Bioinspired Bottlebrush Polymers Effectively Alleviate Frictional Damage Both In Vitro and In Vivo

Bottlebrush polymers (BB) have emerged as compelling candidates for biosystems to face tribological challenges, including friction and wear. This study provides a comprehensive assessment of an engineered triblock BB polymer's affinity, cell toxicity, lubrication, and wear protection in both in vitro and in vivo settings, focusing on applications for conditions like osteoarthritis and dry eye syndrome. Results show that the designed polymer rapidly adheres to various surfaces (e.g., cartilage, eye, and contact lens), forming a robust, biocompatible layer for surface lubrication and protection. The tribological performance and biocompatibility are further enhanced in the presence of hyaluronic acid (HA) both in vitro and in vivo. The exceptional lubrication performance and favorable interaction with HA position the synthesized triblock polymer as a promising candidate for innovative treatments addressing deficiencies in bio‐lubricant systems.


Introduction
Friction, an omnipresent physical phenomenon in the universe, occurs whenever an object slides across a surface, generating a resistive force that opposes the direction of motion.In various instances, this frictional force can harm the integrity of the 5% to 50%, depending on the geographical region and escalating progressively with age. [9]Hence, a clear and urgent clinical need exists to discover technological solutions to prevent or delay frictional damage early and ensure long-lasting efficacy.
Polymeric lubricants have undergone extensive investigation to address these clinical needs. [10]As a main component in healthy cartilage, natural HA is used for injection between the articular joints to treat joint diseases like osteoarthritis.The aim is to restore the viscoelastic properties of the degraded synovial fluid and induce various biological effects, including analgesia, mild anti-inflammatory, and antioxidative actions. [11]However, natural HA has technical limitations, such as rapid clearance and degradation in vivo. [12]Researchers proposed to use chemically crosslinked HA to address the technical shortcomings of natural HA, [10,13] or to use polymer micro/nanoparticles [14] and hydrogels [15] as drug carriers for the delivery of therapeutic agents to articular joints. [10]An alternative approach involves drug-free injectable polymers that can provide a robust protective layer and excellent lubricant properties.Bottlebrush (BB) polymers, drawing inspiration from the architecture of lubricinone of the key protein components of synovial fluid-have attracted considerable attention.These polymers possess a distinctive architecture characterized by an extended backbone with covalently grafted pendant side chains. [16]Over the past decade, advancements in polymerization techniques have afforded precise control over the architectural parameters of BB polymers.This includes the size of both the backbone and side chains, and grafting densities. [17]This architectural versatility empowers researchers to customize BB polymers' rheological and lubricating properties, making them promising candidates for various coating applications. [18]2-Methacryloyloxyethyl phosphorylcholine (MPC)-based bottlebrush (BB) polymers have been demonstrated to have exceptional lubricating properties and can effectively protect surfaces. [16,19]These BB polymers offer remarkably lower coefficients of friction (CoF) compared to HA, typically ranging from 0.001 to 0.03 [16,18,19] as opposed to HA's range of 0.2-0.6, [4,13]and the efficiency is maintained even under pressures near or exceeding physiological levels, and in the presence of salts. [10,16,19,20]Experimental parameters related to this phenomenon and their lubrication performance comparison were summarized in a recent work. [10]However, previous studies have focused mainly on specific substrates (e.g., mica) or formulations. [19]A comprehensive evaluation of the tribological behavior of MPC-based BB polymer and its combination with biological components on different substrates is still lacking.Besides, despite their significant potential in treating diseases related to various bio-lubrication systems, limited efforts have been directed toward real-world applications.In this work, we aim to present the versatile applications of MPC-based BB polymers, both with and without terminal adhesive groups, in treating osteoarthritis and dry eye syndrome using corresponding animal models.The real-time adsorption and antifouling properties of the BB polymers were evaluated by fluorescent microscopy.The lubricant properties and wear resistance were measured using the Surface Force Apparatus (SFA) under close to physiological conditions.Both in vitro and in vivo experiments were performed to validate the safety and efficacy of the polymer formulations for the treatment of joint diseases and dry eye syndrome.Additionally, we demonstrated that the polymers are equally effi-cient in enhancing the lubricating properties of medical device surfaces such as contact lenses, demonstrating these polymers' wide range of possible applications.These proof-of-concept studies bridge the gap between the promising properties of these materials and their practical translation into effective medical interventions.

Low Cell Toxicity and Excellent Lubricant Performance of BB in Combination with Hyaluronic Acid In Vitro
Two BB polymers were designed and used with or without HA in different settings.The chemical structures (left) and morphology (right) of the monoblock (B) and triblock (ABA) polymers are presented in Figure 1A,B, respectively.Both polymers exhibit a bottlebrush structure characterized by a backbone of degree of polymerization DP = 880 and side chains of similar length (DP = 92-109).Analysis of AFM images of the two polymers deposited on a mica substrate reveals that both BB polymers share a comparable apparent contour length, L c .For B, the contour length is L c = 121.6 ± 9.6 nm, and for ABA, L c = 126.5 ± 8.8 nm.This observation conflicts with the theoretical estimate suggesting that the contour length of ABA should be ≈1.5 times longer than that of B. The discrepancy arises from the structure of the terminal A blocks, which lack side chains and are linear, making them prone to folding.This folded structure cannot be distinguished by AFM, resulting in the apparent similarity in length between the ABA and B bottlebrushes in AFM images.Additionally, the average width of B is w = 74.7 ± 5.8 nm, whereas for the triblock polymer ABA, w = 85.1 ± 6.4 nm.These findings align with expected values, considering that the sole difference between the two polymers lies in the small terminal blocks of ABA.
Moreover, salt may affect the structure and the properties of zwitterionic polymers, as previously reported. [21]However, PMPC bottlebrushes exhibit salt-resistant properties, as shown in our previous work. [20]This was also observed for regular PMPC zwitterionic brushes in another study, where the authors also showed that the presence of NaCl and CaCl 2 has little influence on the morphology and lubricant properties of the zwitterionic brushes. [22]he first experimental setting employed to assess the lubricating capacity of the bottlebrush polymers (BB) involved articular joints.In humans, current visco-supplementation treatments aim to postpone cartilage degradation caused by frictional damage, utilizing injectable HA solutions.The effectiveness of this treatment has been vigorously scrutinized in the past decades, [23] to the extent that the American Academy of Orthopedic Surgeons issued clinical guidelines in 2013 and 2016 advising against the use of HA injections for patients with knee osteoarthritis. [24]Consequently, we hypothesized that mixtures of HA and ABA polymer, designed to interact with HA via its lateral, polycationic block, could enhance chondroprotection and impede the progression of cartilage damage compared to HA alone.
The rheological properties of the HA-ABA mixtures were initially assessed to ensure injectability into the knee joint.As depicted in Figure S1 (Supporting Information), the results illustrate that both HA and HA-ABA mixtures exhibit similar rheological behavior.In the linear viscoelastic regime, at low strains, the elastic modulus G' of the solutions remained constant and significantly greater than the viscous modulus, G″, indicating a gel-like behavior.As shear strain increased beyond the linear viscoelastic regime, both G' and G″ decreased, signifying material fluidization.At a critical strain value of 30% for HA and 60% for HA-ABA, G' became smaller than G″, revealing a transition to a liquid-like behavior.The stress required to induce the transition is estimated to be ≈50 Pa, a significantly lower value than other injectable materials. [25]This shift from gel-like to liquid-like behavior at low triggering stress confirms the polymers' injectability and suitability for in vivo use.
To investigate the safety of the ABA polymer alone and when mixed with HA, we evaluated the cell toxicity of HA, ABA, and their mixture (HA-ABA) using three cell lines obtained from human patients who followed a total knee replacement intervention-synoviocytes, chondrocytes, and osteoblastsrepresentative of the articular joint. [26]Before these experiments, ABA underwent sterilization through autoclaving.This process was confirmed not to induce any significant alterations, as evidenced by gel permeation chromatography and AFM image analysis (Figure S2, Supporting Information).At a concentration of 1mg mL −1 , HA exhibited no cell toxicity and even promoted the proliferation of synoviocytes (see Figure 1C,D).In comparison, ABA displayed slight toxicity against synoviocytes (≈5% of cell death) at a concentration of 0.1 mg mL −1 and slightly enhanced the proliferation of the other two cell lines.When combined, the mixture amalgamated the advantages, reducing ABA's slight toxicity against synoviocytes while maintaining favorable characteristics toward the other cell lines.Overall, ABA demonstrated good biocompatibility for the three cell lines, a feature further augmented in the presence of HA.None of the tested formulations resulted in a reduction of cell viability and activity exceeding 30%.Consequently, in accordance with ISO 10993-5, our samples were considered non-cytotoxic. [27]ubsequently, we examined whether ABA could induce the production of Matrix Metalloproteinases (MMPs) in chondrocytes in vitro.MMPs precisely regulate the balance between synthesis and degradation in the cartilage extracellular matrix (ECM), and abnormal amounts of MMPs lead to pathological cartilage destruction. [28]The expression levels of MMP3 and MMP13 were assessed after exposing human chondrocytes from arthritic patients to ABA or HA solutions for 24h.The results indicated a 1.5-fold increase in MMP3 expression for ABA compared to HA (Figure 1E).Mixing HA with ABA resulted in a slight decrease in expression, ≈1.3-fold.In contrast, MMP-13 expression showed a 1.2-fold increase for ABA and a 1.4-fold increase in the HA/ABA combination.The modest increase suggests that the ABA solution did not trigger any adverse reaction in chondrocytes and confirmed that the polymer is well-tolerated and suitable for intra-articular injections overall.
Before conducting in vivo experiments, we confirmed that the HA-ABA mixture at a 10:1 ratio could effectively offer tribo-protection and good lubrication in biological media (Figure 1G,H).Bovine synovial fluid (BSF) from healthy animals was utilized, and Chondroitinase ABC (CASE-ABC) [29] and hyaluronidase (HAase) were added at concentrations of 137 and 200 units/mL, respectively, to degrade HA and aggrecans, simulating arthritic synovial fluid.
Compared to BSF alone, HA solution at 10 mg mL −1 did not present a convincing load-supporting capacity, as assessed by SFA measurements (Figure 1F).When compressing the HA solution between two mica surfaces, the measured interaction forces were short range (onset of interaction below 50 nm of separation) and the hard wall at high normal load (which is related to the load support capacity of the polymer under high confinement) is small (D < 1 nm at F/R = 30 mN/m).On the other hand, BSF exhibited long-range repulsive forces starting at 150 nm and rapidly increasing to reach a hard wall value of D = 21 nm at F/R = 30 mN m −1 .Interestingly, when mixed, HA and ABA could mimic BSF attributes with an interaction onset of 150 nm and a hard wall of D = 27 nm at F/R = 30 mN m −1 .
Healthy BSF exhibited a coefficient of friction (COF) of 0.4 when measured between two mica surfaces, significantly decreasing after adding Chondroitinase ABC and HAase to COF = 0.2 (Figure 1G,H).In contrast to healthy BSF, "arthritic" BSF could not sustain a high normal load, and severe damage to the surfaces was observed at a load of L = 8 mN.This limitation in the formulation can be effectively addressed by incorporating ABA, which not only safeguards the surface from damage, but also brings the COF back to a value closer to the COF of the healthy BSF.These results demonstrate that ABA in synovial fluid can provide surface protection against damage even when HA and other fluid components have been compromised.

Triblock Bottlebrush Polymer Exhibits Superior Chondroprotection in Combination with Hyaluronic Acid In Vivo
To assess the potential benefits of using ABA in conjunction with HA to treat arthritis, we employed a surgical model of OA based on the anterior cruciate ligament transection (ACLT) (Figure 2A).This model (ACLT) replicates the early mechanisms involved in joint instability, cartilage degradation, altered biomechanics, and the overall development of post-traumatic osteoarthritis. [30]n this study, we monitored cartilage composition, compressive modulus, and thickness evolution to demonstrate ABA-HA treatment's effectiveness in delaying the establishment of OA symptoms.
Early signs of OA involve the degradation of cartilage fibers and, consequently, a severe loss of proteoglycans and a significant decrease of the instantaneous compressive modulus, often associated with an increased expression of MMP in the joint cavity.Histology of the condyles provided an evaluation of cartilage composition changes between each group.Proteoglycan-depleted cartilage was observed in the superficial zone of articular cartilage of ACLT rats treated with PBS (Figure 2B).HA and ABA-HA treated joints displayed greater proteoglycan preservation, as indicated by the intensity of Safranin O staining of the articular cartilage (red color).The instantaneous compression modulus of the cartilage was then evaluated using a micro-indentation mapping technique (see Supporting Information).The contra-lateral joint served as each animal's control (CTL) and was compared to the treated ACLT joint.As shown in Figure 2C, CTL groups presented a relatively uniform distribution of compressive modulus with no drastic variations over the entire surface of the condyle.Averaged values of the compressive modulus were obtained for each quadrant of the condyle (Lateral Anterior Condyle, LAC; Lateral Posterior Condyle, LPC; Medial Anterior Condyle, MAC; Medial Posterior Condyle, MPC) for each treatment group and CTL and compared.Mapping of the compressive modulus revealed a systematic decrease in modulus in the MPC quadrant, consistent with the joint instability triggered by the ACLT surgery.For the groups treated with PBS or HA, a significant decrease in compressive modulus was observed in the MPC quadrant and, to some extent, in the LPC quadrant.Therefore, HA alone does not seem to offer any advantages in cartilage protection, even at such an early stage of the disease.The group treated with the ABA-HA mixture exhibited no signs of compressive modulus loss compared to the CTL (see Figure 2C,D).When normalized to control, the decrease in compression modulus in the LPC quadrant was 62% for PBS, 7% for HA, and 2% for ABA-HA treatments.In the MPC quadrant, the variation was ≈37% for PBS, 42% for HA, and 2% for ABA-HA treatment (Figure 2D).Cartilage thickness was also mapped and compared to the CTL group (Figure 2E).The group treated with PBS exhibited a significant increase in cartilage thickness over the whole surface of the condyle.This phenomenon, also observed in human OA, [31] is a characteristic attribute of early-stage OA and is a direct consequence of collagen fiber degradation and cartilage swelling.The group treated with HA showed a similar trend with an average increase in thickness of 33% and 37% in the MPC and LPC quadrants compared to CTL (Figure 2F).Combined histology, compressive modulus, and cartilage thickness data confirmed that intra-articular injections of HA slightly impact the progression of the disease, even at a very early stage.Combining HA and ABA allowed for a significant delay in cartilage swelling, as shown in Figure 2E,F.Therefore, the HA-ABA mixture was significantly more efficient in delaying the deterioration of cartilage and in preventing aggrecan loss at the early stage of the disease, which, combined with its other attributes such as injectability and low toxicity, positions it as a promising candidate for the treatment of OA.

BB Polymers in Combination with HA Lubricate Efficiently the Eye Cornea Surface Ex Vivo
In a second experimental setting, we investigated the lubricating properties of the different polymers on the corneal surface to ex-plore the potential applications of B and ABA polymers as treatment options for dry eye syndrome.Given that HA solutions are already approved for topical application as eye drops, we tested B and ABA polymers alone and in combination with HA to explore potential synergies.
We initially assessed the affinity of different polymer formulations (i.e., B, ABA, B-HA, and ABA-HA) for the rabbit corneal surface using Cy5-labelled BBs. Figure 3A illustrates the time evolution of polymer adsorption during three consecutive injections of increasing concentrations.The results indicated rapid adsorption at each concentration tested, becoming detectable within the first 30 min of experimentation, as evidenced by the steep increase in fluorescent intensity at the cornea's surface.This was followed by a slower continuous process, gradually reaching a plateau.When the polymer solution was replaced with PBS (rinsing step), fast desorption occurred, with the fluorescent intensity decreasing by ≈40-50% within the first 30 min.The fluorescence signal did not significantly change thereafter, indicating the presence of a stable polymer layer on the cornea for all the formulations.Adsorption curves were fitted with a two-binding sites model (1:2 model) to extract quantitative values of the binding constants for each polymer formulation.The choice of the 1:2 model was motivated by the inadequacy of simpler models, such as the 1:1 model, to accurately describe the binding process, as demonstrated in Figure S3 (Supporting Information).This observation might arise from the heterogeneity of the cornea surface or different types of binding sites on the polymer and remains a question to be explored.Nevertheless, the 1:2 model allows for evaluating the affinity of the BB for the cornea surface.
For all the tested formulations, the first association rate (k a1 ) was consistently one order of magnitude higher than k a2 , and the first dissociation rate was always one order of magnitude higher than the second (k d1 > k d2 , as shown in Figure 3B).These differences were more prominently reflected in the dissociation constants of each site.The dissociation constants K D1 = k d1 /k a1 values were 74.1 nM for B, 89.4 nM for ABA, 178.3 nM for HA-B, and 78.9 nM for HA-ABA.For the second binding site, the dissociation constants K D2 = k d2 /k a2 were at least one order of magnitude smaller for all the tested formulations.
Tissue surfaces such as cartilage and the cornea are primarily composed of amino-and hydroxyl-rich components including proteins and carbohydrate polymers.These surfaces typically carry a negative charge in physiological conditions. [32]18a] Basically, B blocks (bearing the pendant chains) can attach to the surface primarily through hydrogen bonding with the exposed amino or hydroxyl groups, In the case of ABA, electrostatic attraction also plays a significant role in surface adhesion.
These observations conclude that the second binding site was significantly stronger than the first one for all the formulations.Comparing the different polymer solutions, it appears that no clear differences in K D for each biding site were observable, suggesting that 1) the presence of HA does not affect the affinity of the two BBs for the cornea surface; 2) the A blocks of ABA have little impact on the polymer affinity toward the surface.
Comparing the lubricating properties of the different formulations revealed interesting differences.Freshly harvested rabbit corneas were tested against a PDMS surface mimicking a contact lens.The time evolution of the friction force, called the friction trace, was monitored for each formulation and is shown in Figure 3C.In tear fluid, B and ABA solutions exhibited a friction trace with a high plateau indicative of a high kinetic friction force.On the other hand, HA and its mixtures with the BB polymers presented slightly lower plateaus.Interestingly, HA alone showed signs of stiction followed by stick-slip events, which indicate transient adhesion of the cornea to PDMS during sliding (Figure 3C).This may arise from insufficient adsorption of HA on the cornea and PDMS, allowing for direct adhesive contact between the two surfaces.On the other hand, the mixtures of BB and HA displayed a smooth trace with low kinetic plateaus.Specifically, the friction signals of B and ABA on dry cornea changed significantly when HA was included under similar applied loads, and the shear process reached a steady state rapidly for the two mixture formulations, especially for ABA-HA.The static friction disappeared, and the dynamic friction decreased compared to HA.
The friction force increases linearly with the normal load, as shown in Figure 3D.The friction curves pass through 0 for all the conditions tested, indicating the negligible contribution from adhesion. [33]For fresh cornea, the friction coefficient, defined as the slope of the friction curve, decreases slightly (≈5-10%) when lubricated by HA, B, or ABA compared to tear fluid.The mixtures B-HA and ABA-HA are more effective lubricants, as they can reduce the friction coefficient by 20% for B-HA and 40% for ABA-HA compared to tear fluid.When performing these experiments with corneas exposed to air for one hour before the experiment, the friction coefficient differences were significantly more pronounced between formulations.In tear fluid, the friction coefficient of the dry cornea is ≈1.7 times higher than that of fresh cornea (Figure 3E).The friction coefficient decreases ≈50% when using B or ABA as lubricants compared to tear fluid (as shown in Figure 3F), highlighting their excellent lubricating properties for cornea surfaces exposed to air for an abnormal amount of time.The decrease percentage reaches 60-70% when mixing BB with HA, confirming the synergistic effect between both macromolecules.Interestingly, the ABA-HA mixture presents the highest decrease in friction coefficient of all tested conditions for both dry and fresh cornea.
The superior lubricating properties of the ABA-HA mixture are not linked to the affinity of the ABA polymer to the cornea surface, as all the tested formulations exhibited similar affinity constants.Instead, the ABA-HA mixture is the only one that exhibits a high affinity between both components via electrostatic interaction between the quaternized amine groups present on the A block of the ABA polymer and the carboxylic functions distributed on HA chains.The strong intermolecular interactions between both components allow the lubricating film to maintain its integrity and sustain higher normal loads than HA alone or B-HA mixtures.19b,c]

BB Polymers in Combination with HA Alleviate Dry Eye Syndrome Symptoms In Vivo
We also tested the ability of the BB polymer, either alone or in combination with HA, to alleviate particular symptoms of dry eye syndrome.Eye dryness was successfully induced by administering scopolamine via a subcutaneous osmotic pump.The rats underwent topical eye treatments from day 5 to day 21 (Figure 4A).Tear volume assessments were carried out on days 0 and 4 and, subsequently, once a week until the study concluded.Notably, scopolamine administration led to a significant tear volume reduction as early as 4 days post-implantation of the osmotic pump (p < 0.0001), with the volume decreasing from 5.2 mm to 2.2 mm.In contrast, the tear volume of rats treated with PBS remained stable throughout the study period, as shown in Figure 4B.Despite the various experimental treatments, no tear volume restoration was observed, even when measurements were conducted both before the first treatment and after the last treatment of each day (Figure 4C-E).
The fluorescein score was assessed on days 0, 21, and 27.Due to its hydrophilic properties, fluorescein penetrates poorly into the lipid layer of the corneal epithelium and does not stain a healthy cornea.However, fluorescein staining effectively detects a wide range of superficial punctate epithelial erosions on the cornea induced by dry eye disease. [34]Due to the unexpected non-zero baseline level, the results were presented as percentages relative to the baseline established on day 0. On day 21, there was a significant increase in the fluorescein score following scopolamine administration in the absence of treatment, as shown in Figure 4F.Notably, the fluorescein score also rose significantly in rats treated solely with ABA or HA.However, combined therapy of both ABA and HA appeared more effective in preventing an increase in the fluorescein score.Despite high variability in this experimental group, the mean score did not significantly differ from the control group's (Figure 4F).By day 27, no increase in fluorescein score was observed in the rats treated with scopolamine compared to those treated with PBS (Figure 4G), even though their eyes remained dry, as indicated by the persistently low tear volume (Figure 4B).The tear break-up time, a measure of tear film stability, was assessed on days 0, 21, and 27.On day 0, the tear break-up time was comparable across all experimental groups (Figure S4A, Supporting Information).Scopolamine administration significantly reduced the tear break-up time by 36% and 46% on days 21 and 27, respectively, compared to the control group (as shown in Figure S4B and S4C, Supporting Information).HA treatment alone increased tear break-up time to levels similar to those of the PBS control group on day 21 (Figure S4B, Supporting Information).However, from day 22 onward, even though scopolamine continued to be administered, the rats did not receive any further treatment.Consequently, the positive ef-fect of HA on tear break-up time was not sustained; its efficacy waned and was no longer evident 6 days after the cessation of treatment (Figure S4C, Supporting Information).
On Day 28, the rats' eyes were sectioned into tissues, which were then subjected to H&E staining and immunofluorescence staining using an antibody targeting the leukocyte cell surface marker CD45.The analysis revealed no significant change in epithelial thickness in rats with dry eye disease compared to the control group, as shown in Figure S5A (Supporting Information).However, a slight trend toward decreased epithelial thickness was noted following scopolamine administration (Figure S5A, Supporting Information).Additionally, there was an increase in the infiltration of CD45+ cells in the cornea post-scopolamine administration, with cell counts rising from 4.2 to 11.3 cells.While treatment with ABA and HA alone did reduce the number of CD45+ cells compared to untreated rats, these results did not reach statistical significance (Figure S5B, Supporting Information).

BB Polymers Alone Form a Stable Lubricant Layer on Contact Lenses and Improve Anti-Fouling Properties
In this setting, we assessed the BB polymer alone as a lubricating and antifouling additive for contact lenses.Similar to previous tests, we examined the adsorption and desorption of various formulations on a silicone contact lens surface, as depicted in Figures 5A and S6 (Supporting Information).Both B and ABA exhibited comparable adsorption-desorption curves.A plateau emerged after an adsorption period of ≈90 min, signifying equilibrium between the adsorption and desorption processes.During the desorption phase, the fluorescent intensity decreased by 60-70%, remaining relatively stable after 90 min.ABA demonstrated a slightly higher adsorption rate and less desorption compared to B after rinsing, likely attributable to the presence of the A blocks that offer additional anchoring to the negatively charged silicone surface of the lens.
As observed with the cornea, the friction force between the contact lens and a flat PDMS surface exhibits a linear increase with the applied load for all tested formulations, as illustrated in Figure 5B.Compared to tear fluid alone, the friction coefficient of the BB polymers in tear fluid is consistently lower, with μ = 0.13 for tear fluid, μ = 0.059 for B, and μ = 0.025 for ABA.This notable difference in lubricating properties is likely due to the A block on the BB polymer, which is expected to strongly anchor the polymer to the surface.This characteristic was also evident when examining the stability of the lubricant film under repeated shearing cycles.The stability of the polymeric layer under repeated shearing is crucial, considering that humans blink their eyes ≈15 000 times a day. [35]We conducted N = 1000 shear cycles to monitor the evolution of the friction of the contact lens in tear solutions with and without polymer under a load of L = 3 mN (Figure 5D).For all tested conditions, the friction force increased with the number of shear cycles, reaching almost ≈0.8 after N = 1,000 shear cycles in the tear solution (Figure 5E).In contrast, the increase in friction force was much less pronounced in the presence of BB polymers, with only 10% increase for B and 5% increase for ABA, emphasizing the role of BB polymers in enhancing lubricant properties.
In addition to evaluating the lubricating properties of the BB polymers, we examined their ability to "passivate" the surface of a contact lens after adsorption.Protein adsorption on contact lenses can lead to bacterial adhesion, potentially reducing optical quality and oxygen permeability. [36]Therefore, it is crucial to assess the anti-fouling properties of BB polymer-coated contact lenses.The results demonstrate that both BB polymers can more effectively reduce lysozyme adsorption/retention on the contact lens surface coated with B or ABA polymer (C p = 100 μg mL −1 for 1 h) compared to the bare surface, as depicted in Figure 5F.The well-hydrated phosphorylcholine groups along each side chain of the BB polymers provide excellent hydration and steric repulsion to the lens surface, resulting in robust anti-fouling properties. [37]ompared to B, the ABA polymer efficiently combines lubrication and anti-fouling properties, both crucial for contact lenses.Prolonged use of contact lenses can lead to dry eye symptoms, and in severe conditions, it may even decrease the entire corneal thickness or increase the risk of bacterial infection. [38]These experiments highlight that immersing the contact lens in the BB solution dramatically improves its surface properties and efficiently prevents protein deposition.18a,39] This characteristic underscores its potential utility in applications related to contact lenses.

Conclusion
Affinity studies have demonstrated that BB polymers can rapidly absorb various bio-surfaces and create a protective layer that lubricates the surface and reduces the risk of infection.This paper showcases the triblock BB polymer ABA, proving its robust lubricating properties when applied to cartilage, the eye, and contact lenses.Notably, ABA's tribological characteristics and biocompatibility can be further enhanced with HA.With its outstanding lubrication performance and favorable interaction with HA, the ABA polymer holds promise as a key component for new treatments addressing deficiencies in bio-lubricant systems.

Experimental Section
Ethical Approvals: All OA model procedures were approved by the Sainte-Justine Hospital Research Center animal ethics committee (protocol #660) and conformed to Canadian Council on Animal Care (CCAC) guidelines. [3]Approval for the dry eye study was acquired from both the Cégep de Lévis Animal Care Committee and the Comité d'éthique de l'utilisation des animaux at the Université de Montréal (protocol 23-215).The animal care facility at TransBIOTech is accredited by the CCAC.The study strictly adhered to CCAC standards and regulations governing the ethical use of animals in research.
Statistical Analysis: The data for the OA study, fluorescein score, and CD45+ cell count were log-transformed before a one-way ANOVA using the transformed data.Outliers were identified using quartile calculations: the first quartile Q1, the third quartile Q3, and the interquartile range IQR (IQR = Q3 -Q1).Points beyond the range of Q1 -(1.5 × IQR) and Q3 + (1.5 × IQR) were considered outliers.Results are presented as means ± SD (n = 5) for in vivo OA study or as means ± SEM (n ≥ 6) for in vivo dry eye study.A two-way ANOVA was employed to assess statistical significance for tear volume, accompanied by Tukey's or Šídák's post hoc test for multiple comparisons, when applicable.A one-way ANOVA was conducted for the OA study, fluorescein score, and histological analysis involving counting CD45+ cells, followed by Tukey's post hoc test for multiple comparisons (* p < 0.05, ** p < 0.01, *** p < 0.001).Tear break-up time and histological analyses involving epithelium thickness utilized the nonparametric Kruskal-Wallis and Dunn's tests for multiple comparisons (* p < 0.05, ** p < 0.01, *** p < 0.001).GraphPad Prism v10.0.3 software was used to generate figures and conduct statistical analyses.

Figure 1 .
Figure 1.ABA polymer exhibits low toxicity and re-establishes lubricating and load-support properties of synovial fluid in vitro A) Chemical structure and morphology of monoblock bottlebrush B; B) Chemical structure and morphology of triblock bottlebrush ABA; C) Cell toxicity of HA, ABA and their mixture to synoviocytes, chondrocytes and osteoblasts; D) Cell proliferation of synoviocytes, chondrocytes and osteoblasts when pretreated with HA, ABA and their mixture; E) Fold change of MMP3 and MMP13 on ABA and ABA-HA vs HA; F) Normal force as a function of distance for different formulations; G) Friction force (F) as a function of the applied load (L) between surfaces lubricated by BSF, BSF-CASE ABC + HAase and BSF-CASE ABC + HAase + ABA; H) Friction coefficients of surfaces lubricated by BSF, BSF-CASE ABC + HAase and BSF-CASE ABC + HAase + ABA.

Figure 2 .
Figure 2. ABA mitigates cartilage degeneration in early post-traumatic OA A) Representation of the treatment timeline B) Representative images of histological slides of condyles of each group.Safranin O staining (which colors proteoglycans red), counterstained with Fast Green FCF (which colors proteins green), and with Weigert's hematoxylin (which colors nuclei black) indicates Proteoglycan's depletion in the superficial zone of the cartilage of ACTL rats treated with PBS, and greater proteoglycans preservation in the superficial zone of the articular cartilage in the HA and ABA-HA treated joints; C) Representative color maps of the instantaneous compression modulus of condyle cartilage from each of the three treated groups PBS, HA, and ABA-HA; D) Average variation of the compression modulus in the posterior quadrant of the condyle.Graphical representation of the measurement setup and data collection method.;E) Representative color maps of the cartilage thickness of condyles from each of the three treated groups PBS, HA, and ABA-HA; F) Average cartilage thickness variation in the condyle's posterior quadrant.Graphical representation of the measurement setup and data collection method.Statistical analysis involved log-transforming the data and assessing statistical significance using one-way ANOVA, followed by Tukey's post hoc test for multiple comparisons (* p < 0.05, ** p < 0.01, *** p < 0.001).Results are presented as means ± SD (n = 5).

Figure 3 .
Figure 3. BB polymers exhibit high affinity to cornea surface and recover lubrication of dried cornea ex vivo A) Adsorption/desorption curves of different formulations on the cornea; B) Binding constants of different polymer formulations on the cornea; C) Friction signals of dry and fresh eyes lubricated by different polymer formulations; D) Friction force as a function of applied load for fresh eyes lubricated by different polymer formulations; E) Friction force as a function of applied load for dry eyes lubricated by different polymer formulations; F) Decrease of COF for both dry and fresh cornea lubricated by different polymer formulations comparing to the neat cornea in tear fluid.BB polymer solutions (100 μg mL −1 ) and their combination with HA solution (1000 μg mL −1 ) for the lubrication tests (Figure 3D-F).

Figure 4 .
Figure 4. ABA polymer mitigates dry eye symptoms in vivo with HA A) Schedule of dry eye disease induction and treatments.B-E) Quantification of the tear volume over the study (B), as well as on days 7 (C), 14 (D), and 21 (E).F-G) The fluorescein score was assessed on days 21 (F) and 27 (G).Statistical analysis involved log-transforming the data and evaluating statistical significance using one-way ANOVA, followed by Tukey's post hoc test for multiple comparisons.(ns: non-significant, * p < 0.05, ** p < 0.01, *** p < 0.001).Results are presented as means ± SEM (n ≥ 8).

Figure 5 .
Figure 5. BB polymers films provide robust and stable antifouling and lubricating properties to contact lenses A) Adsorption/Desorption of the polymers on contact lens; B) Evolution of friction force with applied load for the contact lens with and without polymers; C) Comparison of friction coefficient of the contact lens; D) Friction force signals of the contact lens after N shear cycles; E) Evolution of friction force of the contact lens with the number of shear cycles N; F) Percentage of lysozyme washed out from the surface of a contact lens with BB polymer pre-adsorbed on it.BB polymer solution concentration C p = 100 μg mL −1 for the lubrication tests (Figure 5B-E).