A high molecular weight hyaluronic acid biphasic dispersion as potential therapeutics for interstitial cystitis

Abstract Interstitial cystitis (IC) is a progressive bladder disease characterized by increased urothelial permeability, inflammation of the bladder with abdominal pain. While there is no consensus on the etiology of the disease, it was believed that restoring the barrier between urinary solutes and (GAG) urothelium would interrupt the progression of this disease. Currently, several treatment options include intravesical delivery of hyaluronic acid (HA) and/or chondroitin sulfate solutions, through a catheter to restore the urothelial barrier, but have shown limited success in preclinical, clinical trials. Herein we report for the first time successful engineering and characterization of biphasic system developed by combining cross‐linked hyaluronic acid and naïve HA solution to decrease inflammation and permeability in an in vitro model of interstitial cystitis. The cross‐linking of HA was performed by 4‐arm‐polyethyeleneamine chemistry. The HA formulations were tested for their viscoelastic properties and the effects on cell metabolism, inflammatory markers, and permeability. Our study demonstrates the therapeutic effects of different ratios of the biphasic system and reports their ability to increase the barrier effect by decreasing the permeability and alteration of cell metabolism with respect to relative controls. Restoring the barrier by using biphasic system of HA therapy may be a promising approach to IC.


| INTRODUCTION
Interstitial cystitis (IC) is a chronic disease of the bladder, characterized by bladder pain, increased urinary frequency and urgency, nocturia, and significant lifestyle problems. [1][2][3][4] The etiology of IC remains unclear, but various hypotheses including autoimmune processes, allergic reactions, chronic bacterial infections, toxic or dietary exposure, and psychosomatic factors were considered as key players. 5 The bladder includes a specialized epithelial layer composed of urothelial cells that form a tight, impermeable, protective barrier made up of glycosaminoglycan (GAG) and sulfated glycosaminoglycan (sGAG). [6][7][8][9] Disruption of this bladder wall leads to increased permeability of the bladder, which is a common clinical event in all patients with IC. 10 One of the most common IC treatment is intravesical sGAG or GAGs instillation into the bladder through a catheter. These treatments are targeted to augment a major protective element of the bladder, the urothelial lining and provide permissive healing environment for the urothelium. [11][12][13][14][15] Most of intravesical sGAG/GAG treatments are composed of hyaluronan (HA) solutions either alone or in combination with other GAGs (such as chondroitin sulfate). 11 HA instillations have been shown to decrease secretion of inflammatory cytokines and have been identified as a potential cost-effective treatment for IC patients. [16][17][18][19] Furthermore, in several nonblinded, nonplacebo controlled clinical trials; HA has demonstrated a positive patient response with decreased visual analogue scores for pain symptoms. [20][21][22] However, the clinical formulations use noncrosslinked HA systems that do not recover the urothelium's permeability, which is one of the main clinical symptoms of IC.
The cross-linking chemistry of HA and modified HA derivatives are well described in the literature. 23 It is well accepted that crosslinking of HA can increase chain length and can decrease macromolecular diffusion in the surrounding. 5 HA is recognized as suitable for carrier for cells, [24][25][26][27][28][29] treatment of osteoarthritis 30 and in colon targeted drug delivery applications 31 but to the best of our knowledge few studies report on the effect of cross-linked HA on urothelium specific to IC. HA is degraded mainly by the intestinal bacteria hyaluronidases [32][33][34][35] and the absence of the bladder enzyme hyaluronidases [35][36][37][38][39][40][41] it can be expected that long-chain cross-linked HA delivered locally via the catheter will remain on the surface of the bladder lumen for prolonged time with additive therapeutic benefits.
Considering this, the manuscript describes an in vitro model of urothelium barrier damage in order to examine the effects of native high molecular weight HA solution (commercial), cross-linked HA solution (cHA) and the combination of biphasic solution (native HA solution and cHA). HA cross-linking was carried out as per our reported GMP translatable methodology for the synthesis of cHA. 42 The crosslinking was performed with 4-arm-polyethyeleneamine (PEG) linker using 4-(4,6-Dimethoxy-1,3,5-triazin-2-yl)-4-methylmorpholinium chloride (DMTM) chemistry. The developed HA solutions were subjected to rheological, morphological, cellular level inflammatory cytokine, and GAG assessments. Finally, the HA, cHA, and the biphasic system were studied for concentration dependent effects of the different ratios of HA, cHA, and the biphasic system on the inflammatory cytokines and permeability in the in vitro model to examine their effect on inflammatory cytokines and trans-epithelial permeability using transwell assay units.

| Preparation of HA solutions
Different concentrations of naïve HA (1, 3, 9, 15 mg/ml) were prepared by dissolving naïve HA in PBS buffer (pH 6.0) and stirring overnight on a magnetic stirrer at 500 rpm, 25 C. and then freeze dried at −80 C to obtain the cross-linked HA.

| Synthesis of cHA
2.4 | Determining degree of cross-linking of cHA NMR studies and carbazole assay were performed to determine the degree of cross-linking of cHA. In NMR, the peak integrals were compared for quantitative estimation. Carbazole assay was carried out following previously reported procedures using samples with and without salts. 42,47,48 All the reagents were freshly prepared and no commercial kits were used.

| Preparation of biphasic dispersion of cHA in HA solution
The cHA and HA ratios were formulated 24-48 hrs before each inde-

Inducing inflammation by H 2 O 2 treatment
Cell monolayers in 48 well plates were chemically stripped and inflamed using hydrogen peroxide (1% H 2 O 2 in basal media) for 1 hr before HA intervention.

Inducing inflammation by protamine sulfate treatment
Cell monolayers in 48 well plates were chemically stripped using protamine sulfate (PS) (10 mg/ml) for 1 hr before HA was added.

Inducing inflammation by TNFα treatment
Cell monolayers in 48 well plates were inflamed using TNFα (100 ng/ml) for 1 hr before HA was added.

Basal cell conditions
Cell monolayers were cultured in 48 well plates and media was replaced with basal media.

HA intervention on HTB-2 cells
Cells were washed with HBSS. Control (Basal media) or HA solutions (1, 3, 9, 15 mg/ml) were added to wells for 2 hrs, and then washed with HBSS and replaced with basal media for 24 hours. Cell supernatants were removed and stored at −20 C for cytokines analysis.

Effect of HA intervention on sGAGs secretion
The Blyscan™ GAG assays (Bicolour, UK) were performed as per the manufacturer's instructions. The standard range was from 1.0 to 5 μg/ml.

T84 cells treatment with protamine sulphate and inflammation
Cell monolayers in 48 well plates were chemically stripped and inflamed using protamine sulphate (10 mg/ml) and TNFα (100 ng/ml) mixture for 1 hr before HA intervention.
HA intervention on T84 cell-transwells for permeability testing

| RESULTS
Understanding of viscoelastic behavior is critical as the HA-based system will eventually be administered directly into the bladder via

| DISCUSSION
Currently available intravesical sGAG/GAG treatments are available in a range of concentrations, 0.8, 1.6, and 2.4%. 49,50 There is sufficient evidence to show that the effect of viscosity correlates with increased HA concentration (Figure 1). When examining the effect of HA concentration in a urothelial cell monolayer without a consensus on the exact conditions of mimicking IC pathogenesis can be challenging. To inflammatory factor that has been found to be independently correlated with C-reactive protein and IL-8 in the serum of IC patients. 56 MCP-1 has been reported at elevated levels in preclinical models, which replicate the disease symptoms. The urine of IC patients has also been reported to contain higher levels of sGAG along with IL-6, IL-8, and MCP-1. 16  Cross-linked HA alone at 1 mg/ml HA concentration in PS treated cells increased MCP-1 and IL-6, while cross-linked HA alone at 3 mg/ml HA concentration increased MCP-1 levels (Figures 4 and 5).  Figure 6). TEER remained low in the cells, while 2 hrs of treatment with a 1:1 ratio of cHA:HA had an increased barrier effect as observed by the significant decrease in Papp. A physiochemical barrier effect from the increased chain length in the 1:1 ratio of cHA:HA leads to a decrease in Papp without an alteration in the expression in TEER. In Figure 7 the overall effects of the different ratios and concentrations effects on the key indicators of IC is discussed. It has been observed that the cross-linking of HA decreases the apparent permeability, without altering the disease conditions at a 1:1 ratio of cross-linked HA to naïve HA at both 1 and 3 mg/mlL HA concentrations.

| CONCLUSIONS
We have successfully demonstrated the engineering of a biphasic system developed by combining cHA and naive HA solution to reduce inflammation and permeability in the in vitro interstitial cystitis model.
In our study, we observed that increasing the concentration of naïve Idoia Tolosa from CIDETEC is gratefully thanked for her help in the synthesis of cHA.

CONFLICT OF INTEREST
The authors declare no potential conflict of interest.

AUTHOR CONTRIBUTIONS
The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript.

F I G U R E 7
The effects of the different ratios and concentrations on the key indicators of IC. Cross-linking HA decreased the apparent permeability, without altering the disease conditions at a 1:1 ratio of cross-linked HA to naïve HA at both 1 and 3 mg/ml HA concentrations