(72a) Amphiphilic Macromolecules to Manage Atherosclerosis: Quantitative Structure Activity Relationships

Authors: 
Gu, L., Rutgers University
Poree, D., Rutgers University
Moghe, P. V., Rutgers University
Kholodovych, V., University of Medicine and Dentistry of New Jersey

Introduction: Atherosclerosis is the primary component of pathologies underlying cardiovascular disease (CVD), which is the leading cause of death in western countries. Atherosclerosis is characterized as an inflammatory disease involving macrophage scavenger receptor (SR) interactions with oxidized low-density lipoproteins (oxLDL), leading to plaque growth. Conventional therapies are often  plagued by off-target effects and the inability to address localized inflammation from preexisting lipid deposits. Therefore, it is necessary to develop a therapeutic that can address localized lipid uptake and inflammation. We have designed a new library of bioactive Amphiphilic Macromolecules (AM) to competitively occupy SRs and inhibit oxLDL uptake, foam cell formation and inflammatory signaling in human monocyte derived macrophages (MDM). Modulating the AM architecture resulted in varying degrees of oxLDL uptake and inflammation. Molecular modeling was used to study the correlation between AM structural features and their corresponding biological behavior.

Materials and Methods: AMs were synthesized by acylating the hydroxyl groups of sugar acids with dodecanoyl chloride in the presence of zinc chloride, then 5000 Da monohydroxyl-substituted PEG is coupled using 1,3-dicyclohexylcarbodiimide and 4-(dimethylamino)pyridinium p-toluene-sulfonate as catalyst before precipitation in CH2Cl2 and diethyl ether. In vitro efficacy of AM was investigated in MDM differentiated from peripheral blood mononuclear cells. MDM were treated with oxLDL (5µg/mL DiO labeled for uptake, 50µg/mL unlabeled for foam cell formation, 100µg/mL unlabeled for inflammatory response) in the presence of 10-4-10-6M AM for 6-24 h in RPMI media. OxLDL uptake and foam cell formation were determined by cell associated fluorescence (DiO and post treatment Oil Red O staining, respectively) via epifluorescent microscopy and quantified with ImageJ. The inflammatory response was determined using qRT-PCR (ΔΔCt), for which RNA was extracted and reverse transcribed to cDNA before cycling on a LightCycler480 with SYBR green and primers specific for ACTB, IL-1β, IL-6 and IL-8. ELISA’s were used to validate IL-8 cytokine secretion in post-treatment cell culture supernatant. For modeling studies, all AM were constructed and minimized in Molecular Operating Environment (MOE). Molecular descriptors (only molecular properties, atom types and chemical group counts) were calculated with Dragon software v.5.4- 2006. Partial least squares regression method implemented in MOE was used to model the experimental data.

Results and Discussion: Seven different AM structures were synthesized, varying the overall charge, hydrophobicity, sugar structure (linear vs. cyclic) and stereochemistry by using the sugar acids: mucic (1cM), neutral mucic (0cM), saccharic (1cS), tartaric (1cT), galacturonic(0cG), glucuronic (0cGL), di-mucic (1cMM). Cells treated with oxLDL alone displayed significant amounts of oxLDL uptake and lipid accumulation (foam cell phenotype) in addition to upregulation of IL-1β, IL-6 and IL-8. While some AMs were effective at reducing these endpoints, small changes in AM sugar structures resulted in large differences in potency. This effect was most drastic with the substitution of saccharic acid for mucic acid (the only difference between the structures being one stereocenter), in which 1cM was the most effective at reducing oxLDL uptake (80% inhibited relative to oxLDL alone), whereas 1cS increased uptake. Without an external pro-inflammatory stimulus, linear monosugar AM (1cM, 0cM, 1cS and 1cT) displayed no upregulation of  cytokines IL-1β, IL-6 and IL-8, whereas AM with cyclic sugars (0cG and 0cGL) or two mucic acids (1cMM) upregulated the tested genes to a similar level as oxLDL. Linear monosugar 1cM and 1cT were both effective at reducing oxLDL induced inflammatory gene upregulation. Levels of IL-8 cytokine secretion paralleled IL-8 gene upregulation for all tested conditions. Modeling studies found that a good correlation between predicted and observed results could be achieved for oxLDL uptake with only three descriptors (r2=0.97, q2=0.73) while only two descriptors are needed to model foam cell formation (r2=0.97, q2=0.86).

Conclusions:  Previous work with 1st generation AM structures qualitatively studied placement and quantity of charge, but lacked the sophistication to accurately identify key features. Using quantitative structure activity relationships will be able to guide future development of an optimal AM structure for mitigating adverse athero-relevant endpoints.