Lipophilicity represents arguably the biggest risk factor in small molecule drug design and as such is central to the Rule of Two. High lipophilicity has been attributed to causing poor solubility, high metabolism and high promiscuity. While to a much lesser extent, low lipophilicity has also been associated with detrimental molecular properties such as low permeability and high renal clearance.
what’s lipophilicity
The word lipophilicity comes from the Greek lipos “fat” and filikos “friendly”, which in practical terms refers to the ability of a compound to dissolve in lipids and lipid-like substances. Lipids are a broad group of naturally-occurring molecules which includes fats, waxes, sterols, fat-soluble vitamins, mono, do and triglycerides and phospholipids. Of all these molecules, “friendliness” for the phospholipids in membranes is the most important, as it is key to membrane permeability. But lipophilicity goes beyond this referring to a compounds propensity to interact with lipid-like regions of proteins (more on this in a lter post). Hydrophobicity (water hating) is a term which is often used interchangably with lipophilicity.
a good start with olive oil
Already at the conclusion of the 19th century (in 1899), Ernest Overton and Hans Meyer got lipophilicity off to a good start. Overton showed that the ‘‘selective solubility’’ of a molecule in the cell’s boundary was responsible for its entry into a cell. Importantly, he demonstrated that the more soluble in lipids the molecule is, the greater its permeability. This has become known as Overton´s Permeability Rule.
The Overton Rule states that the entry of any molecule into a cell is governed by its lipid solubility.
Both Overton and Meyer determined their partition coefficients of analgesics between water and olive oil. Olive oil is composed mainly of the mixed triglyceride esters of oleic acid, linoleic acid, palmitic acid.
![](https://i0.wp.com/mcconnellsmedchem.com/wp-content/uploads/2024/01/olive-oil-triglyceride.jpg?resize=772%2C342&ssl=1)
An example of the main type of mixed triglyceride ester found in olive oil. The fatty acids from top to bottom are palmitic, oleic and linolenic acid.
what are membranes made of
It should be noted that it wasn´t known during Overton’s time what a cell membrane actually was. It was two years after Overton´s death in 1935 that the lipid bilayer membrane model would be conceived by Danielli and Davson. What exactly the lipid content of a cell membrane is, is not such a simple question as it differs between tissues and cell type. The lipid fraction of membranes consists of phosphatidic acid derivatives, glycolipids, sphingolipids and cholesterol. Lipids themselves also have an impressive degree of diversity, including variability of the polar head groups (choline, ethanolamine, serine can be added to the phosphate group) and the non-polar fatty acids hydrocarbon chains (hydrocarbon tail length and double bonds number or position can vary). Phosphatidyl cholines (PC) and phosphatidylethanolamines (PE) are the two main cellular lipids in membranes making up 50 and 20 mol% respectively. To make things even more complicated, significant variation of membrane lipid composition has been observed across cancer cells.
![](https://i0.wp.com/mcconnellsmedchem.com/wp-content/uploads/2024/01/phosphatidyl-choline-and-ethanolamine-1.jpg?resize=1024%2C593&ssl=1)
Example structures of phosphatidylcholine and phosphatidylethanolamine
measuring fat friendliness
Lipophilicity is measured as a partition coefficient (log P) which describes the partition equilibrium of an unionized solute between water and an immiscible organic solvent. A logP of zero indicates a compound that has equal concentration in both phases at equilibrium, while a compound with a logP of 3 is 1000-fold more concentrated in the lipophilic solvent versus water. The most common lipophilic solvent used is octanol – which is where the problem starts.
who´s to blame for octanol
It was Collander who first used octanol in 1951 to measure partition coefficients of compounds with water. But it was Meyer who steered him in this direction, suggesting back in 1937 that oleyl alcohol would be a good mimic for membrane lipids. Hansch then in 1962 adopted octanol to measure the partition coefficients of phenylacetic acid derivatives with no mention of why he chose octanol. In the following year, Hansch cemented octanol’s place in history with his QSAR analysis of the biological activity of auxin plant growth regulators. Here he provided a justification for his use of octanol.
We have chosen octanol and water as a model system to approximate the effect of step 1* on the growth reaction in much the same fashion as the classical work of Meyer and Overton rationalized the relative activities of various anesthetics.
*step 1 being auxin going from the solution state to the site of action.
Hansch´s systematic work in the 60s and 70s (Hansch and Fujita and Leo) established the octanol– water as the reference system for lipophilicity measurement.
the problem with octanol
Passive permeability of small molecules through membranes is thought to occur via the successive desolvation of the small molecule as it passes through the membrane. This process is energetically disfavored due to the progressive breakage of ligand-water H-bonds until the ligand is completely desolvated at the center of the bilayer. Hence, it is the central portion of the lipid bilayer that needs to be mimicked for passive permeability.
The problem with octanol is not so much that there are no alcohol groups at the center of a lipid bilayer, but the asymmetry of it’s H-bonding strength versus water. The H-bond donor strength of octanol’s OH group and water are essentially the same (Hunter’s α = 2.7 and 2.8 respectively, for a introduction to Hunter’s H-Bonding Constants see In Drug Design Two Thirds of H-bonds are Weak). So far so good. However, the oxygen atom of alcohols is a much stronger H-bond acceptor than water (β = 5.8 versus 4.5 respectively). This means that H-bond donors in solute molecules will be preferentially solvated by octanol over water, leading to an over-estimation of the lipophilicity of compounds containing H-bond donors.
![](https://i0.wp.com/mcconnellsmedchem.com/wp-content/uploads/2024/01/the-problem-with-octanol.jpg?resize=1024%2C547&ssl=1)
The H-bond acceptor strength assymetry of octanol versus water
how big is the problem
Let’s take a look at an example to see how big the problem is. Anisole (methoxybenzene) and p-cresol (4-methylphenol) have the same molecular weight and differ only on the positioning of the methyl group. The methyl group of anisole is attached to the phenolic oxygen and as such anisole does not contain a H-bond donor. The methyl group of p-cresol is at the 4-position of the benzene ring which leaves it with one phenolic H-bond donor. Given the difference of one H-bond donor between the two molecules of equal molecular weight, p-cresol should be significantly less lipophilic. However, using the octanol-water partitioning system only a minor reduction is observed (2.11 for anisole and 1.95 for p-cresol).
The octanol/water partitioning system is blind to H-bond donors.
Pioneered in recent times by Pete Kenny is the use of alkanes instead of octanol. The alkane-water partitioning system not only better mimics the center of lipid bilayers structures but also does not suffer from the H-bonding assymetry of octanol as alkanes treat H-bond donors and acceptors equally. So let’s see what the logP of our anisol/p-cresol compound pair looks like in alkane-water. Now (as expected) p-cresol is over two orders of magnitude less lipophilic than anisole (-2.52 for p-cresol and -0.23 for anisole). This dramatic difference between octanol and alkane highlights how significantly the lipophilicity of molecules with H-bond donors are over-estimated when using logP(octanol-water).
Recommended Lipophilicity Reading
- Waring, Michael J. “Lipophilicity in drug discovery.” Expert Opinion on Drug Discovery 5.3 (2010): 235-248.
- Arnott, John A., and Sonia Lobo Planey. “The influence of lipophilicity in drug discovery and design.” Expert opinion on drug discovery 7.10 (2012): 863-875.
- Lobo, Sonia. “Is there enough focus on lipophilicity in drug discovery?.” Expert opinion on drug discovery 15.3 (2020): 261-263.
- Ran, Yingqing, Neera Jain, and Samuel H. Yalkowsky. “Prediction of aqueous solubility of organic compounds by the general solubility equation (GSE).” Journal of chemical information and computer sciences 41.5 (2001): 1208-1217
2 comments
Appreciate your writings. Is there a factor that can be developed to adjust the CLogP value for compounds having HBDs?
No I don’t think so. As HBDs differ in strength a simple fudge factor isn’t sufficient and as you can see from the post the effect can be massive. f u are working within a series (eg the H-bond donor functionality is always constant) then the lipophilicity is over-estimated but the delta logPs for matched molecular pairs are correct.
Comments are closed.