The effect of molecular weight of hyaluronic acid on corneal cell viability

Abstract

Hyaluronic acid (HA) is widely utilized in ophthalmology for its viscoelastic and wound-healing properties; however, its effects vary according to molecular weight (MW). This study evaluates the toxicity induced by HA across corneal epithelial cells, keratocytes, and endothelial cells to ensure its safe clinical application. Human corneal epithelial cells, keratocytes, and endothelial cells were cultured under standardized conditions. Viability was assessed using the Cell Counting Kit-8 (CCK-8) and the live/dead cell staining after exposure to HA of varying MW. The MW categories were very low (VLMW, MW range from 8000 to 15,000 Da), low (LMW, MW range from 130,000 to 150,000 Da), medium (MMW, MW range from 750,000 to 1,000,000 Da), and high (HMW, MW range over 1,000,000 Da)), at concentrations ranging from 0.025% to 0.5% for durations of 24, 48, and 72 h. LMW HA exhibited the highest toxicity across all layers of corneal cells, demonstrating a dose-dependent decrease in cell viability. In epithelial cells specifically, 0.5% LMW HA resulted in a reduction of viability by more than 50% within 24–72 h, with concentrations greater than or equal to 0.25% proving hazardous after 24 h. Keratocytes experienced toxicity across all HA MW, with LMW HA leading to the most significant reductions. In corneal endothelial cells, LMW HA decreased viability by more than 50% after 48 h. Conversely, VLMW HA and HMW HA showed minimal toxicity. These findings highlight the importance of selecting appropriate HA formulations based on MW to ensure both efficacy and safety in ophthalmic treatments.

Introduction

Hyaluronic acid (HA) is a naturally occurring glycosaminoglycan found throughout connective, epithelial, and neural tissues. It plays a critical role in maintaining tissue hydration, elasticity, and various cellular functions^1–5. It is present in various ocular tissues, such as the aqueous humor, trabecular meshwork, and vitreous body⁶. HA has been extensively utilized in the medical field for decades^7–9. In 1976, Pharmacia launched high MW HA (HMW HA) under the brand name Healon for treating arthritis, subsequently expanding its applications to ophthalmic surgery. In ophthalmology, HA is frequently utilized in eye drops and surgical applications due to its viscoelastic properties and potential for promoting wound healing^4,6,7,10. By 1978, Healon was utilized in cataract surgeries to safeguard the corneal endothelium, and in 1982, its inaugural clinical usage in eye drops for severe dry eye was documented¹¹. However, studies have demonstrated that the molecular weight (MW) of HA significantly affects its biological activities, particularly in relation to corneal epithelial cells¹².

Previous studies have shown that the MW of HA may be a critical determinant of its biological effects^2,13. High MW HA, typically ranging from 1000 to 2000 kDa, has been linked to enhanced cell proliferation, migration, and wound healing, while low MW HA, ranging from 50 to 200 kDa, and ultra-low MW HA, below 50 kDa, are associated with pro-inflammatory responses and cytotoxic effects^4,12–15. Nevertheless, these effects vary depending on cell and tissue types, which contributes to ongoing debate^12,15,16.

The cornea, the outermost layer of the eye, serves both as a protective barrier against pathogens and a crucial component for vision¹⁷. Damage to the cornea can result in impaired vision and an increased susceptibility to infections. Considering the clinical significance of HA, understanding its effects on corneal cells is vital for developing safe and effective ophthalmic treatments. However, the MW and concentrations of HA utilized in ophthalmology vary,^11,18 and there is limited research on how different MW of HA affect various corneal cell types.

This study aims to assess the cytotoxic effects of HA on human corneal cells—epithelial cells, keratocytes, and endothelial cells—across various MW and concentrations. By examining dose-dependent impacts, we aim to pinpoint optimal formulations for clinical use.

Results

HA toxicity in human corneal epithelial cells

The toxicity profile of hyaluronic acid (HA) on corneal epithelial cells varies depending on its MW. VLMW HA showed no significant toxicity at concentrations up to 0.5% during a 72-h culture period. A Mild decrease of viability (but over 80% of viability maintained) was observed when corneal epithelial cells were exposed to ≥ 0.25% of HMW HA for 24 h or 0.5% of HMW HA for 48 h. In contrast, LMW HA and MMW HA exhibited significant dose-dependent toxicity. LMW HA demonstrated the highest toxicity, with a reduction in cell viability exceeding 50% when corneal epithelial cells were cultured with 0.5% LMW HA from 24 to 72 h (Fig. 1). Moreover, LMW HA concentrations of 0.25% or more were considered unsafe for exposures longer than 24 h. Interestingly, a slight increase in viability was noted with VLMW HA and HMW HA after 48 and 72 h of incubation. The live/dead cell staining revealed a significant increase in cell death following 24-h exposure to 0.5% LMW HA and 0.5% MMW HA. Additionally, the morphology of live cell population exhibited an increase of elongated cells following the exposure to 0.5% LMW HA and 0.5% MMW HA (Fig. 2).

Fig. 1.

The viability results of corneal epithelial cell cultures exposed to very low (VLMW), low (LMW), middle (MMW), and high molecular weight (HMW) hyaluronic acid (HA) at concentrations ranging from 0.025 to 0.5% over periods of 24, 48, and 72 h. Triton X-100 (TX-100) was used as the positive control of cytotoxicity. Values are presented as mean ± SEM and originated from three independent experiments; each experiment was performed in triplicate. * p < 0.05, ** p < 0.01, *** p < 0.001.

Fig. 2.

Live/dead cell staining of corneal epithelial cell cultures following 24-h exposure to hyaluronic acid (HA) of varying molecular weights—very low (VLMW), low (LMW), medium (MMW), and high (HMW)—at concentrations of 0.1% and 0.5%. Viable cells are indicated in green, while non-viable cells appear red. DIC means differential interference contrast image. scale bar (black): 200 μm.

HA toxicity on human keratocytes

The toxicity profile of HA in keratocytes differed from that observed in corneal epithelial cells. All MW of HA induced significant dose-dependent toxicity in keratocytes. While high concentrations of VLMW HA and MMW HA induced mild toxicity, resulting in only a 30% decrease in viability after 72 h with 0.5% HA, HMW HA caused more severe toxicity than both VLMW HA and MMW HA, with LMW HA exhibiting the most pronounced toxicity among all MW tested (Fig. 3). Compared to other MW of HA, the live/dead cell staining revealed a significant increase in cell death following 24-h exposure to 0.5% LMW HA. However, the morphology of live cell population exhibited no significant change after exposure to 0.5% LMW HA (Fig. 4).

Fig. 3.

The viability results of corneal keratocyte cultures exposed to very low (VLMW), low (LMW), middle (MMW), and high molecular weight (HMW) hyaluronic acid (HA) at concentrations ranging from 0.025 to 0.5% for 24, 48, and 72 h showed that the viability of corneal keratocytes decreased as the concentration of HA increased across all molecular weights. Triton X-100 (TX-100) was used as the positive control of cytotoxicity. Values are presented as mean ± SEM and were obtained from three independent experiments; each experiment was performed in triplicate. * p < 0.05, ** p < 0.01, *** p < 0.001.

Fig. 4.

Live/dead cell staining of keratocytes following 24-h exposure to hyaluronic acid (HA) of varying molecular weights—very low (VLMW), low (LMW), medium (MMW), and high (HMW)—at concentrations of 0.1% and 0.5%. Viable cells are indicated in green, while non-viable cells appear red. DIC means differential interference contrast image. scale bar (black): 200 μm.

HA toxicity on human corneal endothelial cells

HA also demonstrated a dose-dependent toxicity profile in corneal endothelial cells. Among all MW, LMW HA induced the most significant toxicity. While VLMW HA, MMW HA, and HMW HA showed only mild toxicity with less than a 20% decrease in viability, the use of LMW HA for over 48 h resulted in a reduction in viability exceeding 50% (Fig. 5). A significant reduction in live cell populations and a corresponding increase in dead cells was observed following exposure to 0.5% LMW HA. However, the morphology of live cells exhibited no significant change (Fig. 6).

Fig. 5.

The viability results of corneal endothelial cell cultures exposed to very low (VLMW), low (LMW), middle (MMW), and high molecular weight (HMW) hyaluronic acid (HA) at concentrations ranging from 0.025 to 0.5% for 24, 48, and 72 h revealed that the viability of corneal endothelial cells decreased as the concentration of HA increased for all molecular weights. Triton X-100 (TX-100) was used as the positive control of cytotoxicity. Values are presented as mean ± SEM and were obtained from three independent experiments; each experiment was performed in triplicate. * p < 0.05, ** p < 0.01, *** p < 0.001.

Fig. 6.

Live/dead cell staining of corneal endothelial cell cultures following 24-h exposure to hyaluronic acid (HA) of varying molecular weights—very low (VLMW), low (LMW), medium (MMW), and high (HMW)—at concentrations of 0.1% and 0.5%. Viable cells are indicated in green, while non-viable cells appear red. DIC means differential interference contrast image. scale bar (black): 200 μm.

Discussion

This study demonstrated the toxicity profiles of HA across different MW on human corneal epithelial cells, keratocytes, and endothelial cells. While VLMW HA and HMW HA exhibited minimal toxicity, LMW HA showed pronounced dose-dependent toxicity across all cell types, with the most severe effects noted in epithelial cells.

HA is a naturally occurring glycosaminoglycan essential for maintaining tissue hydration, elasticity, and cellular functions^8,9. HA is synthesized in cells by hyaluronic acid synthase (HAS) on the cytoplasmic surface of the plasma membrane and is then transported to the pericellular space. Mammalian cells express three HAS enzymes: HAS-1, HAS-2, and HAS-3. HAS-1 and HAS-2 are responsible for producing high MWHA, with HAS-2 being more catalytically active and responsive to shocks, inflammation, and tissue repair. HAS-3 demonstrates the highest catalytic activity, yielding low MW HA. The MW of HA depend on various factors, especially the cell type. In humans, six hyaluronidase genes have been identified in somatic tissues, though only hyaluronidase -1 and hyaluronidase -2 are significantly expressed. Hyaluronidase -2, anchored to the cell membrane, cleaves high MW HA into approximately 20 kDa fragments, while hyaluronidase -1, a lysosomal enzyme, collaborates with hyaluronidase -2 to generate hyaluronan tetra-saccharides.

In ophthalmology, HA is frequently employed in eye drops and as a surgical adjunct^4,6,7,10,19. However, the HA used clinically shows considerable variation in MW and concentration, and there are few toxicity studies on corneal cells^11,18. Certain studies indicate that LMW HA may induce inflammatory responses, whereas HMW HA demonstrates anti-inflammatory properties²⁰. Additionally, the signaling activity of HA is affected by its molecular size, influencing receptor interactions^2,13,15,16.

This study revealed that LMW HA exhibited the highest toxicity across all corneal cell layers. At higher concentrations, LMW HA also induced noticeable morphological changes in corneal epithelial cells. Whether these changes are associated with mesenchymal transition remains to be elucidated through further investigation. Notably, corneal keratocytes demonstrated greater resistance to LMW HA-induced toxicity compared to epithelial and endothelial cells and maintained over 50% viability after 72 h exposure. Further, at higher concentrations, VLMW HA resulted in a lesser reduction of corneal cell viability compared to that caused by LMW HA. Intriguingly, an increase in corneal epithelial cell viability was noted with VLMW HA at elevated concentrations after 48 and 72 h of culture. The proliferative effect of HMW HA on corneal epithelial cells has been well-documented previously^4,12. However, the positive proliferative effect of VLMW HA has not been reported previously.

The varying toxicity profiles associated with different MW of HA require further elucidation through additional studies. While insights can be drawn from previous research, the interaction between HA and biological systems involves HA receptors such as cluster-determined 44 (CD44), receptor for hyaluronate-mediated motility (RHAMM), HA receptor for endocytosis (HARE), and lymphatic vessel endothelial hyaluronan receptor-1 (LYVE-1)¹⁹. Of these receptors, CD44 has been the most extensively studied. CD44 receptors were abundant in the corneal epithelium and present in keratocytes, but absent in the corneal endothelium^21–23.

The binding of HA to CD44 involves complex multivalent interactions influenced by the MW of HA, the number and density of CD44 receptors on the cell surface, and the activation state of CD44^16,19,24. HMW HA exhibits greater affinity for CD44 due to enhanced multivalent binding²⁴. CD44 activation can initiate signaling cascades that facilitate essential cellular functions such as proliferation, migration, angiogenesis, and wound healing^2,25. Consequently, different sizes of HA fragments may enhance or suppress these pathways by competing with endogenous HA for CD44 receptor binding. A recent study shows that HA with MW of 700 kDa and 25 kDa occupy approximately the same number of CD44 receptors⁶. Hence, additional factors beyond MW significantly influence receptor binding.

The findings of this study have significant implications for both the ophthalmic industry and clinical practice. HA plays a vital role in various formulations, including artificial tears, ocular surgery viscoelastics, and regenerative medicine treatments. Understanding its toxicity profile is crucial for optimizing application while minimizing adverse effects. Our data indicate that while both VLMW HA and HMW HA are relatively safe, LMW HA shows significant cytotoxicity, especially with prolonged exposure. Thus, ophthalmic formulations containing LMW HA should be used cautiously, limiting their concentration and assessing their safety rigorously before clinical use. Meanwhile, MMW HA exhibited dose-dependent toxicity, necessitating further exploration of its long-term implications for corneal health. Additionally, the observed mild increase in epithelial cell viability with VLMW HA underscores their potential benefits in promoting corneal health. Future research should delve into their mechanisms of action and potential therapeutic uses in detail.

While this study provides valuable insights into the toxicity profiles of HA on corneal cells, several limitations should be acknowledged. The in vitro nature of the experiments does not fully replicate the complexities of the human eye, where factors such as tear film composition and ocular surface interactions impact HA behavior. For topical ophthalmic application, HA-based formulations are further constrained by rapid clearance from the ocular surface due to blinking and lacrimation, with average retention times reported to be several minutes²⁶. Artificial tear formulations currently available on the market typically contain HA concentrations ranging from 0.1 to 0.3%²⁷. In contrast, ophthalmic viscoelastic devices (OVDs) used in cataract surgery incorporate substantially higher concentrations—commonly between 1 (e.g., Healon PRO, Johnson & Johnson Vision, Santa Ana, CA) and 3% (e.g., Healon EndoCoat, Johnson & Johnson Vision, Santa Ana, CA)—selected according to the surgeon’s preference for rheological properties such as cohesiveness and dispersiveness²⁸. The temporal exposure of corneal endothelial cells to OVDs is notably limited, given that cataract procedures are typically completed within 30 min. Additionally, standard surgical protocols include thorough irrigation and aspiration steps immediately prior to surgery completion, resulting in minimal residual HA in the anterior chamber. Furthermore, the study primarily focused on cell viability as an endpoint; therefore, further research into the molecular mechanisms of HA-induced toxicity is necessary. One limitation is the use of an immortalized cell line to model corneal endothelial cells, while epithelial cells and keratocytes were cultured as primary cells. In vitro, corneal endothelial cells are prone to mesenchymal transition, which complicates obtaining a sufficient number of early-passage endothelial cells for research purposes.

Future studies should utilize in vivo models to validate these findings and examine long-term effects. It is essential to understand how HA interacts with other ocular components, such as tear proteins and inflammatory mediators, to optimize ophthalmic formulations. Additionally, exploring strategies to mitigate HA-induced toxicity, such as modifying HA’s structure or incorporating protective agents, could enhance the safety of HA-based treatments.

Conclusion

This study demonstrated the significant impact of HA’s MW on corneal epithelial cells, keratocytes, and endothelial cells. These results underscore the necessity of meticulously formulating HA-containing ophthalmic products to ensure appropriate MW and concentration, thereby maximizing therapeutic benefits and minimizing adverse effects.

By delineating the toxicity profiles of different HA MW, our research contributes to a more comprehensive understanding of HA’s biocompatibility in corneal applications. As HA remains a vital component in ophthalmic treatments, ongoing research is crucial to refine its utilization and augment its safety profile for clinical applications.

Materials and methods

Hyaluronic acid

In this study, four different MW of HA were used to investigate toxicity in corneal cells: very low (VLMW, MW range from 8000 to 15,000 Da), low (LMW, MW range from 130,000 to 150,000 Da), medium (MMW, MW range from 750,000 to 1,000,000 Da), and high (HMW, MW range exceeding 1,000,000 Da). Details of each HA are presented in Table 1.

Table 1.

Characteristics of hyaluronic acid (HA) used in this study.

Category	Molecular weight range	Manufacturer	Catalog number	Location	Stock state
Very low molecular weight (VLMW)	8000–15,000	Sigma-Aldrich	40,583	St. Louis, MO, USA	Powder
Low molecular weight (LMW)	130,000–150,000	Sigma-Aldrich	75,043	St. Louis, MO, USA	Powder
Medium molecular weight (MMW)	750,000–1,000,000	Sigma-Aldrich	53,163	St. Louis, MO, USA	Powder
High molecular weight (HMW)	> 1,000,000	Johnson & Johnson surgical vision, Inc	10,310,012	Santa Ana, CA, USA	Gel

Cornea cell culture

Epithelial cell culture

The primary culture of human corneal epithelial cells (catalog no. PCS-700–010) was sourced from the American Type Culture Collection (ATCC, Manassas, VA, USA). The cells were resuspended in serum- and calcium-free corneal epithelial basal medium, which was supplemented with a growth kit provided by ATCC. Subsequently, they were seeded into 75-cm² tissue culture flasks that had been pre-coated with an FNC coating mix (Athena Enzyme Systems, cat. 0407, Baltimore, MD, USA) and were maintained at 37 °C in a humidified atmosphere consisting of 5% CO₂ and 95% air. The culture medium was refreshed every three days, and cells were subcultured using 0.05% Trypsin–EDTA (GibcoBRL, Grand Island, NY, USA). For this study, only cells at a passage number of ≤ 5 were used.

Keratocyte culture

The primary culture of human keratocytes was established from a research cornea donated to Eversight Eyebank (Ohio, USA, Tissue number: W403424164196, 51 years old female donor). Descemet’s membrane and the epithelium were meticulously removed with forceps and an ophthalmic knife, followed by the mincing of stromal tissue in a laminar flow hood. Mid-stroma and posterior stroma explants were then suspended in culture medium and plated in 24-well plates. The corneal stroma was quartered and subjected to enzymatic digestion overnight at 37 °C using 2.0 mg/mL collagenase (Roche, Basel, Switzerland) and 0.5 mg/mL hyaluronidase (Worthington Biochemicals, Lakewood, NJ, USA) in DMEM. Isolated cells were then washed with DMEM and cultured in DMEM/F12 supplemented with 10% fetal bovine serum (FBS; Gibco-Invitrogen, Grand Island, NY, USA). The cells were maintained on tissue culture-treated plastic at a density of 4 × 10⁴ cells/cm² until they reached confluency. They were then harvested, resuspended in culture medium, and plated in 75-cm² tissue flasks. Cultures were kept at 37 °C in a humidified atmosphere of 5% CO₂ and 95% air, with medium changes every three days. For subculturing, 0.25% Trypsin–EDTA (Gibco BRL, Carlsbad, CA, USA) was utilized. Only cells with a passage number of ≤ 7 were utilized for this study.

Endothelial cell culture

The established human corneal endothelial cell line, B4G12 cells (Cat no. CSC-C3457-CRA, Shirley, NY, USA), was sourced from Creative Bioarray (Shirley, NY, USA). These cells were cultured in the recommended medium, which comprises human endothelial serum-free medium (Creative Bioarray, Cat no. CM-345L7, Shirley, NY, USA) and 10 ng/ml of fibroblast growth factor-2 (Creative Bioarray, Cat no. CSC-CTK0134, Shirley, NY, USA). The culture medium was refreshed every three days, and cells were passaged using 0.25% Trypsin–EDTA (Gibco BRL, Carlsbad, CA, USA).

Viability assay

The assessment of corneal cell viability was conducted using the Cell Counting Kit-8 (CCK-8) (Dojindo Laboratories, Kumamoto, Japan). Corneal epithelial cells, keratocytes, and endothelial cells were seeded into 96-well plates at a density of 10,000 cells (epithelial cells), 5000 cells (keratocytes), and 20,000 cells (endothelial cells) per well. Following cell adhesion for 24 h, the cultures were treated with hyaluronic acid (HA) of varying MW—very low (VLMW), low (LMW), medium (MMW), and high (HMW)—at concentrations ranging from 0.025 to 0.5% over periods of 24, 48, and 72 h. HA is diluted in culture medium for each type of corneal cells as previously described. Triton X-100 (TX-100), non-ionic detergent, was used as the positive control of cytotoxicity. After the incubation period, 10 μL of CCK-8 reagent was added to each well. After a further 4-h incubation at 37 °C, the absorbance at 450 nm was measured using a microplate reader.

Live and dead cell staining

To evaluate the qualitative aspects of cell viability and cytotoxicity, a live/dead viability/cytotoxicity kit (Molecular Probes, Cat. L3224, Thermo Fisher Scientific, Rochester, NY, USA) was utilized as described in the previous study²⁹. Corneal epithelial cells, keratocytes, and endothelial cells were cultured in confocal imaging dishes and treated with hyaluronic acid (HA) at concentrations of 0.1% and 0.5% for 24 h. Prior to staining, cells were gently washed twice with Dulbecco’s phosphate-buffered saline (DPBS). Staining solutions were prepared in DPBS using calcein AM (2 μM final concentration) to label live cells and ethidium homodimer-1 (EthD-1, 4 μM final concentration) to identify dead cells, in accordance with the manufacturer’s instructions. The cells were incubated with the staining solution at 37 °C for 30 min in the dark to prevent photobleaching. After incubation, excess stain was removed with additional DPBS washes, and fluorescence imaging was conducted using a confocal live-cell imaging system (Leica Microsystems CMS GmbH, Mannheim, Germany).

Statistical analysis

The data were presented as the mean ± standard error. Statistical significance was determined using analysis of variance (ANOVA) followed by Dunnett’s multiple comparison test. P values less than 0.05 were considered statistically significant. Analyses were performed using GraphPad Prism Ver. 9.3.1 (GraphPad Software Inc., La Jolla, CA, USA).

Author contributions

J.P.: data curation, formal analysis, investigation, methodology, project administration, visualization, writing—original draft; C.Y.P.: conceptualization, data curation, funding acquisition, formal analysis, investigation, validation, visualization, writing—review and editing. All authors have read and agreed to the published version of the manuscript.

Funding

The publication of this article was supported by a grant from the Korea Health Technology R&D Project through the Korea Health Industry Development Institute (KHIDI), funded by the Ministry of Health & Welfare, Republic of Korea (RS-2023-KH135936).

Data availability

The original contributions presented in this study are included in the article. Further inquiries can be directed to the corresponding author.

Declarations

Competing interests

The authors declare no competing interests.