the in silico rational drug design methods as an integral part of drug discovery projects for the early prediction and modulation of AOX-mediated metabolism. The current study represents an up-to-date and thorough review of in silico studies of AOX-mediated metabolism and modulation methods. In addition, the challenges and the knowledge gap that should be covered have been discussed. The importance of aldehyde oxidase (AOX) in drug metabolism is a hot topic in drug discovery. Different strategies are available for the modulation of the AOX-mediated metabolism of drugs. Application of the rational drug design methods as an integral part of drug discovery projects is necessary for the early prediction of AOX-mediated metabolism. The current study represents a comprehensive review of AOX molecular structure, AOX-mediated reactions, AOX substrates, AOX inhibition, approaches to modify AOX-mediated metabolism, prediction of AOX metabolism/substrates/inhibitors, and the AOX related structure-activity relationship (SAR) studies. Furthermore, an up-to-date and thorough review of in silico studies of AOX metabolism has been carried out. In addition, the challenges and the knowledge gap that should be covered in the scientific literature have been discussed in the current review.

Synthesis and characterization of new surface modified magnetic nanoparticles and application for the extraction of letrozole from human plasma and analysis with HPLC-fluorescence.

Acetic acid-functionalized magnetic nanoparticles modified by (3-amino-propyl)-tri-ethoxy silane was synthesized and used as a new solid-phases adsorbent. Infrared spectroscopy (FT-IR), scanning electron microscopy (SEM), x-ray diffraction (XRD), energy-dispersive x-ray spectroscopy (EDX), vibrating sample magnetometer (VSM) and Electrophoretic Light Scattering (ELS) were used to characterize the modified nanoparticles. The molecular interaction between letrozole and nanoparticles (NPs) was studied using density functional theory (DFT) calculations. The developed nanoparticles were applied for dispersive solid-phase extraction of letrozole (an anticancer drug) from human plasma. Extracted letrozole was quantified using an isocratic HPLC/FL method. The extraction efficiency was optimized using one experiment at a time optimization method based on the adsorbent quantity, sample pH, adsorption time, desorption time, and elution solvent type/volume. The analysis method was fully validated according to the FDA guideline for bioanalytical method validation. The linear quantification range was 0.01-1 µg/mL and the lower limit of quantification (LLOQ) was 0.01 µg/mL. Plasma samples of 6 patients were analyzed and the measured letrozole concentrations range was 0.04-0.31 µg/mL. The newly synthesized magnetic nanoparticles were used successfully for the extraction of letrozole from spiked and clinical plasma samples. The developed method is a precise and simple method that is suitable for pharmacokinetic studies and clinical applications.

Design and fabrication of clinoptilolite-nanohydroxyapatite/chitosan-gelatin composite scaffold and evaluation of its effects on bone tissue engineering.

The aim of this study was to synthesize an innovative composite scaffold, which structured of clinoptilolite-nanohydroxyapatite/chitosan-gelatin (CLN-nHA/CS-G) with enhanced attributes for utilization in the bone tissue engineering. This composite scaffold was prepared by blending the CLN, nHA, chitosan, and gelatin solution followed by a freeze-drying step. The fabricated composite scaffolds were studied using BET, FTIR, XRD, and SEM techniques. The highly porous composite scaffolds with a pore size of 200 ± 100 µm were synthesized. Moreover, the effects of CLN and nHA on the physicochemical features of the scaffold such as density, swelling ratio, biomineralization, biodegradation, and mechanical behavior were studied. Compared with CS-G scaffold, the presence of CLN and nHA leads to an increased surface area, increased biomineralization, and low rate of degradation in simulated body fluid solution (SBF) and mechanical strength. Cytotoxicity of the CLN-nHA/CS-G scaffold was studied by MTT assay on human dental pulp stem cells (h-DPSCs). The biological response of h-DPSCs showed no toxicity and studied cells proliferated and attached on the pore surfaces of the scaffold. Results indicated that introducing CLN and nHA to composite improves the scaffold characteristics in a way that makes it suitable for bone tissue engineering.

Kinetic and thermodynamic study of beta-Boswellic acid interaction with Tau protein investigated by surface plasmon resonance and molecular modeling methods.

Introduction: Beta-Boswellic acid (BBA) is a pentacyclic terpene which has been obtained from frankincense and its beneficial effects on neurodegenerative disorders such as Alzheimer's disease (AD) have been addressed. Methods: In the present study, thermodynamic and kinetic aspects of BBA interaction with Tau protein as one of the important proteins involved in AD in the absence and presence of glucose has been investigated using surface plasmon resonance (SPR) method. Tau protein was immobilized onto the carboxy methyl dextran chip and its binding interactions with BBA were studied at physiological pH at various temperatures. Glucose interference with these interactions was also investigated. Results: Results showed that BBA forms a stable complex with Tau (KD=8.45×10-7 M) at 298 K. Molecular modeling analysis showed a hydrophobic interaction between BBA and HVPGGG segment of R2 and R4 repeated domains of Tau. Conclusion: The binding affinity increased by temperature enhancement, while it decreased significantly in the presence of glucose. Both association and dissociation of the BBA-Tau complex were accompanied with an entropic activation barrier; however, positive enthalpy and entropy changes revealed that hydrophobic bonding is the main force involved in the interaction.

The validity of self-reported drug use with urine test: results from the pilot phase of Azar cohort study.

Background: The present study aimed at assessing the validity of self-reported drug use in people aged 35 and older in a pilot phase of a population-based cohort study. Methods: A total of 1038 adults over 35 years old in Khamene city in East Azerbaijan province were recruited for the pilot phase of Azar cohort; a province-level of a nationwide PERSIAN cohort study completing a questionnaire and providing biological samples from October to December 2014. Information about the history and duration of smoking tobacco, using drug and medication were obtained by the physician. The validity of the drug use was assessed through comparing the questionnaire response with three urine strip tests for the detection of morphine, amphetamine and methamphetamine among 259 randomly selected subjects. Results: The prevalence of drug use according to self-report was 2.6% (95% CI: 1.7%-3.8%).One-step drug test as the gold standard for the use of drug self-reported demonstrated a sensitivity(95% CI) and specificity 15% (10-22) and 99.7% (98.9%-99.9%) respectively. All participants with positive self-report were male; however, in the urine analysis drug test, it was positive for 7out of 68 randomly selected women. Conclusion: The validity of self-reported drug use in this population was low; therefore, the self reported use of the drug should be used with caution in this population. It is recommended to use alternative techniques to improve the validity of data using the self-report procedure.

An overview of aldehyde oxidase: an enzyme of emerging importance in novel drug discovery.

INTRODUCTION: Given the rising trend in medicinal chemistry strategy to reduce cytochrome P450-dependent metabolism, aldehyde oxidase (AOX) has recently gained increased attention in drug discovery programs and the number of drug candidates that are metabolized by AOX is steadily growing. Areas covered: Despite the emerging importance of AOX in drug discovery, there are certain major recognized problems associated with AOX-mediated metabolism of drugs. Intra- and inter-species variations in AOX activity, the lack of reliable and predictive animal models using the common experimental animals, and failure in the predictions of in vivo metabolic activity of AOX using traditional in vitro methods are among these issues that are covered in this article. A comprehensive review of computational human AOX (hAOX) related studies are also provided. Expert opinion: Following the recent progress in the stem cell field, the authors recommend the application of organoids technology as an effective tool to solve the fundamental problems associated with the evaluation of AOX in drug discovery. The recent success in resolving the hAOX crystal structure can too be another valuable data source for the study of AOX-catalyzed metabolism of new drug candidates, using computer-aided drug discovery methods.

Interaction of an antiepileptic drug, lamotrigine with human serum albumin (HSA): Application of spectroscopic techniques and molecular modeling methods.

Lamotrigine (an epileptic drug) interaction with human serum albumin (HSA) was investigated by fluorescence, UV-Vis, FTIR, CD spectroscopic techniques, and molecular modeling methods. Binding constant (Kb) of 5.74×103 and number of binding site of 0.97 showed that there is a slight interaction between lamotrigine and HSA. Thermodynamic studies was constructed using the flourimetric titrations in three different temperatures and the resulted data used to calculate the parameters using Vant Hoff equation. Decreased Stern Volmer quenching constant by enhanced temperature revealed the static quenching mechanism. Negative standard enthalpy (ΔH) and standard entropy (ΔS) changes indicated that van der Waals interactions and hydrogen bonds were dominant forces which facilitate the binding of Lamotrigine to HSA, the results were confirmed by molecular docking studies which showed no hydrogen binding. The FRET studies showed that there is a possibility of energy transfer between Trp214 and lamotrigine. Also the binding of lamotrigine to HSA in the studied concentrations was not as much as many other drugs, but the secondary structure of the HSA was significantly changed following the interaction in a way that α-helix percentage was reduced from 67% to 57% after the addition of lamotrigine in the molar ratio of 4:1 to HSA. According to the docking studies, lamotrigine binds to IB site preferably.

A dispersive liquid-liquid microextraction and chiral separation of carvedilol in human plasma using capillary electrophoresis.

BACKGROUND: Development of simple, rapid and precise analysis of chiral drugs in biological samples is an important issue. Dispersive liquid-liquid microextraction in combination with CE using field amplified sample injection has been of interest because of its capability to analyze trace amount of drugs. METHODS: Dispersive liquid-liquid microextraction-CE-field amplified sample injection was employed for chiral separation of carvedilol in human plasma using UV-DAD detector and the developed method has been validated according to US FDA method validation guideline for bioanalysis. RESULTS: The method was linear over a concentration range of 12.5-100 ng/ml for each carvedilol enantiomer (R(2) = 0.998) and the mean recoveries ranged from 91 to 107%. CONCLUSION: The method was adapted for sensitive, selective and rapid determination of carvedilol enantiomers in human plasma samples.

Biological sample preparation: attempts on productivity increasing in bioanalysis.

Sample preparation is an important step of any biomedical analysis. Development and validation of fast, reproducible and reliable sample preparation methods would be very helpful in increasing productivity. Except for a few direct injection methods, almost all biological samples should at least be diluted before any analysis. Sometimes dilution is not possible because of the low concentration of the target analyte in the sample, and alternative pretreatments, such as filtration, precipitation and sample clean up using different extraction methods, are needed. This review focuses on the recent achievements in the pretreatment of biological samples and investigates them in six categories (i.e., dilution, filtration/dialysis, precipitation, extraction [solid-phase extraction, liquid-liquid extraction], novel techniques [turbulent flow chromatography, immunoaffinity method, electromembrane extraction] and combined methods). Each category will be discussed according to its productivity rate and suitability for routine analysis, and the discussed methods will be compared according to the mentioned indices.

Quantitative structure-activity relationships of imidazole-containing farnesyltransferase inhibitors using different chemometric methods.

Farnesyltranseferase inhibitors (FTIs) are one of the most promising classes of anticancer agents, but though some compounds in this category are in clinical trials there are no marketed drugs in this class yet. Quantitative structure activity relationship (QSAR) models can be used for predicting the activity of FTI candidates in early stages of drug discovery. In this study 192 imidazole-containing FTIs were obtained from the literature, structures of the molecules were optimized using Hyperchem software, and molecular descriptors were calculated using Dragon software. The most suitable descriptors were selected using genetic algorithms-partial least squares (GA-PLS) and stepwise regression, and indicated that the volume, shape and polarity of the FTIs are important for their activities. 2D-QSAR models were prepared using both linear methods, i.e., multiple linear regression (MLR), and non-linear methods, i.e., artificial neural networks (ANN) and support vector machines (SVM). The proposed QSAR models were validated using internal and external validation methods. The results show that the proposed 2D-QSAR models are valid and that they can be applied to predict the activities of imidazole-containing FTIs. The prediction capability of the 2D-QSAR (linear and non-linear) models is comparable to and somewhat better than that of previous 3D-QSAR models and the non-linear models are more accurate than the linear models.

QSBR study of bitter taste of peptides: application of GA-PLS in combination with MLR, SVM, and ANN approaches.

Detailed information about the relationships between structures and properties/activities of peptides as drugs and nutrients is useful in the development of drugs and functional foods containing peptides as active compounds. The bitterness of the peptides is an undesirable property which should be reduced during drug/nutrient production, and quantitative structure bitter taste relationship (QSBR) studies can help researchers to design less bitter peptides with higher target efficiency. Calculated structural parameters were used to develop three different QSBR models (i.e., multiple linear regression, support vector machine, and artificial neural network) to predict the bitterness of 229 peptides (containing 2-12 amino acids, obtained from the literature). The developed models were validated using internal and external validation methods, and the prediction errors were checked using mean percentage deviation and absolute average error values. All developed models predicted the activities successfully (with prediction errors less than experimental error values), whereas the prediction errors for nonlinear methods were less than those for linear methods. The selected structural descriptors successfully differentiated between bitter and nonbitter peptides.