Read More. Molecules move within the cell or from one cell to another through different strategies. Transport may be in the form of simple diffusion, facilitated diffusion, active transport, osmosis, endocytosis, exocytosis, epithelial transport, or glandular secretion. This tutorial provides elaborate details on each of these mechanisms. Find out how.
Homeostasis is the relatively stable conditions of the internal environment that result from compensatory regulatory responses performed by homeostatic control systems. Know the different components of homeostatic control systems, homeostatic regulators, and the various biological processes that homeostasis entail Proteins have a crucial role in various biological activities. The charges were initially estimated by fitting the electrostatic potential using a Kollman—Singh scheme. All compounds were modeled in their neutral forms and other physiologically relevant charged states.
The parameters for the calibration data set are freely available from the Automated Topology Builder server. The potential of mean force PMF free-energy profiles for the partitioning of compounds was calculated using umbrella sampling simulations.
A single harmonic restraint with a force constant of kJ mol —1 nm —2 was applied to the z -axis normal to the bilayer distance between the center of mass of the compound and the center of mass of the DOPC bilayer. One hundred separate simulations were performed, with the compound harmonically restrained in increments of 0. These simulations completely span the 10 nm distance from bulk water, through the entire membrane and out into bulk water again. The weighted histogram analysis method, 49 as implemented within GROMACS, was used to calculate the PMF for each compound, and all free-energy profiles were normalized to zero in bulk water.
The membrane permeability of a small molecule was calculated using the same methodology as previously published. The position-specific diffusion coefficients were calculated from the MD simulation data using the method of Hummer.
These calculations were repeated for a number of subsamples of each trajectory. The resulting autocovariance curves decay roughly exponentially with increasing lag time. The standard deviation of the diffusion coefficient over the subsamples was also calculated for use in a sensitivity analysis, described below. The integral is over the width of the membrane. As a metric for confidence in our results, the error of P eff was determined.
In order to calculate this overall error of P eff , we first had to separately resample both the PMF and diffusion profiles. The sensitivity analysis used a random resampling of the position-specific diffusion coefficients. As described previously, for each compound, position-specific diffusion coefficients were calculated as well as the standard deviation of each position-specific coefficient.
These calculations were based on independent subsamples of the trajectory from each umbrella-sampling simulation. Since the mean and variation of each calculated coefficient is an independent calculation, the correlation of the variations is zero.
The resampled diffusion coefficients were thus based on independent draws from a set of normal distributions, where each such distribution has a mean and standard deviation corresponding to the calculated mean and standard deviation of each position-specific diffusion coefficient.
We used 50 random draws for each coefficient to produce a set of 50 diffusion profiles that were distributed around the calculated data according to the normal distribution of that data. Again, 50 separate PMF profiles were generated, from which the mean and standard deviations calculated. These standard deviations of the PMF were used as an additional quality control metric.
The PMF histogram overlap was also checked to ensure correct sampling. In order to determine the error of the P eff , the data in the 50 diffusion profiles were combined with the data in the 50 PMF profiles using eq 3 to produce 50 similar, but different, P eff s. The standard deviation for this set of 50 P eff s was taken as the error. PAMPA was applied to screen compounds for passive diffusion. Two fluid-filled compartments donor well and receiver well are separated by a polyvinylidene fluoride PVDF filter plate precoated with a phospholipid-oil-phospholipid trilayer primarily consisting of DOPC phospholipids.
Solutions were added to donor wells at a volume of 0. The filter plate, containing acceptor wells filled with 0. Following incubation, solution was removed from donor and acceptor wells. Compounds were then reserved for analysis via ultra performance liquid chromatography UPLC.
Certain compounds exhibited very low traversal to the acceptor well, such that material could not be detected via UPLC upon initial analysis. Such compounds were concentrated from donor and acceptor solution fractions by extraction into methanol, followed by vacuum concentration and resuspension in the minimal amount of appropriate UPLC solvent.
Detection was based on retention time and UV absorbance. Specific detection protocols were determined, where necessary for each compound Supporting Information. C eq was calculated according to the following equation: 5 where C D t is the concentration in donor well at time t, and the other values are the same as previously defined.
This method is a fragment-based approach, whereby experimentally determined fragments are modeled using QSPR rather than per atom. The CLogP program has become the benchmark, or gold-standard, method for LogP prediction in the last 35 years and is the most extensively tested fragment-based LogP calculator used in drug design.
The authors state that miLogP is calculated using the methodology developed by Molinspiration as a sum of fragment-based contributions and correction factors. This pool of fragments is based on the data in Viswanadhan et al.
This method is based on an atomic contribution model. Our computational methodology is first calibrated against the well-established PAMPA protocol using a set of diverse, highly studied compounds.
The predictive power of the approach is then tested with a synthesized set of structurally similar compounds and compared against the equivalent predictive power of readily available LogP calculators.
We are focused on improving the predictive capability of our membrane permeability model based on our previous work. This set of compounds is comprised of molecules that fit a specific set of criteria: 1 a significant amount of previous experimental work is available to compare against our results, and 2 the compounds must also cover a wide range of permeability values to evaluate the performance of our model across this range.
The potential of mean force PMF free energy curves for these compounds is generated from the umbrella sampling MD simulations Figure 2 , and in keeping with previous results, 36 it shows that as a general rule if a compound has a higher free energy in the middle of the bilayer, it will be less permeable.
Table 1. Using these measured experimental permeability rates, as well as permeability data from literature, we are able to divide our data into four groups; high permeability green , medium permeability yellow , low permeability orange , and impermeable red. The boundaries between these regions are defined by using the P eff PAMPA midpoints between different compounds of known permeability categories identified from literature.
By fitting against the linear regression line, we can use these in vitro cutoff values to define the equivalent cutoff values for our predicted log P eff PMF Figure 3 and Table 1. Using these cutoffs, we attempt to semiquantitatively predict the permeability of 18 of our synthesized set of structurally related compounds. This prediction is significantly more difficult than correlating the control compounds, as the range of permeabilities of these compounds is much smaller.
However, we have successfully identified and categorized all eight of the compounds with some level of permeability low, medium, or high. From a drug design point of view, a false positive is a better outcome than obtaining false negatives that actually do possess permeability, but are predicted to be impermeable.
Table 2. The semiquantitative permeability prediction results generated from our PMF-based methodology are compared with the equivalent calibration and predictions using several commonly used LogP calculation tools Table 3.
The details of these techniques are described in the Methods section. The various LogP calculations are first calibrated against the P eff PAMPA measurements for the control compounds to define LogP cutoffs between the four different permeability categories. Table 3. The LogP prediction tools performed more poorly than our PMF-based methodology in both the correlation against the control compounds and the semiquantitative prediction.
While the correlation coefficient for our PMF-based methodology is 0. The LogP tools have difficulty with the unusual structures of our negative controls that are known to be impermeable. Furthermore, all of the LogP tools have a significant number of false negative permeability predictions, with no method able to correctly identify more than two of the eight compounds that exhibit any form of experimentally categorized permeability.
In fact, two of the methods SLogP and miLogP fail to identify any of these compounds that have categorized permeability. Indeed, regardless of calibration, each of the LogP calculation tools assign a higher LogP greater permeability to some compounds measured to be impermeable than those measured to have a medium or even high permeability.
In comparison, our PMF-based methodology has no such occurrences and correctly identifies all of the compounds with experimentally measured categorized permeability. Indeed, even the four impermeable compounds that our method incorrectly categorize as low permeability compounds still all have lower predicted permeability that the four compounds correctly identified to have low permeability. This is impossible to achieve with any of the LogP data sets.
Assessing permeability is of critical importance to understanding and predicting in vivo efficacy and bioavailability of candidate drug compounds. A range of physiological membranes are relevant to such concerns.
Permeability across gastrointestinal membranes is critical to bioavailability 62 following oral administration. Transdermal permeation must be evaluated for topical administration and management of skin pathologies. Computational prediction of permeability represents a need that has been historically challenging and is often limited by the requirement for a robust training set, which ultimately restricts applicability of the resultant model. This study sought to address this need by refining and prospectively validating an MD-based model, demonstrating effective prediction of permeability for candidate compounds across a physiological lipid membrane.
The results of our study show that the developed computational model can predict the PAMPA-defined permeability category of a compound with greater accuracy than compared LogP-based methods. This is demonstrated both when examining well-characterized calibration compounds exhibiting a wide range of permeation capacity, as well when prospectively studying a set of novel compounds with a more narrow dynamic range. Calculations based on this prediction set are thus a challenging task, especially as the properties of many of the compounds are so similar.
Furthermore, several of the experimental permeabilities of the prediction set are measured at the lower end of the sensitivity threshold of the PAMPA methodology, resulting in higher relative uncertainty in this region due to the inherent challenge of experimentally measuring very low levels of diffusion precisely Figure 4.
Amplified relative uncertainty at the low end of the permeability spectrum is correspondingly observed in computational predictions; we have previously shown that equivalent errors in the PMF profile have a larger effect on the calculated P eff when the compound has a lower permeability versus a higher permeability if there is an energy barrier at the bilayer center rather than an energy well.
These results are a significant improvement in accuracy over predictions made using current high throughput LogP calculation techniques. Indeed, some of the LogP-based methods fail completely when attempting to categorize the permeability of the compounds. Depending on which LogP tool is used, the same compound could be characterized into three different regions. Furthermore, our PMF-based predicted permeabilities are not dependent on training sets, but only the physical properties recapitulated in the MD simulations.
Our first-principles method is especially useful for a novel set of compounds for which there is no QSPR or experimental data available, and we do not run the risk of overfitting our model to a narrow spectrum of physicochemical parameters.
Limitations in permeability and, ultimately, bioavailability, of candidate drug compounds can substantially slow and derail the development process. Ideally, computational model systems would serve as a first pass screen for permeation.
Existing methods such as QSPR and LogP are not computationally intensive, but are restrictive in their input capacity, and were shown in our study to produce a large proportion of false negatives, thereby running the risk of ruling out potentially promising leads. Although MD-based simulations are more computationally intensive than LogP calculations a few days compared to a few minutes , the result is a more accurate, actionable model that does not discard valuable compounds.
Further, the computational time required for the method described in our study is still less than the time required to synthesize, purify, and experimentally characterize the same novel compound.
From a drug design perspective, this predictive capability would facilitate compound evaluation by ruling out impermeable candidates with a demonstrably low false-negative rate. The resultant data would provide a more in depth characterization of the permeability of hits identified though high throughput virtual screening or analogues of existing lead compounds. Application of the described predictive methods would facilitate faster iterative improvement, allowing for selective retention of compounds with access to the intended physiological target.
Such a model could represent an important component of a more comprehensive design pipeline, capable of evaluating leads independent of structure or novelty of the given compound. Supporting Information. Author Information. Timothy S. Brian J. Nicholas A. Emma M. Carlson - U. Carlos A. Michael A. Heather A. Tuan H. Felice C. The authors declare no competing financial interest. Drug interactions with lipid membranes Chem. Seddon, Annela M. Royal Society of Chemistry. A review. The field of drug-membrane interactions is one that spans a wide range of scientific disciplines, from synthetic chem.
Cell membranes are complex dynamic systems whose structures can be affected by drug mols. In this tutorial review we aim to provide a guide for those new to the area of drug-membrane interactions and present an introduction to areas of this topic which need to be considered.
We address the lipid compn. We outline methods by which drugs may cross or bind to this membrane, including the well understood passive and active transport pathways. We present a range of techniques which may be used to study the interactions of drugs with membranes both in vitro and in vivo and discuss the advantages and disadvantages of these techniques and highlight new methods being developed to further this field.
Diffusion of macromolecules in the brain: implications for drug delivery Mol. Pharmaceutics , 10 , — DOI: American Chemical Society. Therapeutics must diffuse through the brain extracellular space ECS in order to distribute within the central nervous system CNS compartment; this requirement holds both for drugs that are directly placed within the CNS i.
The diffusion of any substance within the CNS may be affected by a no. Here, we discuss ECS properties important in governing the distribution of macromols.
We also provide an introduction to some of the methods commonly applied to measure diffusion of mols. Finally, we discuss how quant. Characteristics of compounds that cross the blood-brain barrier BMC Neurol. Substances cross the blood-brain barrier BBB by a variety of mechanisms.
These include transmembrane diffusion, saturable transporters, adsorptive endocytosis, and the extracellular pathways. Here, we focus on the chief characteristics of two mechanisms especially important in drug delivery: transmembrane diffusion and transporters. Transmembrane diffusion is non-saturable and depends, on first analysis, on the physicochemical characteristics of the substance.
However, brain-to-blood efflux systems, enzymatic activity, plasma protein binding, and cerebral blood flow can greatly alter the amount of the substance crossing the BBB. Transport systems increase uptake of ligands by roughly fold and are modified by physiological events and disease states.
Most drugs in clinical use to date are small, lipid soluble molecules that cross the BBB by transmembrane diffusion. However, many drug delivery strategies in development target peptides, regulatory proteins, oligonucleotides, glycoproteins, and enzymes for which transporters have been described in recent years.
We discuss two examples of drug delivery for newly discovered transporters: that for phosphorothioate oligonucleotides and for enzymes. Physicochemical high throughput screening: parallel artificial membrane permeation assay in the description of passive absorption processes J. The majority of pharmaceutical discovery projects aim for an orally available form of a new therapeutic principle. One obstacle for per oral treatment is intestinal absorption.
A simple predictive exptl. The authors describe here the PAMPA Parallel Artificial Membrane Permeation Assay method, a simple, robust, high throughput screen HTS which has been shown to be predictive for passive diffusion through membranes and thus for oral absorption. In addn. PAMPA can deliver information in parallel on the lipophilicity, the ionization state and the soly.
High throughput artificial membrane permeability assay for blood-brain barrier Eur. Editions Scientifiques et Medicales Elsevier. The recent advances in high throughput screening for biol. Rapid screening for BBB penetration potential early in drug discovery programs provides important information for compd. The assay was developed with 30 structurally diverse com. The PAMPA-BBB assay has the advantages of: predicting passive blood-brain barrier penetration with high success, high throughput, low cost, and reproducibility.
Elsevier B. This paper compares the permeability of 19 structurally diverse, com. Log BB values. The false pos. Immobilized-artificial-membrane chromatography: measurements of membrane partition coefficient and predicting drug membrane permeability J.
A , , — 28 DOI: Immobilized artificial membranes IAMs are chromatog. IAM surfaces mimic fluid cell membranes. For 23 structurally unrelated compds. This indicates that solute partitioning between the IAM bonded phase and the aq.
IAMs also predicted oral drug absorption in mice and drug permeability through Caco-2 cells. IAM chromatog. Solute retention on IAMs was found to be dominated by a partitioning mechanism.
The structural requirements for HPLC bonded phases to predict solute-membrane partitioning are briefly discussed. Immobilized artificial membrane liquid chromatography: proposed guidelines for technical optimization of retention measurements J. A , , 39 — 53 DOI: Development of the first sphingomyelin biomimetic stationary phase for immobilized artificial membrane IAM chromatography Chem.
Cambridge, U. Evaluated in a proof-of-concept model for blood-brain barrier passage, partial least squares regression demonstrated its potential as an in vitro prediction tool. Immobilized artificial membranes - Screens for drug membrane interactions Adv. Drug Delivery Rev. A review with refs. Immobilized artificial membranes IAMs are monolayers of phospholipid analogs covalently bonded to the surface of silica particles.
These similar interfacial properties have resulted in IAMs' being useful as a physicochem. Chromatographic retention of drug molecules on immobilised liposomes prepared from egg phospholipids and from chemically pure phospholipids Eur. Elsevier Science Ireland Ltd. The partitioning of a chem.
The drug partitioning was assessed from the retention vol. The partitioning of drugs on liposome columns log Ks vs. Statistical anal.
Repeated anal. A close relationship was obsd. EPL columns. The short 'quick screen bilayer columns' permit anal. A moderate to fair rectilinear relationship was obsd.
The drug fraction absorbed in humans showed a similar relationship to log Ks values as to surface plasmon resonance signals representing drug-liposome interaction Danelian et al. Experimental and computational screening models for the prediction of intestinal drug absorption J. The aim of this study was to devise exptl. Both the required exptl. In vitro intestinal Caco-2 cell monolayer permeabilities were detd.
Computational models were built using four different principles for generation of mol. A theor. The results indicate that it is possible to predict intestinal drug permeability from rather simple models with little or no loss of accuracy. A new, fast computational model, based on partitioned mol.
Immobilized liposome and biomembrane partitioning chromatography of drugs for prediction of drug transport Int. Elsevier Science B. Drug partitioning into lipid bilayers was studied by chromatog. The drug retention vol. Log Ks values for pos. A fair correlation was obsd. Compared to the data obtained with liposomes, the retentions of hydrophilic drugs became larger and the range of log Ks values more narrow on the vesicles and the membranes, which expose hydrophilic protein surfaces and oligosaccharides.
Lower correlations were obsd. Absorption of orally administered drugs in humans literature data was nearly complete for drugs of log Ks values in the interval 1.
Both vesicles and liposomes can thus be used for chromatog. QSAR model for drug human oral bioavailability J. The quant. The oral bioavailability detd. A systematic examn. Lipophilicity, expressed as the distribution coeff.
The observation that acids generally had better bioavailability characteristics than bases, with neutral compds. The addn. Rs of 0. In leave-one-out tests, an av. The relationship formulated identified significant factors influencing bioavailability and assigned them quant. The predictive power of the model was evaluated using a sep. Since the necessary physicochem. Also, the model has the advantage of transparency in that it indicates which factors may affect bioavailability and the extent of that effect.
This could be useful in designing compds. Refinement of the model is possible as more bioavailability data becomes available. Potential uses are in drug design, prioritization of compds. Pharmaceutics , 8 , — DOI: Intestinal drug absorption in humans is a central topic in drug discovery. In this study, we use a broad selection of machine learning and statistical methods for the classification and numerical prediction of this key end point.
Our data set is based on a selection of small druglike compds. Using easily available tools, we calcd. We then used decision tree induction DTI , fragment-based lazy-learning LAZAR , support vector machine classification, multilayer perceptrons, random forests, k-nearest neighbor and Naive Bayes anal. Best performance for classification was seen with DTI using the chi-squared anal.
In numeric predictions, the multilayer perceptron performed best, achieving a root mean squared error of In line with current understanding is the importance of descriptors such as lipophilic partition coeffs. However, we are able to highlight the utility of gravitational indexes and moments of inertia, reflecting the role of structural symmetry in oral absorption. Our models are based on a diverse data set of marketed drugs representing a broad chem.
These models therefore contribute substantially to the mol. Relationships between structure and high-throughput screening permeability of peptide derivatives and related compounds with artificial membranes: application to prediction of Caco-2 cell permeability Bioorg. Development of quantitative structure-property relationship models for early ADME evaluation in drug discovery. Blood-brain barrier penetration J. Blood-Brain Barrier Penetration. A new mol. The descriptor was defined based on Kier and Hall's atom-type electrotopol.
Its evaluation requires 2-D mol. A multiple linear regression equation using this descriptor and mol. The results indicate that the lipoaffinity descriptor defined in this paper may be a significant descriptor for mol.
Back to the future: can physical models of passive membrane permeability help reduce drug candidate attrition and move us beyond QSPR? Drug Des. It is widely recognized that adsorption, distribution, metab. The development of computational models to predict small mol.
Empirical qual. Others and we have shown that implicit solvent models to predict passive permeability for small mols. Given the vast increase in computer power, more efficient parallelization schemes, and extension of current atomistic simulation codes to general use graphical processing units, the development and application of phys. Preliminary results from rigorous free energy calcns. Here, we review the current state-of-the-art phys. Lipophilicity and Its relationship with passive drug permeation Pharm.
In this review, we first summarize the structure and properties of biol. Lipophilicity is then introduced in terms of the intermol. Finally, lipophilicity indexes from isotropic solvent systems and from anisotropic membrane-like systems are discussed for their capacity to predict passive drug permeation across biol.
The broad evidence presented here shows that beyond the predictive power of lipophilicity parameters, the various intermol. In silico prediction of membrane permeability from calculated molecular parameters J. Refsgaard, Hanne H. A data set consisting of compds. Nine mol. A model based on five descriptors, no.
In an external test set of compds. Among the compds. Testing physical models of passive membrane permeation J.
Leung, Siegfried S. The biophys. Here, we investigate mol. The exptl. The phys. A systematic approach is taken to analyze the contribution from each component in the physics-based permeability model.
A primary factor in detg. Other factors that improve agreement with exptl. Quantitative correlation of physical and chemical properties with chemical structure: utility for prediction Chem. Katritzky, Alan R. An archetypical example of such transport diseases is Cystinuria , an inherited autosomal recessive disease that is characterized by abnormally high amino acid cystine concentration level in the urine, that may result in the formation of cystine stones in the kidneys.
Another example can be Cystic Fibrosis CF which is caused by a mutation in the cystic fibrosis transmembrane conductance regulator, CFTR, a protein that helps move salt and water across the membrane. It is a genetic disorder that affects mostly the lungs but also the pancreas, liver, kidneys, and intestine.
Long-term issues include breathing problems and coughing up mucus as a result of frequent lung infections. In a patient with CF, the cells do not secrete enough water; when it happens in the lungs, it causes the mucus to become extremely thick. It is also worth mentioning that most fatal toxins like Dendrotoxin black mamba snake of Africa and Batrachotoxin Colombian frog Phyllobates aurotaenia act directly on specific ion channels of the plasma membrane to disrupt the action potentials.
To put it simply, this fatal toxin binds to anionic sites near the extracellular surface of the channel and physically blocks the path and ion conductance. In a nutshell, batrachotoxin irreversibly binds to the sodium channels, enforcing them to remain open.
The permeability of a membrane can be defined as the passive diffusion rate of permeated molecules across the biomembrane. It is unanimously accepted that permeability of any specific molecule depends mainly on charge number, polarity, size, and to some extent, to the molar mass of the molecule.
It should be noted though that both the nature of the bilayer and the prevalent environments can play a significant role too. As mentioned before, because of the inevitable hydrophobic nature of the biomembranes, small uncharged molecules pass across the membrane more easily than charged, large ones [6].
With charged species e. Most cells are characterized by a membrane potential difference of mV V inside - V outside.
Let us first consider an example of Cl - ion to clarify the issue. So, there is a driving force of diffusion for Cl - to diffuse along the concentration gradient into the cell. Therefore, an equilibrium is achieved when influx and efflux of Cl - level each other. The membrane potential at which this equilibrium occurs is called equilibrium potential that can be calculated by Nernst equation [7]:.
Note that this relation was obtained from ion transport equation for zero Gibbs free energy change i. By this definition, negative V DF means passive uptake and exit of cations and anions, respectively.
In such conditions, passive protein channels or active transporters are required for the ion transfer. Superscripts "aq" and "m" denote solute concentrations at bulk aqueous solutions and surfaces of the membrane, respectively. As it can be seen, the concentration gradient is considered to be from S 1 to S 2 , providing the chemical driving force of the transport. To mathematically describe the permeability, let us first introduce the useful concept of partition coefficient.
At thermodynamic equilibrium, the equality of the chemical potentials of solute j in two different intracellular and extracellular phases can be expressed as.
Selectivity of Biomembranes When a membrane separates two aqueous compartments, some chemicals can move across the membrane while others cannot. Based on the transport mechanism and permeability, solutes can be divided into three main groups as follows [2]: Small lipophilic lipid soluble molecules that transfer through the membrane by the sole diffusion.
Molecules that cross the membrane with the aid of protein channels. Very large molecules that do not cross the membrane at all. Small Lipophilic Molecules Passive Diffusion Certain substances easily pass through the membrane by passive diffusion. Polar and Charged Molecules Protein-Mediated Transfer Biological membranes are permeable not only to gases and small lipophilic molecules by passive diffusion processes , but also to many polar and charged molecules, including water, but through a different path.
Large Molecules Membrane Barriers Very large molecules like proteins, polysaccharides or nucleic acids, do not diffuse across the cell membranes at all. Passive and Active Transport Most biologically important solutes require protein carriers to cross cell membranes, by a process of either passive or active transport. Therefore, to summarize, transport of solutes across cell membranes by protein carriers can occur in one of two ways [2]: Downhill movement of solutes from regions of higher to lower concentration level, with the assistance of the protein carrier to pass through the membrane.
This process is called passive transport or facilitated diffusion, and does not require energy. Uphill movement of solute against the concentration gradient driving force from regions of lower to higher concentration. Based on the chemical driving force, this process is unfavorable and requires some form of chemical energy to occur active transport. By forming a protein-lined pathway across the membrane, proteins can appreciably speed up the transfer rate of such solutes.
Moreover, many of these channels are gated. To simply explain the issue, consider that the pathways are closed and unavailable for transport unless specific signals are given.
0コメント