2262: Albumin-hitchhiking MMP13 siRNA Conjugate (siMMP13< (EG₁₈L)₂) for the Treatment of Rheumatic Disease

Juan Colazo¹, Ella Hoogenboezem¹, Fang Yu¹, Veeraj Shah¹, Justin Lo¹, Hongsik Cho², Karen Hasty², Leslie Crofford³ and Craig Duvall¹, ¹Department of Biomedical Engineering, Vanderbilt University, Nashville, Nashville, TN, ²Department of Orthopaedic Surgery and Biomedical Engineering, Memphis VA Medical Center, Memphis, TN, ³Vanderbilt University Medical Center, Nashville, TN

Background/Purpose: Osteoarthritis (OA) and rheumatoid arthritis (RA) decrease quality of life due to joint destruction, pain, and decreased function. Multiple joint osteoarthritis (MJOA) can occur in up to 50% of OA cases. The high incidence of MJOA motivates the development of systemic therapies. Although there are successful therapies for RA treatment, many patients do not benefit from them and they can cause systemic side effects. In OA/RA, matrix metalloproteinases (MMPs) drive joint degeneration. Small molecule MMP inhibitors were tested clinically, but were dose limited by toxicities caused by lack of MMP selectivity. Short-interfering RNA (siRNA) therapies can be designed to silence "undruggable targets" but are limited due to significant biological barriers such as extra-hepatic targeting. Albumin-based drug delivery is promising due to its high half-life, recycling ability, and its accumulation in diseased joints. Herein, we characterize a chemically stabilized MMP13 siRNA conjugate, siMMP13< (EG₁₈L)₂, which spontaneously binds albumin 'in situ'.

Methods: The siRNA was synthesized with alternating 2′-OMe and 2′-F ribosugar modifications to stabilize against endonucleases and phosphorothiate linkages on the backbone to block exonucleases. For albumin-binding abilities, a splitter phosphoramidite (< ) (to allow for diacyl addition i.e., 2 lipid tails) at the 5ʹ end of the sense strand was added followed by 3 repeats of a hexa-ethylene glycol (EG₆) phosphoramidite. Thus, 18 EGs were added prior to each eighteen-carbon (C18) acyl (2x, L₂ nomenclature). A Cy5 phosphoramidite was used in fluorescence tracking experiments. An overload-induced osteoarthritis mouse model (9N, 250 cycles, 3x/week) and the K/BxN serum transfer arthritis (STA) mouse model were used for OA and RA studies, respectively.

Results: RNA stabilization chemistries increased serum/synovial fluid stability, while maintaining potent MMP13 silencing. Albumin-binding dye Evan's Blue and fluorescent albumin accumulated more in OA knees/K/BxN hindpaws than healthy knees/hindpaws. siMMP13< (EG₁₈L)₂ showed preferential delivery to OA knees/K/BxN hindpaws over healthy knees/hindpaws and demonstrated greater pharmacokinetics than non-end-modified siRNA and cholesterol-conjugated siRNA (Figures 1 and 2). In the K/BxN model, siMMP13< (EG₁₈L)₂ reached/treated multiple joints including forepaws, knees, and hindpaws (Figure 3). Overall, siMMP13< (EG₁₈L)₂ treatment was safe, provided potent MMP13 mRNA/protein knockdown, provided joint pain benefits, reduced arthritis clinical score, reduced arthritis-related genes, provided joint protection, and performed better than or equal to gold standard clinical controls.

Conclusion: This novel albumin-hitchhiking siRNA shows promise as a platform technology that can be readily adapted for targeting of many rheumatic disease-driver genes. Furthermore, this siRNA technology is modular allowing potential for the engineering of many other chemistries such as targeting moieties, etc. In these studies, siMMP13< (EG₁₈L)₂ provided to be therapeutic in pre-clinical animal models of OA and RA, showing the importance of MMP13 in both degenerative and inflammatory arthritis.
Figure 1. Effective Joint Delivery of siMMP13< (EG18L)2 in a Mechanical Overload Model of Osteoarthritis. A) Mechanical Overload Mouse Model Protocol. B) Schematic of Albumin-hitchhiking Mechanism and Cy5 Fluorophore Placement. C) Ex-Vivo IVIS Fluorescence Imaging. D) Quantification of Ex-Vivo Imaging.

Figure 2. Effective Multi-joint Delivery of siMMP13< (EG18L)2 in the K/BxN Serum Transfer Arthritis Model with Presence in Important Joint Homeostatic Tissues. A) Ex-Vivo IVIS Fluorescence Imaging. B) Quantification of Ex-Vivo Imaging. C) Cryohistology of Hindpaw Tissues.

Figure 3. siMMP13< (EG18L)2 Treatment in the K/BxN Serum Transfer Arthritis Model. A) Hindpaw Clinical Scores. B) Ankle Width Change Measurements. *200 uL intra-peritoneal injection of K/BxN serum was performed on day 0, injections of therapeutics (all dosed at 10 mg/kg) were performed on day 2 after serum transfer. Clinical score: 0 = no swelling or erythema, 1 = slight swelling or erythema, 2 = moderate erythema and swelling in multiple digits or entire paw, 3 = pronounced erythema and swelling of an entire paw.
Disclosures: J. Colazo, None; E. Hoogenboezem, None; F. Yu, None; V. Shah, None; J. Lo, None; H. Cho, None; K. Hasty, None; L. Crofford, None; C. Duvall, None.