Intralacrimal Sustained Delivery of Rapamycin Shows Therapeutic Effects without Systemic Toxicity in a Mouse Model of Autoimmune Dacryoadenitis Characteristic of Sjögren's Syndrome

Abstract

Sjögren's syndrome (SS) is an autoimmune disease associated with severe exocrinopathy, which is characterized by profound lymphocytic infiltration (dacryoadenitis) and loss of function of the tear-producing lacrimal glands (LGs). Systemic administration of Rapamycin (Rapa) significantly reduces LG inflammation in the male Nonobese Diabetic (NOD) model of SS-associated autoimmune dacryoadenitis. However, the systemic toxicity of this potent immunosuppressant limits its application. As an alternative, this paper reports an intra-LG delivery method using a depot formulation comprised of a thermoresponsive elastin-like polypeptide (ELP) and FKBP, the cognate receptor for Rapa (5FV). Depot formation was confirmed in excised whole LG using cleared tissue and observation by both laser-scanning confocal and lightsheet microscopy. The LG depot was evaluated for safety, efficacy, and intra-LG pharmacokinetics in the NOD mouse disease model. Intra-LG injection with the depot formulation (5FV) retained Rapa in the LG for a mean residence time (MRT) of 75.6 h compared to Rapa delivery complexed with a soluble carrier control (5FA), which had a MRT of 11.7 h in the LG. Compared to systemic delivery of Rapa every other day for 2 weeks (seven doses), a single intra-LG depot of Rapa representing 16-fold less total drug was sufficient to inhibit LG inflammation and improve tear production. This treatment modality further reduced markers of hyperglycemia and hyperlipidemia while showing no evidence of necrosis or fibrosis in the LG. This approach represents a potential new therapy for SS-related autoimmune dacryoadenitis, which may be adapted for local delivery at other sites of inflammation; furthermore, these findings reveal the utility of optical imaging for monitoring the disposition of locally administered therapeutics.

Figure 1.. Schematic illustration of intra-LG injection of 5FV-Rapa relative to subcutaneous administration of a soluble Rapa carrier, FAF.

A) Graphical illustration of depot formation by 5FV-Rapa but not 5FA-Rapa. B) Schematic illustration of the study evaluating the therapeutic efficacy of a single intra-LG injection of 5FV-Rapa (0.44 mg Rapa/kg × 1 dose) versus subcutaneous injection every other day of FAF-Rapa (1mg Rapa/kg × 7 doses), a systemic administration control, over 14 days in diseased male NOD mice. A control treatment was intra-LG with PBS on Day 1 of the two week study, also shown. Created with biorender.com.

Figure 2.. Biophysical characterization of 5FA and 5FV.

A) Coomassie blue staining of purified 5FA and 5FV resolved by SDS-PAGE. B) Optical density at 350 nm of 5FA was monitored as a function of temperature at different concentrations. C) Optical density at 350 nm of 5FV was monitored as a function of temperature at different concentrations. D) Phase transition temperature was plotted vs. concentration as a phase diagram for 5FA and 5FV. The 95% confidence interval around each best-fit line is indicated with dashed lines. E) Surface plasmon resonance (SPR) sensorgram of different concentrations of Rapa monitored on a surface with immobilized 5FA at 18°C. F) SPR sensorgram of indicated concentrations of Rapa monitored on a surface with immobilized 5FV at 18°C.

Figure 3.. Intra-LG injection does not promote tissue damage or fibrosis.

Male C57 mice were either used as controls (CON) or administered with 32 μl of 5FA or 5FV (366 μM) in total, 16 μl per side with 4 injections of 4 μl each. LG were obtained after 2 weeks. A) H&E staining of a representative LG section from each group (n=3 LGs from 3 different mice in each group, 2 sections per mouse). Scale bar: 200 μm. B) Trichrome staining of a representative LG from each group). Scale bar: 200 μm. C) Quantification of the percentage of the area stained blue with Trichrome. (n=3 LGs from 3 different mice in each group, 2 sections per LG) D) HSP47 immunofluorescence staining of LG representative of each group. White: F-actin, Blue: DAPI, Green: HSP47. Scale bar: 20 μm. E) Quantification of the number of HSP47+ cells in each picture (n=3 LGs from 3 different mice each group, 3 pictures from each LG). Data is presented as mean ± SD.

Figure 4.. Lightsheet imaging of Rhodamine-labeled 5FV-Rapa following intra-LG administration reveals formation of a depot.

Male C57 mice were administered with a single 5 μl injection of rhodamine-labeled 5FV-Rapa or rhodamine-labeled 5FA-Rapa (366 μM of ELP) and LG were collected 1 day after injection. A) A reconstructed 3D image of LG injected with rhodamine-labeled 5FA-Rapa. B) A cross-section of the 3D image in panel A showing three orthogonal planes. C) Serial sections of LG injected with rhodamine-labeled 5FA-Rapa. D) Images of each cross section in panel C. E) A reconstructed 3D image of LG injected with intra-LG rhodamine-5FV-Rapa. F) A cross-section of the 3D image in panel F showing three orthogonal planes at the site of injection. G) Serial sections of LG injected with rhodamine-labeled 5FV-Rapa. H) Images of each cross section in panel G. Magenta: Rhodamine-ELP, Cyan: nuclear stain, Scale bar: 500 μm. Asterisk: depot formed after intra-LG delivery of 5FV-Rapa. Arrows: 5FV-Rapa trapped beneath the capsule after intra-LG injection.

Figure 5.. Confocal fluorescence microscopy imaging of Rhodamine-labeled 5FV-Rapa injected intra-LG reveals a depot formation.

Male C57 mice were injected with a single intra-LG injection of 5 μl containing rhodamine-labeled 5FV-Rapa or rhodamine-labeled 5FA-Rapa (366 μM of ELP). LG were collected after 1 day. Serial sections were acquired in the cleared tissue at 15 μm intervals following following A) intra-LG injection with Rhodamine-labeled 5FA-Rapa and B) intra-LG injection with Rhodamine-labeled 5FV-Rapa. 5FV-Rapa is retained throughout the LG, especially in the interlobular (arrowhead) and interacinar (arrows) areas after 24 hrs. Red: Rhodamine-ELP, White: nuclear stain, Scale bar: 100 μm.

Figure 6.. The 5FV-Rapa depot is retained in the LG for significantly longer than its soluble control formulation (5FA-Rapa).

Male and female C57 mice aged 12–14 weeks old are injected with 32 μl of 5FA or 5FV (1.2 mM Rapa, 366 μM ELP) in total, 16 μl per side with 4 injections of 4 μl each in the LGs. Amount of Rapa in the LG at 30 min, 2 days, 7 days and 14 days after intra-LG injection of 5FA-Rapa or 5FV-Rapa is assessed by LC-MS (n=4 mice per group). *p<0.05, **p<0.01.

Figure 7.. Intra-LG 5FV-Rapa significantly enhances basal and stimulated tear secretion.

13 week (diseased) male NOD mice were treated with the dose regimen in Figure 1B. Tear production was evaluated 2 weeks after the initial treatments at the conclusion of the protocol. A) The differences in basal tear production as measured by phenol red thread tests for values post-treatment (mm) and pre-treatment (mm) were plotted. Each data point represents one individual eye (n=28 eyes from 14 mice per group). B) After treatments, tear production from both eyes after topical carbachol stimulation of both LG was plotted. Each data point represents one individual mouse (n=7 mice per group). *p<0.05, **p<0.01.

Figure 8.. Both 5FV-Rapa and FAF-Rapa reduced lymphocytic infiltration.

Representative H&E staining of NOD mouse LG after treatment as in Figure 1 with A) a single treatment with intra-LG PBS; B) a single treatment of intra-LG 5FV-Rapa (0.44 mg/kg with 0.22mg/kg to each gland.) C) subcutaneous FAF-Rapa injections (1mg/kg × 7doses). D) Quantification of the percentage of area of LG covered by lymphocytes after different treatments (n=14 LG from 14 different mice per group). **p<0.01***p<0.001.

Figure 9.. 5FV-Rapa has significantly less systemic toxicity than FAF-Rapa.

Body weight and serum chemistry were evaluated 2 weeks after the initial injections in 13 week male NOD mice at study endpoint as shown in Figure 1. A) Percentage of body weight (BW) change post-treatment compared with pre-treatment (n=14). B) Change in blood glucose post-treatment compared with pre-treatment(n=14). C) Serum cholesterol level after different treatments (n=6). D) Serum triglycerides levels after different treatments (n=6). **p<0.01***p<0.001.