Skip to main content
NCBI home page
ACS Omega logoLink to ACS Omega
. 2026 Feb 27;11(9):14616–14625. doi: 10.1021/acsomega.5c10320

Biomimetic Partitioning and Pharmacokinetic Insights into Phytocannabinoids

Jared Kainalu Martin †, Derek J Muensterman †, Boris Droz ‡, Klara Valko §, Lisa Truong †, Robyn L Tanguay †, Jennifer A Field †,*
PMCID: PMC12980171  PMID: 41835504

Abstract

Phytocannabinoids from Cannabis sativa exhibit therapeutic potential, yet historical research restrictions have limited the ability to acquire key physicochemical data. This study employs biomimetic chromatography to characterize the partitioning behavior of eight neutral and seven acidic phytocannabinoids using octadecylsilica (C18), an immobilized artificial membrane (IAM), human serum albumin (HSA), and α-1-acid glycoprotein (AGP) columns. Chromatographic hydrophobic index (CHI) scaled to octanol–water partition coefficient (log D) values ranged from 4.16 to 6.06 for neutral compounds and 2.09–3.15 for acidic ones. Despite lower lipophilicity, acidic phytocannabinoids showed stronger phospholipid and HSA binding at pH 7.4, driven by electrostatic interactions. Predicted pharmacokinetic parameters indicate high lung and brain-tissue affinity, consistent with psychoactive and therapeutic effects, while human-ether-a-go-go-related gene inhibition suggests cardiotoxicity risk. These findings provide a comprehensive data set for predicting in vivo distribution and assessing safety, highlighting biomimetic chromatography as an effective approach for advancing cannabinoid pharmacology.

graphic file with name ao5c10320_0011.jpg

graphic file with name ao5c10320_0009.jpg

Introduction

Cannabis (Cannabis sativa) is a versatile crop cultivated for a wide range of commercial and consumer products, including industrial materials, food ingredients, and pharmaceuticals. The crop is used as feed for both terrestrial and aquatic livestock, as fibers for textiles, or for their naturally occurring phytochemicals. − Some of these naturally occurring, bioactive chemicals in C. sativa are phytocannabinoids, which are structurally diverse and include well-known molecules such as Δ9-tetrahydrocannabinol (Δ9-THC) and cannabidiol (CBD) along with numerous minor cannabinoids with therapeutic potential. Despite growing interest in their pharmacological properties, research has been constrained by decades of regulatory restrictions, particularly in the United States, after implementation of the Controlled Substances Act in 1970. Restriction on access to C. sativa has led to the limited acquisition of experimental data on key physicochemical parameters such as lipophilicity, protein binding, and tissue distribution. − The lack of foundational data hampers accurate prediction of pharmacokinetics and safety profiles, particularly for less-studied acidic and neutral phytocannabinoids.

The 2018 Farm Bill in the United States legalized hemp, defined as C. sativa containing 0.3% or less Δ9-THC by dry weight, expanded access to hemp, and opened new opportunities for hemp-related research. Advances in the therapeutic applications of phytocannabinoids include the treatment of cancer, inflammation, and neurological diseases. ,− While more information is available for Δ9-THC and CBD, there is relatively little information for minor phytocannabinoids, such as cannabigerol (CBG) or cannabichromene (CBC). Most phytocannabinoids lack measured properties including octanol–water partition coefficients and protein binding that are needed for predicting biological distribution and bioactivity. − Therefore, while the lipophilic nature of some phytocannabinoids is well documented, distribution throughout the body is not well understood.

Biomimetic chromatography is a high-throughput technique that relies on separations performed on analytical columns packed with biologically relevant stationary phases. Retention times on an octadecylsilica (C18) column at three different pHs (2.6, 7.4, and 10.5) are used to determine acid–base characteristics and lipophilicity. − Retention times on immobilized artificial membrane (IAM), human serum albumin (HSA), and α-1-acid glycoprotein (AGP) stationary phases are used to calculate phospholipid–water (log k IAM), human serum albumin–water (log k HSA) partition coefficients as well as α1-glycoprotein logarithmic retention factors (log k AGP), respectively. ,− The pharmaceutical industry uses the rapid-gradient biomimetic chromatography approach for early drug development. Biomimetic chromatography is used to estimate pharmacokinetic parameters, such as tissue and cell partitioning, and for estimating the potential for toxicological parameters, such as gene and protein inhibition. ,, To date, biomimetic chromatography has been used to determine IAM and HSA partition coefficients for six neutral phytocannabinoids phospholipids. Compared to HSA, AGP is typically present in lower concentrations in blood, yet it is an important protein that is involved in storing and transporting endogenous and exogenous lipophilic compounds throughout the body. , To the best of our knowledge, there are no partition coefficients for any acidic phytocannabinoids or for retention factors for any phytocannabinoids for AGP. Furthermore, no biomimetic binding data have been used to confirm acid–base characteristics or to estimate pharmacokinetic parameters of phytocannabinoids. Rapid-gradient biomimetic chromatography is a high-throughput technique for acquiring a large quantity of data for a large group of compounds than can be easily expanded to include new compounds.

The objective of this study was to determine the acid–base characteristics, lipophilicity, phospholipid, and protein partition coefficients for 15 phytocannabinoids (Table ). Each phytocannabinoid was evaluated on four biomimetic columns including C18 at three different pHs (2.6, 7.4, and 10.5), IAM, HSA, and AGP. The partition coefficients from each column were used to estimate % lung-tissue binding (% LTB), % brain-tissue binding (% BTB), and human-ether-a-go-go-related gene (hERG) inhibition potential.

1. Phytocannabinoid Names, Abbreviations, Structures, Neutral Formulas, CAS Numbers, and Molecular Weight .

graphic file with name ao5c10320_0007.jpg

graphic file with name ao5c10320_0008.jpg

Open in a new tab
a

An *, ∧, +, or # indicates a phytocannabinoid belonging to a group of constitutional isomers.

Materials and Methods

Standards and Reagents

Acetonitrile, isopropyl alcohol (IPA), methanol, and deionized water were purchased from Fisher Chemical (Waltham, MA) in LC-MS grade optima. Buffers of ammonium acetate and ammonium hydroxide were purchased from Fischer Chemical (Waltham, MA) and formic acid was purchased from Sigma-Aldrich (St. Louis, MO) in LC-MS grade.

Specific pharmaceutical calibration test mixes for each column were purchased from Biomimetic Chromatography Ltd. (Stevenage, U.K.). Adaptations to the pharmaceutical test mixture compositions were made to improve detection using a quadrupole time-of-flight (QToF) mass spectrometer. Standards for neutral phytocannabinoids (Table ) were obtained from Cerilliant (Round Rock, TX), or RESTEK (Bellefonte, PA), while standards for acidic phytocannabinoids were obtained from Supelco (Darmstadt, GER). Neutral and acidic phytocannabinoid standard mixes, composed of constitutional isomers (Table ), were purchased from Cerilliant (Round Rock, TX).

Phytocannabinoid neutral and acidic standard mixtures contained 100 μg/mL of each phytocannabinoid except CBD and cannabidiolic acid (CBDA), which were present at 250 and 500 μg/mL, respectively. Samples consisted of 15 μL of either the neutral cannabinoid mixture or the acidic cannabinoid mixture combined with 135 μL of 1:1 water/mobile phase B using acetonitrile for C18 and IAM experiments and IPA for HSA and AGP experiments. Because there were constitutional isomers present in the standard mixtures (Table ), phytocannabinoid isomers were injected as individual standards. Constitutional isomers with the same molecular weight (e.g., Δ8-tetrahydrocannabinol (Δ8-THC), Δ9-THC, CBD, and CBC; Table S2), when injected as a mixture, gave a single broad peak. Therefore, injections of individual isomer standards were used, and samples consisted of 7.5 μL of an individual phytocannabinoid standard at 100 μg/mL and 143 μL of a 1:1 water/solvent mixture, as described above. For a phytocannabinoid that belonged to a group of isomers, the retention time from its analysis as an individual standard was used to derive chromatographic hydrophobicity index (CHI) values on the C18 and IAM columns and log k values on the HSA and AGP columns. Data for nonisomers were obtained from the injection of mixtures. Pairs of neutral and acidic analogs included Δ9-THC and Δ9-tetrahydrocannabinolic acid (THCA); CBD and CBDA; CBG and cannabigerolic acid (CBGA); cannabidivarin (CBDV) and cannabidivarinic acid (CBDVA); tetrahydrocannabivarin (THCV) and tetrahydrocannabivarinic acid (THCVA); and cannabinol (CBN) and cannabinolic acid (CBNA). Additional information on mobile phase composition, calibration curves, and column care are provided in the SI.

Biomimetic Chromatography

Biomimetic separations were performed on an Agilent 1260 infinity II high performance liquid chromatograph using an injection volume of 10 μL and chromatographic columns kept at 25 °C. A column temperature of 37 °C (e.g., physiological temperature) gives retention times that are shorter by ∼0.05 min for test compounds and drug calibrants (unpublished data). However, because the HSA and other binding values are calculated from relative retention times, temperature shifts cancel out. Furthermore, operating the HSA columns at that temperature significantly reduced column lifetimes. For these reasons, all measurements were performed at 25 °C.

The C18 retention times were measured with a 3 × 50 mm2 Gemini NX C18 column with a 5 μm particle size (Phenomenex Ltd. Macclesfield, U.K.) at pH 2.6 (10 mM formic acid buffer), 7.4, and 10.5 (both 50 mM acetate buffer) to obtain CHI values. Retention times were measured at pH 7.4 on an IAM column (4.6 × 150 mm2 IAM.PC.DD2 with a 10 μm particle size; Regis Technologies Inc., Morton Grove, IL) to obtain CHI IAM and log k IAM values. Retention times on HSA and AGP columns were measured at pH 7.4 on a Chiralpak HSA and Chiralpak AGP column (each 3 × 50 mm2, Daicel Corporation, West Chester, PA), each with a 5 μm particle size, respectively, to obtain log k HSA and log k AGP values, respectively. The calculated deadtimes (see SI for details) for the C18, HSA, and AGP columns were 0.23 min, while the calculated deadtime for the IAM column was 1.06 min. Detection of phytocannabinoids was performed on an SCIEX X500R QToF under a resolution of 31,000 at 350 m/z interfaced with an electrospray ionization source that was operated in positive-ion mode (Table S3) using <10 ppm mass error and <20% isotope ratio difference. Calibration of each column using pharmaceutical standards is described in the SI. Pharmaceutical calibration standards were injected in triplicate after every 10 samples to check for any drift in the calibration curve.

Data Analysis

Average retention times (Tables S4–S6) computed from triplicate injections for each phytocannabinoid on each column and the regression equation for each column calibration curve were used to calculate partition coefficient for each column. The standard deviation for triplicate injections was <0.1 for each phytocannabinoid. Regression equations for each column can be found in the SI (Table S7 and Figures S1–S3). Values of CHI at each pH on the C18 column were converted into CHI log D values (Table S8 and eq S3). The retention times obtained from the IAM and HSA columns were converted into retention factors (log k IAM and log k HSA, respectively). These factors were then used to compute log k IAM (Table S9) and log k HSA values (Table S10), respectively, for comparison with other linear free energy parameters. ,,, Retention times on the AGP column was used to compute the log k AGP retention factors (Table S11); no transformation of log k AGP was necessary since the k values directly express equilibrium partitioning. Various calculated parameters were then used to predict in vivo distributions including % lung-tissue binding, hERG inhibition, and % brain-tissue binding (eqs S7–S14).

Results and Discussion

Lipophilicity (C18 Column)

The largest CHI log D (Table S2) value determined over three pH values on a C18 column is comparable to the octanol–water partition coefficient, which is a measure of lipophilicity. Neutral phytocannabinoids gave CHI log D values that did not vary significantly by pH (Figure ), which is consistent with their lack of ionizable functional groups with pK a values in the ranges of 2 and 10 (see Table S8 for the acid–base characterization of each phytocannabinoid). The greatest CHI log D values across all three pH values were obtained for neutral phytocannabinoids at pH 7.4 and ranged from 4.1 to 5.85, which are in good agreement (71–113%) with reported octanol–water partition coefficients from the literature (Table S2). Poorer agreement (40–94%) was obtained with values determined by Estimation Program Interface Suite v4.1 from the U.S. EPA (Table S2), which were up to 2 log units greater. Ciura et al. measured CHI values on a C18 column at pH 7.4 for six neutral phytocannabinoids, which were in good agreement (96–135%) with those obtained in this study (Table S8).

1.

1

Open in a new tab

CHI log D (lipophilicity) values for neutral and acidic phytocannabinoids at pH 2.6, 7.4, and 10.5. Groups of isomers are denoted by common symbols *, ∧, +, and #. The error bars are too small to be clearly seen in the figure (SD < ±0.11).

The largest CHI log D values for acidic phytocannabinoids were obtained at pH 2.6 (Table S2) where the carboxylic acid groups are protonated; see the SI for the acid–base characterization for each phytocannabinoid acid. The highest CHI log D values for acidic phytocannabinoids ranged from 4.33 to 5.83 (Table S2 and Figure ) and were in good agreement with measured values for four acidic phytocannabinoids (71–113%) and within 62–94% of those estimated using EPI Suite (Table S2). At a physiologically relevant pH of 7.4, the acidic phytocannabinoids are 2 orders of magnitude less lipophilic, as indicated by CHI log D values in Figure . ,

The structural differences among constitutional isomers appear in their CHI log D values. CBD has two hydroxyl groups, while Δ8- and Δ9-THC and cannabichromene (CBC) have only one hydroxyl group and an ether. This slight structural change results in CBD having the lowest CHI log D value for that group of constitutional isomers (Figure ). The CHI log D values between the four constitutional isomers Δ9-THC, Δ8-THC, CBD, and CBC at the three pH values gave standard deviations of 0.47, 0.45, and 0.43, respectively, while those for CBDV and THCV isomers were 0.71, 0.69, and 0.69, respectively. The isomers CBDA, THCA, and cannabicyclolic acid (CBLA) isomers gave standard deviations of 1.40, 0.32, and 0.22, respectively, while CBDVA and THCVA were 1.5, 0.35, and 0.29, respectively. Thus, biomimetic chromatography is sensitive enough to detect small differences among constitutional isomers.

Phospholipid Binding on Immobilized Artificial Membrane (IAM) Column

For neutral phytocannabinoids, log k IAM values ranged from 4.46 to 8.03, while those of acidic phytocannabinoids ranged from 1.86 to 8.32 (Figure and Table S9). Cannabichromene (CBC) gave the highest log k IAM value for neutral phytocannabinoids, while CBDV had the lowest. Of the acidic phytocannabinoids, THCA gave the greatest log k IAM value, while CBGA gave the lowest log k IAM. Good agreement (78–99%) was obtained between the CHI IAM values for neutral phytocannabinoids, including Δ9-THC, CBD, CBC, CBG, CBN, and CBDV compared to those of Ciura et al. (Table S9). No comparative data are available for acidic phytocannabinoids. Among the four sets of isomers, variation in log k IAM values (Figure and Table S9) indicates that the rapid-gradient approach is sensitive enough to small changes in the structure.

2.

2

Open in a new tab

Log k IAM values for neutral and acidic phytocannabinoids. The red threshold (log k IAM = 6.24) indicates strong phospholipid binding, strong tissue partitioning, and low drug efficiency. Groups of isomers are denoted by common symbols *, ∧, +, and #. The error bars are too small to be clearly seen in the figure (SD < ±0.05).

Neutral phytocannabinoids binding to the IAM membrane is almost exclusively driven by lipophilicity as indicted by values above the 1:1 line in a plot of log k IAM vs CHI log D (Figure S4), which indicates a stronger affinity for the IAM membrane than predicted by lipophilicity alone (CHI log D). Acidic phytocannabinoids appear above the 1:1 line and are attributed to electrostatic interactions between the positively charged region of the zwitterionic phosphatidyl choline group of the IAM membrane and the negatively charged acidic phytocannabinoids. For pairs of acidic and neutral analog phytocannabinoids (e.g., Δ9-THC vs THCA and CBD vs CBDA), the acidic analog gave a higher log k IAM value (pH 7.4) than the neutral analog, except for THCA and Δ9-THC (Table S9).

Compounds that give log k IAM values greater than a threshold value of 6.24 (Table S9) are expected to exhibit strong phospholipid binding, strong tissue partitioning, and low drug efficiency. , None of the neutral phytocannabinoids exceeded the threshold (Figure ), which is consistent with reports that they partition into adipose tissue, remain associated with cell membranes, and cross the blood–brain barrier while remaining bioactive. , Only two acidic phytocannabinoids exceeded the log k IAM threshold: THCA and CBNA. Should any of these compounds produce therapeutic effects, their doses would need to be calculated in a way that accounts for high retention in fatty tissue. Similar to the isomeric differences observed on the C18 column, the binding effect of two hydroxyl groups versus that of one hydroxyl group and one ether group produces a lower log k IAM for CBD compared to that of Δ9-THC.

HSA Protein Binding

The neutral phytocannabinoids yielded log k HSA values ranging from 3.89 to 4.18 (Figure and Table S10). The log k HSA values for six neutral phytocannabinoids (e.g., Δ9-THC, CBD, CBC, CBG, CBN, and CBDV) in the present study are within 67–75% (Table S10) of those obtained on an HSA biomimetic column. Compared to log k IAM values, phytocannabinoids, isomers yielded very similar log k HSA values (Figure and Table S10). Acidic phytocannabinoids yielded higher log k HSA values, ranging from 4.42 to 5.01 (Figure and Table S10).

3.

3

Open in a new tab

Log k HSA and retention factor log k AGP for neutral and acidic phytocannabinoids. Groups of isomers are denoted by common symbols *, ∧, +, and #. The error bars are too small to be clearly seen in the figure (SD < ± 0.02.)

Binding of neutral and acid phytocannabinoids to HSA is within the same order for all compounds (log k HSA values between 3.89 and 5.01). However, neutral phytocannabinoids fell below the 1:1 relationship between log k HSA and CHI log D at pH 7.4, indicating low hydrophobicity (Figure S5). In contrast, acidic (anionic) phytocannabinoids fell above the 1:1 line due to electrostatic interactions with the HSA protein, which has multiple binding sites. Values of HSA binding <99% are consistent neutral phytocannabinoid biological and pharmacological activity. ,,,− Low (<99%) HSA binding of the acidic phytocannabinoids is consistent with their bioactivity and therapeutic potential. The isomeric differences observed on the C18 and IAM columns were not observed on the HSA column (Figure ). This is likely because binding to HSA is not driven by lipophilicity but electrostatic interactions and conformation of the protein and its binding sites.

α-1-Acid Glycoprotein Binding

Neutral phytocannabinoids gave retention factor log k AGP values that ranged from 0.43 to 0.53, which translates to 73–78% AGP binding (Table S11). Phytocannabinoid acids yielded lower log k AGP values (0.18–0.39), which may be due to electrostatic repulsion of the anionic acidic phytocannabinoids at pH 7.4. Isomers gave more variation in log k AGP values (Table S11) and may result from more complex binding with contributions from lipophilicity, electrostatic interactions, and steric limitations. The factor log k AGP of neutral phytocannabinoids were strongly correlated with those of their acidic analog (Figure S6). The log k AGP values of neutral phytocannabinoids were, on average, 0.17 units higher than their acidic phytocannabinoids analogs (Figure ). The low (<1) log k AGP values for all of the phytocannabinoids suggests that AGP might not be a major transporter for these compounds. To the best of our knowledge, there are no comparator data in the peer-reviewed literature for phytocannabinoid interactions with AGP. The relationship between log k AGP and lipophilicity (CHI at pH 7.4) indicates that lipophilicity contributes to neutral and acidic phytocannabinoid binding to AGP (Figure S6).

Unlike HSA, which has a more stable presence in the plasma, AGP plasma concentrations vary and respond to insults to the body, such as inflammation or illness. Thus, phytocannabinoids with AGP binding greater than 75% are expected to have variable potency, depending on AGP concentrations in the plasma as a function of disease state of an individual. Of the neutral phytocannabinoids, only CBDV and THCV did not exceed the 75% AGP binding threshold (Table S11) that correlates with variable biological activity in clinical settings. , Future pharmaceutical applications based on neutral cannabinoids need to be designed to account for the disease states of recipients. In contrast, all acidic phytocannabinoids fell below the 75% threshold (Table S11), which indicates that their efficacy in a clinical or therapeutic setting is not dependent on the disease state of the individual (e.g., AGP concentration). Although HSA and AGP both have an effect on the biological transport and therapeutic efficacy of phytocannabinoids, HSA would be expected to have a stronger, more consistent effect due to its higher, more constant concentration in the plasma.

Pharmacokinetic Models

Potentially, the most impactful aspect of biomimetic chromatography is its ability to predict in vivo distributions and various types of tissue binding. Through the measurement of retention and interactions of well-characterized pharmaceuticals on biomimetic columns, models have been developed to estimate biological parameters on any analytical instrument calibrated using the same calibrants. Correlative models developed through the screening of many pharmaceuticals allows for the high-throughput screening of pharmacokinetic parameters using already-established biomimetic models.

The % LTB for each phytocannabinoid exceeded the 98% threshold (Figure and Table S12), which indicates that these phytocannabinoids can be expected to be retained in lung tissue. Although % LTB is dependent on the interactions with HSA and AGP proteins (eqs S8–S10), % LTB is influenced by liphophilicity, as indicated by CHI log D values at 7.4 (eq S10). Variations in % LTB among phytocannabinoid isomers stem from differences in lipophilicity (CHI log D) among isomers (Figure ). Lung-tissue binding is particularly relevant because inhalation via smoking and vaping are the most popular modes of recreational cannabis administration. The highest amount of Δ9-THC in pigs and in post-mortem humans that received cannabis orally was found in the lungs. ,

4.

4

Open in a new tab

Percent lung-tissue binding (% LTB) of neutral and acidic phytocannabinoids and the red-line threshold (98%) associated with retention by lung tissue. Groups of isomers are denoted by common symbols *, ∧, +, and #.

All 15 phytocannabinoids exceeded the threshold hERG inhibition value of ≥5, which is indicative of the potential for cardiotoxicity. Estimates of hERG are weighted toward interactions with AGP protein and to a lesser extent with IAM membranes (eq S11), thus variations among isomers are likely due to differences in k AGP values (Table S11). The highest values were obtained for two acidic phytocannabinoids, CBGA and CBNA (Figure and Table S12). The inhibition of hERG in guinea pigs and rabbits following CBD exposure has been previously reported. Inhibition of hERG may explain physiological effects of cannabis, including neuromodulation and altered cardiac function. The inhibition of hERG by phytocannabinoids must be considered for any future therapeutic application of phytocannabinoids, especially when considering drug–drug interactions and patients with cardiovascular conditions. The hERG inhibition correlative model has the potential to direct further toxicity testing.

5.

5

Open in a new tab

Human-ether-a-go-go-related gene (hERG) inhibition potential for neutral and acidic phytocannabinoids with a red-line threshold (hERG inhibition potential = 5) associated with potential for cardiotoxicity. Groups of isomers are denoted by common symbols *, ∧, +, and #.

The % BTB is an estimate of a compound’s ability to bind to brain tissue after it has crossed the blood–brain barrier. Only CBGA and cannabicyclolic acid (CBLA) had less than 99% BTB (Figure ). The closely related parameter, brain-to-plasma ratio, is an estimate of a compound’s partitioning between plasma proteins and brain tissue (Table S12). Only CBDV, CBGA, and CBLA gave a brain-to-plasma ratio <1, which indicates that all other neutral phytocannabinoids would be expected to partition into the brain from the plasma. Values exceeding the % BTB threshold (Figure ) and brain-to-plasma ratios exceeding 1 (Table S12) are consistent with the well-known psychoactive properties of phytocannabinoids (e.g., Δ9-THC) and indicate that many phytocannabinoids have therapeutic potential for the treatment of central nervous system diseases. ,

6.

6

Open in a new tab

Percent brain-tissue binding (% BTB) of neutral and acidic phytocannabinoids with red-line threshold (% BTB = 99%) associated with bind to brain tissue after molecules have crossed the blood–brain barrier. Groups of isomers are denoted by common symbols *, ∧, +, and #.

Log k p skin is an estimation of a compound’s ability to permeate through the skin (Table S12). Neutral phytocannabinoids had log k p skin values ranging from −1.50 to −0.47 while acidic phytocannabinoids had a range of −3.23 to −2.65. A negative log k p skin value suggests that a phytocannabinoid would be a poor candidate for transdermal delivery. Because of the poor skin permeability of phytocannabinoids, formulations of phytocannabinoid products intended for dermal application include chemical penetration enhancers, such as oleic acid or ethanol. ,,

Conclusions

This study presents the largest collection of biomimetic partition coefficients for neutral and acidic phytocannabinoids. Biomimetic chromatography offered a high-throughput method to measure partition coefficients for phytocannabinoids that were then used to calculate certain pharmacokinetic parameters. Phytocannabinoids exhibited high lipophilicity (CHI log D and log k IAM) and high affinity for HSA (log k HSA). Phytocannabinoids are expected to be highly retained in the lungs (% LTB) and the brain (% BTB) and to potentially inhibit hERG but are not expected to partition into skin (log k p skin). Using biomimetic information, drugs that incorporate synthetic cannabinoids can be better designed since synthetic cannabinoids have molecular targets similar to those of phytocannabinoids and endogenous endocannabinoids. Moreover, biomimetic data can significantly enhance our understanding of phytocannabinoids bioactivity by complementing the results obtained from in vitro and in vivo biological activity and toxicity screening through the use of the pharmacokinetic models.

Supplementary Material

ao5c10320_si_001.pdf (324KB, pdf)
ao5c10320_si_002.pdf (485.5KB, pdf)

Acknowledgments

This work is supported, in part, by the Oregon Agricultural Experiment Station with funding from the Hatch Act capacity funding program, award numbers NI25HFPXXXXXG022 and/or NI25HMFPXXXXG029, from the USDA National Institute of Food and Agriculture and by the United States Department of Agriculture (USDA) Agricultural Research Service (ARS) Non-Assistance Cooperative Agreement (NACA) Project #2072-2100-054-00-D.

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsomega.5c10320.

  • Additional information on the biomimetic chromatography preparation and calibration as well as high-resolution mass spectrometer details, results on acid–base characterization, data set obtained from biomimetic chromatography and phytocannabinoid parameter values (PDF)

  • Biomimetic TOC (PDF)

J.K.M.: data curation, formal analysis, investigation, writingoriginal draft, review and editing; D.J.M.: data curation, formal analyses, investigation, writingreview and editing; B.D.: data curation, investigation, writingreview and editing; K.V.: formal analyses, resources, validation; L.T.: writingreview and editing; R.L.T.: supervision, writingreview and editing; J.A.F.: conceptualization, funding acquisition, project administration, supervision, writingreview and editing.

The authors declare no competing financial interest.

References

  1. Sampson P. B.. Phytocannabinoid pharmacology: medicinal properties of Cannabis sativa constituents aside from the “Big Two”. J. Nat. Prod. 2021;84(1):142–160. doi: 10.1021/acs.jnatprod.0c00965. [DOI] [PubMed] [Google Scholar]
  2. Amar M. B.. Cannabinoids in medicine: A review of their therapeutic potential. J. Ethnopharmacol. 2006;105(1–2):1–25. doi: 10.1016/j.jep.2006.02.001. [DOI] [PubMed] [Google Scholar]
  3. Dickson K., Janasie C., Willett K. L.. Cannabinoid conundrum: a study of marijuana and hemp legality in the united states1. Arizona J. Environ. Law Policy. 2019;10(20):132. [PMC free article] [PubMed] [Google Scholar]
  4. Pagano C., Navarra G., Coppola L., Avilia G., Bifulco M., Laezza C.. Cannabinoids: therapeutic use in clinical practice. Int. J. Mol. Sci. 2022;23(6):3344. doi: 10.3390/ijms23063344. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Abioye A., Ayodele O., Marinkovic A., Patidar R., Akinwekomi A., Sanyaolu A.. Δ9-Tetrahydrocannabivarin (THCV): a commentary on potential therapeutic benefit for the management of obesity and diabetes. J. Cannabis Res. 2020;2:1–6. doi: 10.1186/s42238-020-0016-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Patti F., Chisari C. G., Solaro C., Benedetti M. D., Berra E., Bianco A., Bruno Bossio R., Buttari F., Castelli L., Cavalla P.. et al. Effects of THC/CBD oromucosal spray on spasticity-related symptoms in people with multiple sclerosis: results from a retrospective multicenter study. Neurol. Sci. 2020;41:2905–2913. doi: 10.1007/s10072-020-04413-6. [DOI] [PubMed] [Google Scholar]
  7. Lah T. T., Novak M., Pena Almidon M. A., Marinelli O., Žvar Baškovič B., Majc B., Mlinar M., Bošnjak R., Breznik B., Zomer R., Nabissi M.. Cannabigerol is a potential therapeutic agent in a novel combined therapy for glioblastoma. Cells. 2021;10(2):340. doi: 10.3390/cells10020340. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Cherney J. H., Small E.. Industrial hemp in North America: production, politics and potential. Agronomy. 2016;6(4):58. doi: 10.3390/agronomy6040058. [DOI] [Google Scholar]
  9. Abernethy, A. Hemp Production and the 2018 Farm Bill; US Food and Drug Administration, 2019. [Google Scholar]
  10. Hong M., Kim J.-H., Han J.-H., Ryu B.-R., Lim Y.-S., Lim J.-D., Park S.-H., Kim C.-H., Lee S.-U., Kwon T.-H.. In Vitro and In Vivo Anti-Inflammatory Potential of Cannabichromene Isolated from Hemp. Plants. 2023;12(23):3966. doi: 10.3390/plants12233966. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Valko K., Du C. M., Bevan C., Reynolds D. P., Abraham M. H.. Rapid method for the estimation of octanol/water partition coefficient (log Poct) from gradient RP-HPLC retention and a hydrogen bond acidity term (∑α2H) Curr. Med. Chem. 2001;8(9):1137–1146. doi: 10.2174/0929867013372643. [DOI] [PubMed] [Google Scholar]
  12. Valkó K.. Application of high-performance liquid chromatography based measurements of lipophilicity to model biological distribution. J. Chromatogr. A. 2004;1037(1–2):299–310. doi: 10.1016/j.chroma.2003.10.084. [DOI] [PubMed] [Google Scholar]
  13. Valkó, K. Physicochemical and Biomimetic Properties in Drug Discovery: Chromatographic Techniques for Lead Optimization; John Wiley & Sons, 2013. [Google Scholar]
  14. Valkó K. L.. Lipophilicity and biomimetic properties measured by HPLC to support drug discovery. J. Pharm. Biomed. Anal. 2016;130:35–54. doi: 10.1016/j.jpba.2016.04.009. [DOI] [PubMed] [Google Scholar]
  15. Camurri G., Zaramella A.. High-throughput liquid chromatography/mass spectrometry method for the determination of the chromatographic hydrophobicity index. Anal. Chem. 2001;73(15):3716–3722. doi: 10.1021/ac001388j. [DOI] [PubMed] [Google Scholar]
  16. Valko K. L., Rava S., Bunally S., Anderson S.. Revisiting the application of Immobilized Artificial Membrane (IAM) chromatography to estimate in vivo distribution properties of drug discovery compounds based on the model of marketed drugs. ADMET DMPK. 2020;8(1):78–97. doi: 10.5599/admet.757. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Valko K., Nunhuck S., Bevan C., Abraham M. H., Reynolds D. P.. Fast gradient HPLC method to determine compounds binding to human serum albumin. Relationships with octanol/water and immobilized artificial membrane lipophilicity. J. Pharm. Sci. 2003;92(11):2236–2248. doi: 10.1002/jps.10494. [DOI] [PubMed] [Google Scholar]
  18. Hollósy F., Valkó K., Hersey A., Nunhuck S., Kéri G., Bevan C.. Estimation of volume of distribution in humans from high throughput HPLC-based measurements of human serum albumin binding and immobilized artificial membrane partitioning. J. Med. Chem. 2006;49(24):6958–6971. doi: 10.1021/jm050957i. [DOI] [PubMed] [Google Scholar]
  19. Stergiopoulos C., Tsopelas F., Valko K.. Prediction of hERG inhibition of drug discovery compounds using biomimetic HPLC measurements. ADMET DMPK. 2018;9(3):191–207. doi: 10.5599/admet.995. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Stergiopoulos C., Tsopelas F., Ochsenkühn-Petropoulou M., Valko K.. Predicting the acute aquatic toxicity of organic UV filters used in cosmetic formulations. ADMET DMPK. 2024;12(5):781–796. doi: 10.5599/admet.2364. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Ciura K., Jarzyńska K., Greber K. E., Sawicki W., Jagiello K., Wielgomas B., Kierkowicz M.. Is the potential for bioaccumulation of cannabinoids underestimated? Insights from biomimetic chromatography in cannabinoid exposure evaluation. Food Chem. Toxicol. 2025;197:115252. doi: 10.1016/j.fct.2025.115252. [DOI] [PubMed] [Google Scholar]
  22. Bteich M.. An overview of albumin and alpha-1-acid glycoprotein main characteristics: highlighting the roles of amino acids in binding kinetics and molecular interactions. Heliyon. 2019;5(11):e02879. doi: 10.1016/j.heliyon.2019.e02879. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Smith S. A., Waters N. J.. Pharmacokinetic and pharmacodynamic considerations for drugs binding to alpha-1-acid glycoprotein. Pharm. Res. 2019;36(2):30. doi: 10.1007/s11095-018-2551-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Muensterman D. J., Valko K., Chen R., Ng C., Droz B., Vyas S., Benotti M., Field J.. Biomimetic Chromatography: A Novel Approach for Measuring Phospholipid Membrane–Water and Protein–Water Partition Coefficients for Target and Suspect PFAS. Environ. Sci. Technol. 2025;60(1):1207–1217. doi: 10.1021/acs.est.5c11716. [DOI] [PubMed] [Google Scholar]
  25. Anderson L. L., Low I. K., Banister S. D., McGregor I. S., Arnold J. C.. Pharmacokinetics of phytocannabinoid acids and anticonvulsant effect of cannabidiolic acid in a mouse model of Dravet syndrome. J. Nat. Prod. 2019;82(11):3047–3055. doi: 10.1021/acs.jnatprod.9b00600. [DOI] [PubMed] [Google Scholar]
  26. Valko K. L., Zhang T.. Biomimetic properties and estimated in vivo distribution of chloroquine and hydroxy-chloroquine enantiomers. ADMET DMPK. 2018;9(2):151–165. doi: 10.5599/admet.929. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Licitra R., Marchese M., Naef V., Ogi A., Martinelli M., Kiferle C., Fronte B., Santorelli F. M.. A review on the bioactivity of cannabinoids on zebrafish models: Emphasis on neurodevelopment. Biomedicines. 2022;10(8):1820. doi: 10.3390/biomedicines10081820. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Monfort A., Ferreira E., Leclair G., Lodygensky G. A.. Pharmacokinetics of cannabis and its derivatives in animals and humans during pregnancy and breastfeeding. Front. Pharmacol. 2022;13:919630. doi: 10.3389/fphar.2022.919630. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Curry S., Brick P., Franks N. P.. Fatty acid binding to human serum albumin: new insights from crystallographic studies. Biochim. Biophys. Acta, Mol. Cell Biol. Lipids. 1999;1441(2–3):131–140. doi: 10.1016/S1388-1981(99)00148-1. [DOI] [PubMed] [Google Scholar]
  30. Shahzad A.. Hemp fiber and its composites – a review. J. Compos. Mater. 2012;46(8):973–986. doi: 10.1177/0021998311413623. [DOI] [Google Scholar]
  31. Della Rocca G., Di Salvo A.. Hemp in veterinary medicine: from feed to drug. Front. Vet. Sci. 2020;7:558206. doi: 10.3389/fvets.2020.00387. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Zhu L., Pei W., DiCiano P., Brands B., Wickens C. M., Le Foll B., Kwong B., Parashar M., Sivananthan A., Mahadevan R.. Physiologically-based pharmacokinetic model for predicting blood and tissue concentrations. Comput. Chem. Eng. 2021;154:107461. doi: 10.1016/j.compchemeng.2021.107461. [DOI] [Google Scholar]
  33. Citti C., Braghiroli D., Vandelli M. A., Cannazza G.. Pharmaceutical and biomedical analysis of cannabinoids: A critical review. J. Pharm. Biomed. Anal. 2018;147:565–579. doi: 10.1016/j.jpba.2017.06.003. [DOI] [PubMed] [Google Scholar]
  34. Varshney A., Sen P., Ahmad E., Rehan M., Subbarao N., Khan R. H.. Ligand binding strategies of human serum albumin: how can the cargo be utilized? Chirality. 2010;22(1):77–87. doi: 10.1002/chir.20709. [DOI] [PubMed] [Google Scholar]
  35. Kaliszan R., Nasal A., Turowski M.. Binding site for basic drugs on α1-acid glycoprotein as revealed by chemometric analysis of biochromatographic data. Biomed. Chromatogr. 1995;9(5):211–215. doi: 10.1002/bmc.1130090504. [DOI] [PubMed] [Google Scholar]
  36. Fournier T., Medjoubi-N N., Porquet D.. Alpha-1-acid glycoprotein. Biochim. Biophys. Acta, Protein Struct. Mol. Enzymol. 2000;1482(1–2):157–171. doi: 10.1016/S0167-4838(00)00153-9. [DOI] [PubMed] [Google Scholar]
  37. Israili Z. H., Dayton P.. Human alpha-1-glycoprotein and its interactions with drugs. Drug Metab. Rev. 2001;33(2):161–235. doi: 10.1081/DMR-100104402. [DOI] [PubMed] [Google Scholar]
  38. Valko K., Chiarparin E., Nunhuck S., Montanari D.. In vitro measurement of drug efficiency index to aid early lead optimization. J. Pharm. Sci. 2012;101(11):4155–4169. doi: 10.1002/jps.23305. [DOI] [PubMed] [Google Scholar]
  39. Farrelly K. N., Wardell J. D., Marsden E., Scarfe M. L., Najdzionek P., Turna J., MacKillop J.. The impact of recreational cannabis legalization on cannabis use and associated outcomes: a systematic review. Subst. Abuse: Res. Treat. 2023;17:11782218231172054. doi: 10.1177/11782218231172054. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Schaefer N., Kettner M., Laschke M. W., Schlote J., Ewald A. H., Menger M. D., Maurer H. H., Schmidt P. H.. Distribution of synthetic cannabinoids JWH-210, RCS-4 and Δ 9-tetrahydrocannabinol after intravenous administration to pigs. Curr. Neuropharmacol. 2017;15(5):713–723. doi: 10.2174/1570159X15666161111114214. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Saenz S. R., Lewis R. J., Angier M. K., Wagner J. R.. Postmortem fluid and tissue concentrations of THC, 11-OH-THC and THC-COOH. J. Anal. Toxicol. 2017;41(6):508–516. doi: 10.1093/jat/bkx033. [DOI] [PubMed] [Google Scholar]
  42. Orvos P., Paszti B., Topal L., Gazdag P., Prorok J., Polyak A., Kiss T., Toth-Molnar E., Csupor-Loffler B., Bajtel A.. et al. The electrophysiological effect of cannabidiol on hERG current and in guinea-pig and rabbit cardiac preparations. Sci. Rep. 2020;10(1):16079. doi: 10.1038/s41598-020-73165-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Richards J. R., Blohm E., Toles K. A., Jarman A. F., Ely D. F., Elder J. W.. The association of cannabis use and cardiac dysrhythmias: a systematic review. Clin. Toxicol. 2020;58(9):861–869. doi: 10.1080/15563650.2020.1743847. [DOI] [PubMed] [Google Scholar]
  44. Croxford J. L.. Therapeutic potential of cannabinoids in CNS disease. CNS Drugs. 2003;17:179–202. doi: 10.2165/00023210-200317030-00004. [DOI] [PubMed] [Google Scholar]
  45. Zhang S.-S., Zhang N.-N., Guo T.-T., Sheen L.-Y., Ho C.-T., Bai N.-S.. The impact of phyto-and endo-cannabinoids on central nervous system diseases: A review. J. Tradit. Complementary Med. 2023;13(1):30–38. doi: 10.1016/j.jtcme.2022.10.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Valko K. L., Kindy M., Evans J., Ko D.. In vitro biomimetic HPLC and in vivo characterisation of GM6, an endogenous regulator peptide drug candidate for amyotrophic lateral sclerosis. ADMET DMPK. 2018;6(2):176–189. doi: 10.5599/admet.547. [DOI] [Google Scholar]
  47. Kirk R. D., Akanji T., Li H., Shen J., Allababidi S., Seeram N. P., Bertin M. J., Ma H.. Evaluations of skin permeability of cannabidiol and its topical formulations by skin membrane-based parallel artificial membrane permeability assay and Franz cell diffusion assay. Med. Cannabis Cannabinoids. 2022;5(1):129–137. doi: 10.1159/000526769. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Mahmoudinoodezh H., Telukutla S. R., Bhangu S. K., Bachari A., Cavalieri F., Mantri N.. The transdermal delivery of therapeutic cannabinoids. Pharmaceutics. 2022;14(2):438. doi: 10.3390/pharmaceutics14020438. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

ao5c10320_si_001.pdf (324KB, pdf)
ao5c10320_si_002.pdf (485.5KB, pdf)

Articles from ACS Omega are provided here courtesy of American Chemical Society

RESOURCES