Protein-protein interactions (PPIs) are critical therapeutic targets for cancer treatment , but despite 650,000 potential PPI targets, only one drug targeting a PPI is approved in the U.S. for clinical use . PPIs are often mediated by multiple residues spanning surface areas an order of magnitude larger than small molecules can bind [3,4]. Peptides, by contrast, are larger and can bind multiple residues simultaneously to disrupt PPIs with high affinity and specificity by mimicking one of the binding partners [5-7]. However while they mimic the sequence, they fail to mimic the secondary structure, which is often alpha-helical in the context of the whole protein . To overcome this, hydrocarbon “stapled” peptides were developed to maintain alpha-helical structure . Despite stapled peptides’ specificity of binding and stability of structure, they lack multiple delivery advantages of nanoparticle platforms including increased bioavailability and concentration of delivery , active cellular targeting [11-13], and co-delivery of synergistic therapeutics .
To deliver therapeutics to specific molecular targets, a delivery platform must account for many pharmacologic obstacles . I seek to develop a nanoparticle platform for disrupting specific intracellular PPIs in specific cell types by combining two complementary technologies: (i) hydrocarbon-stapled peptides for binding specificity and alpha-helical stability, and (ii) peptide amphiphile (PA) nanoparticles for increased bioavailability and concentration of delivery, co-delivery of synergistic therapeutics, and active cellular targeting. I am uniquely positioned to develop this platform with co-mentors, James LaBelle and Matthew Tirrell, as experts in stapled peptide therapeutics and targeted PA nanoparticles, respectively [16-21]. My goal is to combine a known p53-reactivating stapled peptide with PA nanoparticles decorated with single-chain variable fragment (scFv) targeting domains to overcome therapeutic resistance in diffuse large B cell lymphoma (DLBCL).
Aim 1: Structurally and biochemically characterize p53-specific stapled PA (p53-sPA) nanoparticles designed for disrupting intracellular p53:MDM2/MDMX interactions. Hypothesis: A p53-sPA nanoparticle will combine the therapeutic advantages of stapled peptides and PAs. Our group has previously shown that unstapled p53-PA (i) self-assembles into nanoparticles and (ii) drives therapeutic peptide uptake [22,23], while a p53 stapled peptide (iii) disrupts p53-inactivating PPIs and (iv) induces p53-dependent cell death [24,25]. I will synthesize p53-sPA and study its self-assembled nanoparticles using dynamic light scattering (DLS) and transmission electron microscopy (TEM). Then I will study its intracellular accumulation and p53-reactivation using a split-luciferase reporter cell line . Lastly, I will evaluate its ability to induce p53-dependent cell death in a broad panel of diffuse large B cell lymphoma (DLBCL) cell lines with both wildtype and mutant p53.
Aim 2: Actively deliver p53:MDM2/MDMX PPI-disrupting nanoparticles specifically to DLBCL using scFvs specific for B-cell surface antigen CD19. Hypothesis: p53-sPA nanoparticle-scFv conjugates will target DLBCL B cells with specificity and trigger endocytosis to eventually disrupt p53 sequestration within targeted cells. Targeted nanoparticles have been used to deliver conventional chemotherapeutics, but not stapled peptides, to specific cells. B cell lymphomatous cells express surface CD19, and liposomes decorated with anti-CD19 scFvs have been used to selectively target B cells and trigger nanoparticle endocytosis [27,28]. Our group has functionalized PA nanoparticles with targeting peptides [17-20,29], and I have designed an anti-CD19 scFv to do the same to B cells. I aim to functionalize our nanoparticles with scFvs and monitor their targeting specificity using fluorophore-conjugation, flow cytometry, and super-resolution confocal microscopy.
Aim 3: Overcome therapeutic resistance by disrupting synergistic PPIs using a single, mixed PA nanoparticle loaded with BIM-BH3 and SAH-p53 bioactive peptides. Hypothesis: Using combination PA nanoparticles to target non-redundant and synergistic PPIs overcomes adaptive resistance. The pathways through which BIM-PA and p53-sPA induce cell death are non-redundant [30-34], and I aim to test for synergism in chemotherapeutically resistant DLBCL cells. It has recently been shown that delivering two drugs in one nanoparticle overcomes adaptive resistance better than in separate nanoparticles , so I will compare delivery of BIM-PA and p53-sPA in mixed and unmixed PA nanoparticles targeting DLBCL.
While stapled peptides have been optimized to disrupt PPIs ex vivo, these proposed studies develop a delivery platform to assist therapeutic peptide delivery to their molecular targets. The results of these studies will establish a targeted delivery nanoparticle to disrupt specific PPIs in specific cell types, thus laying the groundwork for targeting previously “undruggable” PPIs with unprecedented specificity.
Protein-Protein Interactions: An Evasive Linchpin of Cancer Biology: Protein-protein interactions (PPIs) are crucial mediators of cellular physiology and cancer biology. There are an estimated 650,000 PPIs in the human interactome , the dysregulation of which leads to diseases such as cancer . Such PPIs can be drugged by blocking the binding surface of one of the partners, and multiple compounds targeting aberrant PPIs have entered clinical trials. Despite the promise of these targets, the U.S. Food and Drug Administration (FDA) has approved only one clinical therapeutic to target a PPI: Venetoclax .
PPIs have evaded biopharmaceutical targeting for multiple reasons. One reason is simply the size of the interaction sites. Most PPIs are mediated by multiple residues spanning surface areas of 1500-3000Å2, an order of magnitude larger than small molecules’ binding areas [3,4], although certain PPIs relying on a single “hot spot” have been successfully targeted [36,37]. Peptides, by contrast, can span large surface areas to simultaneously bind greater numbers of residues with higher specificity [5,6,38], and synthetic peptides can be designed to mimic the native sequences of alpha-helical segments that are known to be critical for many PPIs .
While peptides are a promising tool for the PPI target, important barriers remain for in vivo delivery. First, circulating peptides have short plasma half-lives due to clearance and hydrolysis, and they can fail to reach the target cell in an efficacious dosage. Second, peptides that do reach the target cell generally fail to penetrate the cell membrane unless they are highly cationic and/or amphiphilic, and therefore most peptide therapeutics entering clinical trials target extracellular targets . Finally, if they do reach their intended intracellular molecular target, they often fail to bind due to lack of alpha-helical secondary structure inherent to the protein they are mimicking . Although unstructured peptides can sometimes bind via induced fit , peptide binding typically relies on a stable alpha-helical structure. Delivery platforms have overcome subsets of these barriers individually, but simultaneously overcoming all obstacles will require a unique combination of chemistry, physics, and biology.
A Molecular Engineering Moonshot: Man reached the moon using a modular vehicle to overcome various obstacles at differing length scales; a comprehensive drug delivery platform should do the same. While peptide therapeutics show great promise at the molecular level, their major downfalls lie in their delivery. A recent article in Nature Biotechnology highlighted this urgent need: “Site-specific delivery of therapeutics will remain a distant reality unless nanocarrier design takes into account the majority, if not all, of the biological barriers that a particle encounters upon intravenous administration” . For example, a therapeutic peptide targeting an intracellular PPI must first be long-lived in circulation, reach the target cell, enter the target cell, and bind its molecular target intact.
I have designed a modular nanoparticle platform for delivering therapeutic peptides to disrupt specific PPIs in specific cell types. To do this, I combined two molecular engineering approaches used by my co-mentors. One of my co-mentors, James LaBelle, is an MD-PhD hematologist-oncologist and expert in stapled peptide modulation of oncogenic PPIs. My other co-mentor, Matthew Tirrell, is a world expert in peptide amphiphile (PA) nanoparticles that deliver therapeutic peptides to precise targets. Their co-mentorship has positioned me to understand this biological problem, develop chemical tools to address it, and test them biologically. The chemical modifications used for PAs and stapled peptides modify different parts of the peptide, as shown in Fig. 1, yet they impart non-redundant benefits for therapeutic peptide delivery. By combining stapled peptides and PAs, I can account for each step of a therapeutic peptide’s journey from administration to molecular target (Fig. 2). I have begun developing this platform (Fig. 2A-D) and now aim to expand its utility (Fig. 2E-G).
Stapled peptides are synthesized by adding a hydrocarbon “staple” across helical turns of a peptide to physically lock it in an alpha-helical secondary structure [9,40], as shown in Fig. 1A. This more accurately mimics the native protein to bind and modulate PPIs and expands the collection of PPIs targetable by therapeutic peptides (Fig. 2D and 2E). “Stapling” is done by incorporating into the peptide sequence non-natural amino acids with olefin-tether side chains then closing the “staple” using ruthenium-catalyzed ring closing metathesis (RCM). In addition to improved target binding, stapling also imparts protease resistance . However, cell penetration of stapled peptides is sequence-dependent and requires optimization .
PA nanoparticles overcome another subset of obstacles facing therapeutic peptides (Fig. 2A,B,F,G), and they are formed by conjugating a hydrophobic “tail” domain to one end of the peptide, which drives hydrophobic self-assembly of micellar nanoparticles as depicted in Fig. 1B. As opposed to other types of nanoparticles, their mass is almost entirely composed of peptide. PA nanoparticles are of optimal size (i.e. 20-150nm) to avoid clearance by the liver, kidney, and spleen , and the structure prevents hydrolysis and proteolysis during circulation (Fig. 2A). Once the nanoparticle reaches the target cell, its non-covalent assembly allows monomers to leave the nanoparticle and enter the cell (Fig. 2B) as the amphiphilic tail facilitates peptide internalization [22,23,43]. Moreover, targeting moieties can be conjugated to these nanoparticles (Fig. 2F) for active targeting to specific cell types [17-20,29], and multiple PAs can be self-assembled into one nanoparticle for combination therapy (Fig. 2G).
Targeting BIM – Chief Regulator of B-Cell Apoptosis (Preliminary Data): As a first step for targeting intracellular PPIs using PAs, we used an unstructured yet bioactive peptide mimicking the BH3 death domain of BIM, a pro-apoptotic member of the BCL-2 family of proteins and central regulator of B-cell ontogeny and lymphomagenesis [44-48]. The BCL-2 family of proteins is a network of PPIs that integrates pro- and anti-apoptotic signals to determine cell fate by either blocking or initiating apoptosis at the level of the mitochondrion [49-52]. These proteins share four BCL-2 homology domains (BH1-BH4), of which the BH3 domain mediates most functional PPIs in the network. BH3-only protein BIM is one of the most potent pro-apoptotic mediators in the family as it directly activates pro-apoptotic effectors while simultaneously inactivating anti-apoptotic suppressors. Therapeutic peptides mimicking the BIM BH3 domain potently induce B-cell death. While most peptides require alpha-helical stabilization to tightly bind their target proteins, the BIM BH3 domain is an exception, binding with nanomolar affinity with or without alpha-helical stabilization [44,51,53,54].
The unstructured BIM BH3 peptide induces cancer cell death, but only if the cell membrane is artificially permeabilized to facilitate its uptake. A stapled BIM BH3 peptide has been used to make the peptide sufficiently amphiphilic to pass the cell membrane , but as a general peptide delivery strategy, we instead aimed to use an amphiphilic tail to ferry the original, unstructured peptide into the cell (Fig. 2B) and then release it as an unmodified peptide once internalized (Fig. 2C). To do this, we first synthesized the unstructured BH3 peptide sequence (Fig. 3A, “BH3”) and added C-terminal lysines (K) to enhance water solubility and provide amine functional groups for tail coupling (Fig. 3A, “BIMAK”). We then coupled the C-terminal lysine side-chain to an amphiphilic tail domain with a PEG2000 spacer and a DSPE lipidic tail to create a self-assembling BIM-PA (Fig. 3A, “BIMAKPA2”). Our preliminary data showed PAs being internalized, at least in part, via endocytosis, so we sought to free the peptide from the tail after endocytosis. As endosomes mature, they activate enzymes called cathepsins, which preferably cleave specific amino acid sequences and can be harnessed to release therapeutic payloads after endocytosis [55,56]. We therefore synthesized a cathepsin-cleavable BIM-PA by adding a cathepsin cleavage sequence between the peptide and C-terminal lysines (Fig. 3A, “BIMAcathKPA2”).
We first tested these compounds’ abilities to bind their target proteins (e.g. BCL-XL and MCL-1) in vitro using fluorescence polarization assays (FPA). As expected, the unmodified BH3 peptide bound all its targets with nanomolar affinity (data not shown). Adding the amphiphilic tail slightly but significantly inhibited that binding, but incubating the cleavable BIM-PA (“BIMAcathKPA2”) with Cathepsin B released the tail and restored the native affinity of the peptide, as demonstrated by the leftward shifts in Fig. 3B.
We next tested the ability of the cleavable amphiphilic tail to enhance cellular internalization and subsequent bioactivity (Fig. 2B-2D). We added an N-terminal fluorophore to each of the compounds and tracked them with live cell confocal microscopy for two hours. We observed no uptake of the peptide alone (Fig. 3C, “BIMAK”). Adding an amphiphilic tail enhanced cellular uptake (Fig. 3C, “BIMAKPA2”), but the cleavable amphiphilic tail enhanced uptake and allowed the peptide to co-localize with mitochondria, the site of its BCL-2 family targets (Fig 3C, “BIMAcathKPA2”, indicated by white arrow). The cleavable BIM-PA potently induced dose-dependent cell death (Fig. 3D) by upregulating canonical apoptosis mechanisms, such as caspase-3/7 (Fig. 3E), while the non-
cleavable BIM-PA was slower-acting and less potent.
These preliminary data demonstrate the value in targeting intracellular, oncogenic PPIs using PAs (Fig. 2A-D). While the BIM BH3 domain binds its targets with nanomolar affinity without alpha-helical stabilization, most peptides do not. I now aim to expand this platform’s collection of targetable PPIs by creating and testing a cleavable stapled PA.
Awakening p53 – A Sleeping Guardian in DLBCL: Tumor suppressor protein p53 is responsible for maintaining the genomic integrity of cells by initiating cell-cycle-arrest or apoptosis upon damage. In healthy cells, p53 remains bound by inactivating proteins in the cytoplasm, such as MDM2 and MDMX, until it is activated and released. Many cancers, despite having wildtype p53, upregulate inactivating proteins to bind and functionally eliminate the “sleeping guardian,” p53 . Diffuse large B cell lymphoma (DLBCL) is the most common lymphoid malignancy and most clinically relevant example in which nearly 80% of cases have inactivated yet wildtype p53 [57-62], making it a clinically relevant model for p53 reactivation. Stapled peptides have recently been developed to mimic p53 and potently block both negative regulators, MDM2 and MDMX [24,25,63,64], and PAs have independently been used to deliver an unstructured p53 peptides into cells . This existing literature and the dire need for clinical p53 reactivation in DLBCL make this a high-impact first target for stapled PAs (Fig. 2E).
Specific Aim #1: Structurally and biochemically characterize p53-specific stapled PA (p53-sPA) nanoparticles designed for disrupting intracellular p53:MDM2/MDMX interactions. Hypothesis: A p53-sPA nanoparticle will combine the therapeutic advantages of stapled peptides and PAs.
Preliminary Data: My goal is to expand the utility of this PPI-targeting platform by delivering a bioactive p53-sPA and demonstrate the same intracellular delivery benefits as shown with unstructured BIM-PA (Fig. 3). To accomplish this aim, I developed a stapled PA synthesis and purification strategy. First, I used Fmoc-based, solid-phase peptide synthesis (SPPS) techniques to synthesize peptides with non-natural amino acids used for stapling (Fig. 1A). Due to the bulky side chains on the non-natural amino acids, I used specialized reaction conditions as described in the stapled peptide synthesis literature to synthesize a stabilized-alpha-helical p53 (SAH-p53-8) stapled peptide [64-66], as depicted in Fig. 4A. I then purified them using High Performance Liquid Chromatography and Mass Spectrometry (HPLC-MS). By implementing stapled peptide synthesis techniques in our group, I enabled us to produce pure stapled peptides with yields near ~60%, comparable to published expected yields , and successfully automated it using our group’s PS3 peptide synthesizer.
To synthesize stapled PAs, I included a non-natural amino acid on the N-terminus of the peptide (L-propargylglycine, PRA) with an alkyne functional group side-chain. After purifying these alkyne-peptides via HPLC, I coupled them to an amphiphilic tail, DSPE-PEG2000-azide, using a Cu-catalyzed azide/alkyne cycloaddition (CuAAC) “click” reaction in water between the alkyne-peptide and azide-tail [67,68]. This reaction provided 100% yield in a few hours, and the resulting products can be purified via HPLC. The resulting amphiphiles have a polymeric molecular weight distribution (due to the PEG spacer) centered near 5000 Da, as depicted in a representative MALDI-TOF mass spectrum in Fig. 4B.
Approach: I am now synthesizing a panel of compounds with which to study p53 reactivation using p53-mimicking stapled PAs. I will compare a p53-mimicking (a) unstructured peptide, (b) stapled peptide, (c) unstructured peptide amphiphile, (d) stapled peptide amphiphile, and (e) cathepsin-cleavable stapled peptide amphiphile in their abilities to (i) self-assemble into nanoparticles, (ii) enter cells, (iii) disrupt p53-inactivating PPIs, and (iv) induce p53-dependent cell death.
First, I will physically characterize p53-sPA nanoparticles’ self-assembly properties using dynamic light scattering (DLS), transmission electron microscopy (TEM), and the critical micelle concentration (CMC). These experiments will show the size and shape of the resulting nanoparticles, which I expect to be spherical and within the optimal size range to avoid clearance by the liver, kidney, and spleen (i.e. 20-150nm) , as are other DSPE-PEG nanoparticles [21,69-72]. Meanwhile, the CMC measurement will give the concentration required for the monomers to remain self-assembled as nanoparticles. Next, I will biofunctionally characterize the compounds’ abilities to enter cells and bind p53’s inactivating proteins. First I will observe cellular uptake and localization by incubating MEFs with fluorophore-conjugated compounds for 2 hours and using live cell confocal microscopy. I will then test for intracellular p53 reactivation using a split luciferase reporter cell line (ReBiL) with recombinant p53 and MDM2 that luminesce when p53 is bound and inactivated by MDM2 . I will incubate ReBiL cells with a range of compound concentrations and measure luminescence every 30 minutes to quantify intracellular p53-MDM2 interactions in cells treated with compounds normalized to vehicle-treated cells. I will further test its ability to bind recombinant MDM2 and MDMX using FPA and western blots. Finally, I will characterize the p53-sPA nanoparticle’s ability to induce p53-dependent cell death. First, I will rule out the possibility of non-specific cell membrane disruption by the lipid tails by performing lactic acid dehydrogenase (LDH) release assays, a sensitive screening tool used to detect cytoplasmic LDH inadvertently released from the cells due to non-specific membrane disruption. Our preliminary data with BIM-PA using the same DSPE-PEG tail showed no LDH release, so I expect similar results here. After ruling out non-specific activity, I will test p53-sPA’s ability to induce p53-dependent cell death in a broad panel of DLBCL cell lines. I will measure this using XTT, CellTiter-Glo and caspase-activation measurement assays.
Anticipated Results and Challenges: Due to preliminary data with BIM-PA (Fig. 3) and successful synthesis of stapled PAs (Fig. 4), I anticipate few remaining experimental challenges for this aim. I expect p53-sPA nanoparticles to self-assemble into spherical micelles as any other DSPE-PEG PA [21,69-72]. I expect cleavable p53-sPA to deliver SAH-p53 into the cell, as it did for BIM BH3 (Fig. 3), and I expect delivered SAH-p53 to reactivate p53, as it has already been shown to do [25,66].
In pursuing this aim, I will acquire a broad skill-set spanning chemistry, physics, and biology, including techniques such as peptide synthesis, liquid-chromatography/mass-spectrometry (LCMS), organic and inorganic synthesis, nanoparticle characterization, cell culture, confocal microscopy, and cell death and proliferation assays. With this broad skill-set, I can plan multifaceted experiments I otherwise could not.
Specific Aim #2: Actively deliver p53:MDM2/MDMX PPI-disrupting nanoparticles specifically to DLBCL using scFvs specific for B-cell surface antigen CD19. Hypothesis: p53-sPA nanoparticle-scFv conjugates will target DLBCL B cells with specificity and trigger endocytosis to disrupt p53 sequestration within targeted cells.
Preliminary Data: I next aim to deliver our PPI-disrupting PAs to specific cell populations by conjugating B-cell-specific anti-CD19 single chain variable fragments (scFvs) to PA nanoparticles (Fig. 2F). Our group has previously delivered PA nanoparticles to other cell types using targeting peptides, and has demonstrated that this platform is amenable to the inclusion of targeting domains [17-20,29]. Anti-CD19 scFvs have previously been used to deliver nanoparticles to B-cells with high specificity [27,28], and they functionally mimic Type I antibodies to induce cellular uptake upon binding CD19 , which would further enhance intracellular delivery of our PA nanoparticles.
I first engineered the anti-CD19 scFv with a linker for nanoparticle conjugation, as depicted in Fig. 5. I cloned a gene construct with covalently-linked variable heavy and light chain domains from the B43 hybridoma anti-CD19 antibody . On the C-terminus, I added a 6xHIS tag for affinity purification and secondary antibody recognition, and I added a short, flexible linker with a terminal cysteine (Cys) residue for thiol-maleimide conjugation to our PA nanoparticles [27,28,74-76]. I cloned the gene sequence into a pET21b inducible-expression vector and expressed it in E. coli. After expression, I purified it using Ni-NTA affinity chromatography followed by fast protein liquid chromatography (FPLC). A western blot showing the purified protein is shown in Fig. 5C.
Approach: First, I will characterize the scFv’s ability to bind specifically to B cells using flow cytometry. Since scFvs retain the entire binding domain of a full IgG antibody, I will incubate cells with concentrations similar to those commonly used with flow cytometry antibodies. I will incubate CD19+ cells (i.e. immortalized DLBCL B-cells) and CD19- cells (e.g. immortalized Jurkat T-cells) with the scFvs for one hour, wash the cells, then use a secondary, fluorophore-conjugated anti-HIS antibody to label the scFvs bound to cell surface CD19. I will ensure CD19 specificity in another experiment by first pre-incubating the cells with a 10-fold excess of commercially available anti-CD19 IgG for one hour to block all potential scFv binding sites and thereby knock-down its B-cell targeting.
Next I will characterize the scFv’s ability to induce cellular uptake. To do this, I will use the C-terminal cysteine linker to conjugate a maleimide-fluorophore to the scFv. I will incubate the scFv-fluorophores with CD19+ DLBCL B cells and CD19- Jurkat T cells for one hour at 37C to facilitate energy-dependent uptake, wash them, and inspect them using confocal microscopy to look for intracellular fluorophore localization specifically in the CD19+ B cells. I will then perform the same experiment at 4C to block energy-dependent uptake and see if intracellular uptake is inhibited.
After characterizing the scFv, I will conjugate it to p53-sPA nanoparticles by incorporating commercially-available amphiphilic DSPE-PEG-maleimide monomers into p53-sPA nanoparticles. By including them at the time of self-assembly, the outer layer of the nanoparticle will be coated in maleimide functional groups, which react rapidly and specifically with thiols such as on the C-terminal cysteine of the scFv. Static light scattering (SLS) will then measure the molecular weights of the decorated and undecorated nanoparticles to understand the number of scFvs coupled per nanoparticle.
Finally, I will test the decorated nanoparticle’s therapeutic specificity for B cells. I will incubate CD19+ malignant DLBCL cells and CD19- Jurkat T cells with p53-sPA nanoparticles decorated with anti-CD19 scFvs and measure their internalization and induction of cell death, as described above, using flow cytometry, confocal microscopy, and cell death and proliferation assays. I will also measure cell killing in non-malignant CD19+ B cells compared to malignant DLBCL cells.
Anticipated Results and Challenges: I expect my purified anti-CD19 scFv to bind CD19+ B cells, couple to nanoparticles via the C-terminal cysteine, and deliver a payload into B cells all as previously described. Since p53-sPA induces programmed cell death via p53 activation, I expect non-malignant cells to be less susceptible to killing due to their lower levels of genotoxic stress. One potential challenge to this aim is if the DSPE-PEG-maleimide conjugated scFvs dissociate from the self-assembled nanoparticles and separate the targeting moiety and payload. While this will likely occur to some degree, our previous experience in targeting PA nanoparticles to specific cell types using targeting peptides suggests that the non-covalent supramolecular structure is at least stable enough in circulation to deliver a payload to the target [17-20,29]. If it somehow fails to do so, I will explore stabilization mechanisms for the nanoparticle by covalently cross-linking the core. In case of difficulties with p53-sPA in Aim 1, I can alternatively use BIM-PA for this aim as it targets another important PPI network in DLBCL.
In pursuing this aim, I will gain expertise in physical chemistry techniques, including SLS and DLS, protein engineering techniques, including molecular design, cloning, expression, and purification, and biological techniques including flow cytometry and confocal microscopy.
Specific Aim #3: Overcome therapeutic resistance by disrupting synergistic PPIs using a single, mixed PA nanoparticle loaded with BIM-BH3 and SAH-p53 bioactive peptides. Hypothesis: Using combination PA nanoparticles to target non-redundant and synergistic PPIs overcomes adaptive resistance.
Preliminary Data: An estimated 90% of patients with metastatic cancer fail treatment due to drug resistance , so finding new ways to combat resistance is of paramount clinical importance. Resistance arises either at the cellular level, by a single cell altering its molecular machinery after encountering a drug , or at the tissue level, due to tumor heterogeneity selecting for cell populations inherently resistant to the drug . A common strategy to combat the resistance problem involves dosing with two or more drugs simultaneously to target non-redundant pathways within the cell , though resistance can develop even then.
Our platform, however, can deliver multiple therapeutics simultaneously in a single nanoparticle (Fig. 1B), as opposed to in two separate nanoparticles, and this form of spatially constrained delivery was recently shown to prevent adaptive resistance in cancer . When dosing with two drugs simultaneously, some cells receive both, some receive one or the other, and some receive none. The cells receiving both die, and the cells receiving none remain unchanged. However, cells receiving one or the other can still acquire adaptive resistance. Instead, if every nanoparticle contains both drugs, every cell is either killed by the synergistic combination or completely naïve such that adaptive resistance has no chance to form before the next treatment.
One-third of DLBCL patients relapse with refractory and resistant disease , and finding ways to prevent that resistance is clinically important. Two PPI networks commonly associated with relapse are the two targeted by p53-sPA and BIM-PA [30-34]. Using these tools to target these pathways, I aim to overcome adaptive resistance in DLBCL by combining synergistic BIM-PAs and p53-sPAs into mixed nanoparticles targeting CD19.
Approach: First, I will test for synergism, additivity, or antagonism between unmixed p53-sPA and BIM-PA nanoparticles. I will test this using XTT cell death and viability assays to establish a therapeutic combination index (CI) in a broad panel of DLBCL cells lines, including both wildtype and mutant p53 strains with variable levels of chemotherapeutic resistance, as well as in non-malignant primary human B cells. After establishing their degree of synergism, I will perform the same experiment with pre-mixed p53- and BIM-nanoparticles to test for differences in synergism. Finally, I will perform the same cell death experiments using scFv-conjugated nanoparticles from Aim 2, this time co-loaded with both p53-sPA and BIM-PA, to determine if active targeting enhances their synergism. I expect this synergistic combination to activate cell death even in chemotherapeutically resistant strains. By performing these experiments in both malignant DLBCL cell lines as well as non-malignant primary human B cells, I will be able to establish a therapeutic window between killing healthy and mutant B cells. I expect that targeting two non-redundant programmed cell death pathways will potently kill mutant DLBCL cells but merely arrest the cell cycle of primary human B cells.
Anticipated Results and Challenges: I anticipate our PPI-disrupting therapeutics to have similar synergistic effects to those shown in the literature targeting these two pathways. In case of setbacks with p53-sPA in Aim 1, this aim can alternatively be done with BIM-PA nanoparticles loaded with nutlin-3, a small molecule blocking MDM2, in the hydrophobic core. While nutlin-3 only targets one of the p53 negative regulators targeted by SAH-p53, this combination should still be synergistic.
In pursuing this aim, I will learn to quantify pharmaceutical synergism, which will be invaluable in my career as a physician-scientist, both in developing new therapeutics and prescribing them to patients.
...(download the rest of the essay above)