Polygalacturonic Acid Partially Inhibits Matrix Metalloproteinases and Dehydration in Wounds

The inflammatory cells in chronic wounds secrete cytokines and proteases, which impair normal healing processes. Proteases that notably present at abnormally high levels in chronic nonhealing wounds are MMPs such as MMP-2 and MMP-9, and NE.¹ Although some protease-driven granulation tissue remodeling is desirable, prolonged exposure to high levels of MMPs and NE destroys newly deposited granulation tissue, preventing maturation and ultimately reepithelialization and wound closure.² Key factors in enabling wound beds to develop mature granulation tissue and to reepithelialize include reducing chronic inflammation by eradicating microbial biofilm and maintaining a local environment of optimal moisturization, pH, and reduced protease activity.

Wound ointments and dressings have been developed to facilitate healing of chronic wounds. Previously, the authors of the current study reported the ability of a combination of caprylic acid, derived from coconut and other natural oils, and PG, derived from citrus or other fruit pectins, to eradicate a broad range of microbial biofilm with greater efficacy and less cytotoxicity (inflammation) than commonly used antiseptic ointments (ie, benzalkonium chloride in polyhexamethyl biguanide–based ointments) or naturally derived wound ointments (honey).³ The authors also previously showed that PG enhances the bioavailability and produces a pH (~4.5) that is optimal for both antimicrobial activity of caprylic acid and wound healing.^4,5 PG has been shown to have other physical and chemical utilities, such as molecular binding and gelation.^6-8 Therefore, in the current study, the authors evaluated PG’s ability to bind and inhibit wound proteases, such as MMP-2, MMP-9, and NE, as well as its ability to inhibit dehydration in a model where evaporative water loss occurs. 

PG was purchased (Sigma Aldrich) and tested as an inhibitor of MMP-2 and MMP-9 enzymes using Abcam colorimetric MMP-2 (ab 139446) and colorimetric MMP-9 (ab 139448) inhibitor screening assay kits.^9,10 PG was prepared at various concentrations (0.4%, 1%, and 2%). PG was fully dissolved in deionized water at pH 4.25 by slowly adding ammonium hydroxide (dropwise), stirring, and filtering through a 0.22-nm cellulose acetate filter to achieve a clear, non-turbid solution. The proprietary colorimetric inhibitor screening assay kits are designed to screen MMP-2 and MMP-9 enzyme inhibitors using a thiopeptide as a chromogenic substrate. The MMP cleavage site peptide bond is replaced by a thioester bond in the thiopeptide substrate. Hydrolysis of this bond by an MMP produces a sulfhydryl group, which reacts with the substrate DTNB (Ellman reagent) to form 2-nitro-5-thiobenzoic acid, whose concentration can be measured by absorbance at 412 nm.^9,10

To measure inhibition by PG, the MMP substrate, inhibitor, and enzymes (MMP-2 and MMP-9) were diluted as instructed in the assay kits.^9,10 Per kit instructions, the diluted solutions and PG solutions at various concentrations were brought to 37°C, then assay buffer was pipetted into designated wells of a microplate. The microplate was allowed to equilibrate to 37°C. Diluted MMP-2 and MMP-9 were added to the negative control, MMP inhibitor, and PG test wells (not to the blank). Then, diluted MMP inhibitor and PG solutions were added to the appropriate wells of the microplate. The positive controls were reported in the kit instructions to inhibit 94% of MMP-2 and 100% of MMP-9. The microplate was incubated for 30 minutes at 37°C to allow inhibitor–enzyme interaction. The reaction was then initiated by the addition of MMP substrate at 37°C. Absorbance at 412 nm was continuously read in a microplate reader (Fluostar Omega; BMG Labtech). Data (optical density at 412 nm) were recorded each minute for 10 minutes of total elapsed reaction time. Three replicate plates were run for each MMP, over the range of PG concentrations.

PG inhibition of NE enzyme was tested using a purchased Neutrophil Elastase inhibition assay kit (BML-AK497-ENZO; Enzo Biochem). PG was prepared as for the MMP inhibitory assay kits. The NE inhibitor kit uses the chromogenic substrate MeOSuc-Ala-Ala-Pro-Val-pNA; cleavage of pNA from the substrate by NE increases absorbance at 405 nm. Elastinal (a known NE inhibitor supplied by the kit) was used as the positive control. Briefly, NE substrate, inhibitor, and enzyme were diluted per kit instructions.¹¹ Diluted substrate solutions and PG solutions at various concentrations were brought to 37°C. Assay buffer was pipetted into microplate wells per instructions and allowed to equilibrate to 37°C. NE enzyme was added to appropriate wells. The plate was incubated for 30 minutes at 37°C to allow inhibitor–enzyme interaction. The reaction was initiated by the addition of BML-P213-9090 diluted substrate at 37°C. Absorbance (optical density) was continuously read at 405 nm in a microplate reader and data were recorded each minute for 10 minutes of total elapsed reaction time. Three replicate plates were run for NE, over the range of PG concentrations. 

PG inhibition of MMPs and NE (%) was calculated per kit manual instructions. The slope of optical density increase over 10 minutes was computed by linear regression and compared with the slope and percent inhibition of the positive control. Percent inhibition for PG solutions was then calculated from the percentage change in slope differences between PG-containing solution (PG slope to negative control slope) and the positive control (positive control slope to negative control slope) multiplied by the known percent inhibition of the positive control. Linear regression analysis was performed using Microsoft Excel (Microsoft Corporation).

Transepidermal evaporative water loss from a polyvinyl alcohol skin substitute hydrogel was gravimetrically measured. Crosslinked polyvinyl alcohol hydrogel was used as a skin substitute (Chill Pal). Briefly, 5-cm squares of hydrogel were cut and dried for 24 hours, then weighed to determine the dry mass. The hydrogel squares were then soaked in deionized water, blotted dry, and reweighed until saturated with water. The external surfaces of the saturated hydrogel squares were then dip-coated in deionized water (0% PG), 1% PG, or 2% PG solutions. The hydrogel squares were then placed in a chemical hood and reweighed periodically to record mass of water evaporated over time. The total mass of water that could evaporate was calculated as the difference between dry weight and initial coated weight. Percent evaporation was computed from the mass of water evaporated relative to the total mass of water that could potentially evaporate (initial coated weight less the dry weight). Data for triplicate runs of each PG concentration were collected. Data analysis (linear regression, mean, and standard deviation calculations) was performed using the aforementioned spreadsheet software program. The decrease in evaporative water loss due to PG was computed as the difference in percent water loss for the PG-coated samples and non-PG control, divided by the non-PG control.

The ability of PG at various concentrations to inhibit MMP-2 is summarized in Table 1, and its ability to inhibit MMP-9 is shown in Table 2.

Slopes of the change in optical density were calculated by linear regression, and both slopes and correlation coefficient (r²) for goodness of fit are reported (where r² = 1 denotes a perfect linear fit). PG-induced inhibition of both MMPs increased with higher concentrations of PG; at 2% PG concentration, MMP inhibition was greater than 60% for all replicates. The ability of PG to inhibit NE is shown in Table 3, where slopes and correlation coefficients similar to Table 1 and Table 2 from linear regression analyses are reported. PG-induced inhibition of NE increased with higher concentrations of PG; at 2% PG concentration, NE inhibition was greater than 25% for all replicates.

Evaporative water loss over time for each replicate at various PG concentrations is shown in Table 4. The slope of mass of water changes due to evaporation vs time (in 100-minute increments) by linear regression analysis was reported as the correlation coefficient. PG reduced the rate of evaporative water loss by 26.3% at 1% PG concentration and by 42.1% at 2% PG concentration vs the rate of evaporative water loss without PG.

Previous studies have shown that the combination of caprylic acid and PG was effective in inhibiting and eradicating microbial biofilm, which is known to play a role in prolonged inflammatory response and impaired wound healing.⁵ PG, in addition to enhancing the bioavailability of caprylic acid, which has limited aqueous solubility, maintains a pH below the pKa of caprylic acid (pH 4.8), which is essential to caprylic acid’s antimicrobial activity. Elevated pH is also a hallmark of wound infection. Intact skin has an acidic pH (4-6), which is believed to be important for reducing the potential for microbial ingress and infection. Studies have shown that acidic moieties such as honey and citric acid can improve wound healing by maintaining an acidic environment in the wound bed similar to that of intact skin. PG also serves to regulate pH in the wound bed so that it remains in this acidic range.^12,13 The results of the current study suggest that PG may have the potential to aid the transition from inflammatory to proliferative phase by partially inhibiting MMPs and NE. This requires further in vivo study verification. Although some protease-mediated remodeling is desirable, excessive protease activity can stall the proliferative phase by degrading newly deposited extracellular matrix. Partial inhibition of these key proteases by PG therefore may potentially aid in preserving a desirable balance of some remodeling without excessive tissue destruction. This benefit also requires additional in vivo study verification.

In the current study, PG was noticeably more potent at inhibiting MMP-2 and MMP-9 at concentrations that could be used in ointments and dressings (at 2% concentration, PG could inhibit >60% protease activity). PG was a less potent inhibitor of NE but did generate some inhibition. MMP-2 and MMP-9 have zinc +2 coordination centers. Pectins with low esterification degrees (eg, PG) are known to have an affinity for zinc ions in acidic solutions.¹⁴ This could explain the potency difference, because NE is not a metalloenzyme but rather a basic serine protease whose optimal activity is at alkaline pH.¹⁵ PG’s more limited inhibition of NE may be a consequence of the acidity of PG rather than affinity chelation of zinc +2 in the MMPs.

Maintaining moisture balance has been recognized as a key factor for efficiently healing wounds.¹⁶ Maintaining a moist environment has been shown to be optimal for accelerating wound healing.¹⁷ Moist environments promote reepithelialization of wounds and reduce scarring, whereas dry environments promote inflammation, which is detrimental to healing. However, excess moisture in wounds can delay healing and can cause skin breakdown.¹⁸ The results of the current study indicate that PG can contribute to maintaining a healthy moisture balance by providing a barrier that substantially reduces evaporation rates. PG is known to be a gelling agent, and this property of moisture retention may explain its ability to assist in reducing evaporative water loss.   

This study has limitations. To isolate the interaction between PG and specific proteases, in vitro testing was performed, which did not encompass the full complexity of a wound bed. Extrapolation to performance in wounds requires further in vivo study, and extrapolation to clinical benefit requires further clinical testing.

PG is a versatile natural agent for potential use in wound healing. In addition to maintaining an optimal pH, PG might help in regulating wound environments to promote optimal healing. Its benefits potentially include partially inhibiting key proteases to maintain a balance between tissue remodeling and destruction of new granulation tissue, as well as helping to maintain optimal wound moisture levels by preventing evaporative water loss from wounds. 

Authors: Bahgat Z. Gerges, MD; Joel Rosenblatt, PhD; Y-Lan Truong, BS; and Issam I. Raad, MD

Affiliation: Department of Infectious Diseases, Infection Control and Employee Health, The University of Texas MD Anderson Cancer Center, Houston, TX

Acknowledgments: The authors thank Ms Sally Saxton for her help in submitting this manuscript, and Ms Erica Goodoff in the Research Medical Library at The University of Texas MD Anderson Cancer Center for editing the manuscript.

Disclosure: Drs Raad and Rosenblatt are co-inventors of this technology, which is owned by The University of Texas MD Anderson Cancer Center. The other authors have no competing interests. All authors approve the submission. This study was supported by institutional funds.

Correspondence: Bahgat Z. Gerges, MD; Department of Infectious Diseases, Infection Control and Employee Health, The University of Texas MD Anderson Cancer Center, 1515 Holcomb Blvd, Houston, TX 77030; BZGerges@mdanderson.org

Manuscript Accepted: May 24, 2024