July 11, 2024
by Bara Townsend, Marketing Communications Specialist
The Importance of Physicochemical Properties in Drug Design
In pharmaceutical development, the balance between lipophilicity and hydrophilicity of a drug candidate is crucial. For decades after he published his paper “Experimental and computational approaches to estimate solubility and permeability in drug discovery and development settings”, Lipinski’s Rule of Five (Ro5) served as a guideline for identifying orally active drugs.
The rules proposed that a good “druggable” compound should follow at least three of the following:
-
- Fewer than 5 H-bond donors
- No more than 10 H-bond acceptors
- The molecular weight should not exceed 500 Da
- The calculated octanol-water partition coefficient, logP, should be less than 5
These rules were widely utilized across the industry to speed up drug discovery, forcing chemists to consider the physicochemical properties of a molecule, such as size and lipophilicity, during the drug design process.
However, it is now widely believed that overreliance on these rules could lead to missed opportunities for viable therapeutics, especially when considering borderline compounds that may be deemed as unsuitable based on Ro5. Indeed, as the explored chemical space expands beyond small molecules, there is an increasing number of approved oral drug compounds that go beyond the rule of 5 (bRo5).
Challenging the Norm—Beyond the Rule of 5
Even Dr. Lipinski later acknowledged that some protein target sites will only bind to larger compounds outside the remits of Ro5. The expansion into bRo5 space gives scientists the freedom to explore potential therapeutic candidates with much higher molecular weights. These often have the ability to fold to hide their hydrogen-bond donors and acceptors, leading to often unexpected properties of molecules, including surprising oral bioavailability.
This revelation has led to scientists moving away from Ro5, with some groups defining the bRo5 chemical space which is thought to contain “possible oral compounds”.
The revised parameters for the bRo5 space proposed by some research groups include:
-
- Molecular weights of < 1000 Da
- Calculated logP between -2 and 10
- Fewer than 6 H-Bond donors
- No more than 15 H-Bond acceptors
Additional constraints since the original Ro5 include:
-
- Maximum polar surface area of 250 A2
- Fewer than 20 rotatable bonds1
It should be noted that, similar to Ro5, these “rules” should still be viewed as guidelines, as oral drugs exist even outside this redefined space.
Given the increased interest in bRo5, it may come as no surprise that many researchers are now exploring larger compounds, such as macrocycles, protein-based agents, and multispecific drugs, including antibody-drug conjugates (ADCs) and proteolysis targeting chimeras (PROTACs) as potential therapeutic modalities. While much of the focus during the drug design process remains on oral bioavailability, chemists must also consider the absorption, distribution, metabolism, excretion and toxicity (ADMET) behavior of these compounds, all of which are highly dependent on the lipophilicity, hydrophilicity, and crucially, the ionization state of the drug candidate.
Although drug design is shifting beyond the rule of five, one of the properties that is still considered key in determining the oral bioavailability of compounds is their lipophilicity. It should be noted that the overall lipophilicity of any compound is dependent on its ionization state at a given pH. However, logP, the value that was considered in the original Ro5, does not account for ionization.
Which leads us to the question:
Is considering logP of a compound enough to gain a comprehensive understanding of its behavior in varying biological environments or do we need to delve deeper?
LogP—the Basic Measure of Lipophilicity
Most, if not all, chemists will have heard of the partition coefficient at some point during their career, although not everyone will quickly recall exactly what it means in terms of drug design.
The partition coefficient, logP, quantifies a compound’s distribution between two immiscible liquids, typically octanol and water. The higher the value, the more lipophilic the compound. This will influence its ability to cross cell membranes, making logP a vital metric to account for in the early stages of drug development.
Yet it’s not this simple.
LogP calculations assume the compound is in its unionized form. If a compound ionizes, it’s no longer the same structure and the calculations become obsolete. Since a large proportion of compounds investigated in pharmaceutical research contain ionizable sites, as discussed in our post on the Importance of Ionization in Pharmaceutical R&D, it’s not logP that should be considered, but logD.
LogD—the pH Dependent Lipophilicity Value
So what is logD, and what is the difference between logP and logD?
Unlike logP, which is pH-independent, the distribution coefficient, logD, changes with pH as its calculations account for all forms of a compound at a specific pH, including ionized, partially ionized, and unionized species. Similarly to logP, the higher the value of logD, the more lipophilic the compound. Equally, the lower the value, the more hydrophilic, and therefore more water soluble, the compound.
Therefore, logD offers a much more accurate picture of a compound’s behavior in various biological environments, where the pH can be expected to differ, making logD predictions an invaluable tool for drug development.
Changing pH Environments in Physiological Conditions
Let’s consider the gastrointestinal (GI) tract as an example. Although chemists are moving away from the Rule of 5, oral drug administration remains the most desirable from of drug delivery. As illustrated by Figure 1, the GI tract has a variety of pH environments, all of which an orally administered compound is likely to encounter.
Figure 1 Changing pH environment in the GI tract.
A compound such as 5-methoxy-2-[1-(piperidin-4-yl)propyl]pyridine, which has two ionization centers (pyridine with pKa 4.8 and piperidine with pKa 10.9), would ionize to a different extent in each compartment of the GI tract, as shown in Figure 2.
Figure 2 The changing ionic forms of 5-methoxy-2-[1-(piperidin-4-yl)propyl]pyridine with pH.
At physiologically relevant pH (1–8), it is evident that ionization drastically affects the distribution coefficient of the compound as shown in Figure 3, and therefore its behavior in aqueous media.
Figure 3 The logD curve of 5-methoxy-2-[1-(piperidin-4-yl)propyl]pyridine.
Crucially, Figure 2 also shows that at physiological pH, the neutral form of the compound is practically non-existent. The logP of this compound suggests a lipophilic compound with high membrane permeability, and although this may be true at pH > 12, it certainly wouldn’t be the case as physiologically relevant pH. Based on the logD prediction, we can conclude that at pH < 8, the compound is likely to have high solubility in aqueous media and low lipophilicity, contrary to the predicted properties based on logP alone.
Uses of Partition Coefficients Beyond Pharma/BioPharma
Although the example above is relevant to drug discovery, the reasoning applies to other fields. Environmental chemists might study the behavior of chemicals affected by the pH of different soils or bodies of water. They would be interested in the partitioning of the species that actually exist at that pH, not just in the neutral species that may not be the dominant form.
Additionally, an understanding of the partition and distribution coefficients is often needed when developing new separation and chemical extraction methods. Knowledge of a compound’s logP and logD values allows chemists to select appropriate separation conditions, for example, the predictive calculations can be used to determine an optimal pH for the separation of positional isomer, maximizing the extraction yield and product purity as a result.
Tell your Ps from your Ds
In summary, logP tells us about the properties of a compound in its neutral state; logD is pH dependent and should be reported so. Both coefficients are valuable tools when assessing a compound’s lipophilicity and hydrophilicity, which give us information about the likely membrane permeability and solubility in aqueous media. However, if a compound contains ionizable sites, it is more accurate and appropriate to use logD over logP.
The Future of Physicochemical Property Prediction
In the current digital age and the rising use of artificial intelligence, it would be naïve of us to ignore its potential impact in drug discovery. There is already emerging evidence of increasing investment in AI in the pharmaceutical industry, however, these investments have not yet been shown to impact attrition in drug discovery.
Watch the presentation below to learn more about the importance of selecting and using accurate physicochemical predictive models, and the use of generative AI alongside predictive software.
Software for Physicochemical Property Prediction
ACD/Labs’ predictive software on the Percepta Platform provides scientists with the information needed for data-driven decisions without the need to measure physicochemical properties of target compounds. PhysChem Suite offers predictors for both logP and logD, as well as calculators for other important physicochemical properties such as ionization (pKa), aqueous solubility, rule of 5 compliance and more, enabling you to investigate compound properties and structure modifications with efficiency.
References
- B. Doak, B. Over, F. Giordanetto, J. Kihlberg. (2014). Oral Druggable Space beyond the Rule of 5: Insights from Drugs and Clinical Candidates. Chemistry & Biology, 21, 1115-1142. DOI: 10.1016/j.chembiol.2014.08.013
About the Author
