Anti-biofilm efficacy of a medieval treatment for bacterial infection requires the combination of multiple ingredients

Jessica Furner-Pardoe, Blessing O Anonye, Ricky Cain, John Moat, Catherine A. Ortori, Christina Lee, David A. Barrett, Christophe Corre, Freya Harrison

Posted on: 26 May 2020

Preprint posted on 27 April 2020

Article now published in Scientific Reports at


The safety profile of Bald’s eyesalve for the treatment of bacterial infections

Blessing O Anonye, Valentine Nweke, Jessica Furner-Pardoe, Rebecca Gabrilska, Afshan Rafiq, Faith Ukachukwu, Julie Bruce, Christina Lee, Meera Unnikrishnan, Kendra P. Rumbaugh, Lori AS Snyder, Freya Harrison

Posted on:

Preprint posted on 24 April 2020

Article now published in Scientific Reports at

Modern problems, medieval solutions? A study of the anti-biofilm properties and safety profile of a medieval treatment for bacterial infections

Selected by Snehal Kadam

Categories: microbiology

Context and background: The number of deaths worldwide attributed to antimicrobial resistance is increasing and is predicted to continue with the rise in multidrug resistant pathogens and the lack of new antimicrobial therapeutics [1]. One of the major issues with treating bacterial infections is the formation of biofilms. Biofilms are 3D structures formed by aggregates of bacteria, with an extracellular matrix surrounding them and making them highly resistant to antibiotics [2]. Interactions with host cells, changes in growth rates and the presence of persister cells can make clearing biofilms from infections even more difficult. Biofilms have been implicated as the major cause of various chronic infections, including wounds [3]. A recent push in the field has been towards rediscovering medicinal plant-based formulations from historical texts to identify anti-biofilm treatments. Traditional formulations usually combine various plants and plant extracts as opposed to identifying single compounds with antimicrobial activity. Another important aspect is to test against biofilms, and not only planktonic cultures.

Two recent preprints look at a medieval treatment, called Bald’s eyesalve, traditionally used for eye infections, against a range of pathogen biofilms and the safety profile of the formulation using various in vitro cytotoxicity tests and in vivo tests on mice.


Experimental setup: Bald’s eyesalve was prepared by mixing garlic, onion, wine and bile salts. Multiple batches were also made and tested. The effects of individual components were also tested by using ‘dropout’ batches, where one ingredient was left out from the formulation. The highest concentration of Bald’s eyesalve tested was 33%. The remedy and its batches were tested against planktonic cultures in 96-well plates and Colony Forming Units (CFU) were determined to obtain MIC values. The effect of the treatment on biofilms was tested using a previously described in vitro collagen-based wound model. The biofilm eradicated was determined by dissolving the wound matrix, serial dilutions and CFU counts. In order to understand the safety of use for this remedy, several tests were carried out. The remedy was tested for toxicity against two cell lines (representing skin and immune cells) using alamarBlueTM and lactate dehydrogenase (LDH) release as readouts. Bovine corneal opacity and permeability assays, slug mucosal assays and finally, in vivo tests in mice were also carried out.


Important Results:

Bald’s eyesalve shows antimicrobial activity and anti-biofilm activity against common Gram-negative and Gram-positive wound pathogens

The remedy was seen to eradicate planktonic cultures of P. aeruginosa, A. baumannii, E. cloacae, S. aureus Newman, S. epidermidis and S. pyogenes. S. maltophilia and S. aureus USA300 did not show complete eradication, but a 3-4 log drop in CFU. When tested against mature biofilms, the eradication was strain dependent. A 2-6 log drop in CFU was observed for S. aureus Newman, S. aureus USA300, S. epidermidis, S. pyogenes and A. baumannii, whereas no effect was seen for the other strains. The remedy was also tested against an eye pathogen, N. gonorrhoeae. Disk diffusion assays showed significant zones of inhibition and planktonic cultures showed 7-log reductions in CFU.


Garlic alone contributes to the antimicrobial activity in planktonic cultures, but is insufficient for anti-biofilm activity

A component of garlic, allicin, has been shown to have antimicrobial activity. To test the contributions of garlic, individual components of the remedy were evaluated for their MIC against planktonic cultures of four bacterial strains. Individually, wine, onion or bile had less effect than the whole remedy. The MIC of garlic alone was similar to the whole remedy, indicating that garlic contributed to the antimicrobial activity against planktonic cultures. Similarly, dropout batches of the remedy omitting garlic lost their ability to kill planktonic cells.

The effects of individual components of the remedy against S. aureus Newman biofilms was also tested. Unlike planktonic cultures, biofilms required the whole remedy to show biofilm killing, where omission of a single ingredient resulted in loss of anti-biofilm activity.


Multiple models, in vitro, ex vivo and in vivo, indicate low toxicity, no inflammation and irritation from the remedy

This study makes use of multiple models to test the safety profile of Bald’s eyesalve. The in vitro assay detecting LDH levels indicate the remedy is less toxic to human skin and immune cells as compared to chloramphenicol eye drops (available on the market without prescription). However, the alamarBlueTM assay did not show any difference between Bald’s eyesalve and chloramphenicol treated cells. An ex vivo test using bovine eyes showed mild irritation to the cornea, seen as a change in opacity, which resolved within ten minutes. Comparison with a positive control indicated that different batches of the remedy caused no irritation. A slug mucosal test for irritation also showed only mild irritation for the remedy, and significantly lower mucosal secretion by the slugs as compared to a positive control. Finally, in vivo studies with mice showed that the treatment caused no differences in wound healing, with no visible signs of inflammation and irritation, indicating the safety of the remedy with a live vertebrate.


Interesting aspects of the study: These two studies address various aspects of Bald’s eyesalve in treating bacterial infections and its safety as a remedy. Despite it being traditionally used to treat eye infections, the ability of Bald’s eyesalve to eradicate biofilms by wound pathogens indicates its possible use in more infections. With a push to reduce animal testing, I appreciate the authors’ approach of testing the safety profile of the remedy and choosing to move to animal models only after multiple other in vitro and ex vivo tests showed promise. The systematic approach adopted here was one of the highlights of the study for me. The tests to confirm anti-biofilm activity as well as multiple tests to confirm safety give a strong case for the remedy to be tested further as a therapeutic.


Questions for the authors:

  1. The absence of killing by Bald’s eyesalve on  aeruginosa, E. cloacaeand S. maltophilia biofilms was observed using 33% of Bald’s eyesalve. Is it possible that higher concentrations may have an effect and was that tested?
  2. In testing the effect of Bald’s eyesalve on gonorrhoeae, biofilms in particular were not tested. Was there a specific reason for this? Is the incidence of biofilm formation by N. gonorrhoeae in neonatal conjunctivitis low?
  3. The cytotoxicity testing using alamarBlue and LDH level tests showed a discrepancy in the cytotoxicity. For alamarBlue assays, the chloramphenicol treated cells were consistently more viable as compared to the same Bald’s eyesalve treatment. The LDH assays however, showed more LDH release (and hence more plasma membrane rupture) in the diluted chloramphenicol treated cells compared to Bald’s eyesalve. This means by one assay (alamarBlue), the Bald’s eyesalve was more cytotoxic than the chloramphenicol treatment, but vice versa in the other assay (LDH). Is there a particular reason for the difference in the assays?


References/Further Reading:

[1] Review on Antimicrobial Resistance. Tackling drug-resistant infections globally: final report and recommendations. Review on antimicrobial resistance, 2016.

[2] Flemming, Hans-Curt, et al. “Biofilms: an emergent form of bacterial life.” Nature Reviews Microbiology 14.9 (2016): 563.

[3] Percival, Steven L., et al. “A review of the scientific evidence for biofilms in wounds.” Wound repair and regeneration 20.5 (2012): 647-657.

[4] Benkeblia, N. “Antimicrobial activity of essential oil extracts of various onions (Allium cepa) and garlic (Allium sativum).” LWT-food science and technology 37.2 (2004): 263-268.

[5] Yuan, Haidan, et al. “The traditional medicine and modern medicine from natural products.” Molecules 21.5 (2016): 559.

[6] Sun, Fengjun, et al. “Biofilm-associated infections: antibiotic resistance and novel therapeutic strategies.” Future Microbiology 8.7 (2013): 877-886.

Tags: antibiotic alternatives, antimicrobial, bacterial infections, biofilm, wound infection


(1 votes)

Author's response

Blessing O Anonye, Lori AS Snyder, Freya Harrison shared about The safety profile of Bald’s eyesalve for the treatment of bacterial infections

Author’s Response to Question 1: When we started this project, we had to make a judgement call about the amount of eyesalve we could add to the wounds and fit in the wells of the microtitre plates we make them in. When we initially tried this with S. aureus in our early work, using 0.5 volumes of eyesalve, and it killed pretty much all of the bacteria in the biofilms, we decided to stick with this as a tractable volume to work with and a handy benchmark of activity – if a batch of the eyesalve causes ≥3 log-fold killing of S. aureus Newman in this experimental set-up we count it as “active” for our purposes. It is possible that if we could prepare a higher concentration of active compounds within the eyesalve – or set up experiments where we repeatedly add fresh doses of eyesalve to biofilms – we might see killing of these other species. Our focus now is on identifying the key biologically-active compounds within the eyesalve, and then we can conduct much more controlled and standardised MBEC testing with different species and strains. But I think that the massive variability we see in how well different species and strains make biofilms, and the differences in the composition of the biofilm matrix, may mean that some strains are just impervious to the eyesalve. If the active components can’t penetrate some biofilm matrices due to their size or charge, then simply increasing the concentration is unlikely to achieve results. For bacteria where we see this protective power of the biofilm, then the most sensible route to go down to render them treatable would be to look for adjuvants which help antibacterial compounds get through the matrix.


Author’s Response to Question 2: We did not look at them in biofilms. These species can form biofilms, but their role in eye infections is not known. Biofilms do not appear to play a major role in the antimicrobial resistance issues associated with gonorrhoea infections; isolates from patients retain their resistance when grown on plates in the lab. In neonatal conjunctivitis with N. gonorrhoeae, ophthalmia neonatorum, the infection progresses rapidly, therefore prompt treatment with an antimicrobial that is able to eliminate the bacteria before they can cause damage and potentially perforate the globe of the eye is essential. This is why we are especially interested in antimicrobials such as Bald’s eyesalve.


Author’s Response to Question 3: This is an interesting observation and likely due to differences in the mechanism or specification of the assays. Alamar blue uses a redox indicator to determine proliferation of cells growing in culture with a change from the oxidized (non-viable, blue) to the reduced forms (viable cells, pink). This change in the colour of the media is caused by the metabolic activity of the growing cells in culture and the cells must be actively growing for this change to occur. Like you rightly mentioned, lactate dehydrogenase (LDH) measures the plasma membrane damage of the cell. This assay is performed using the cell culture media that the treated cells have been grown in. The catalysis of lactate to pyruvate by LDH (a cytosolic enzyme), through NAD+ reduction to NADH is first step in this process. This is followed by the oxidation of NADH by diaphorase to the reduction of a tetrazolium salt. This leads to the red formazan product that is measured indicating the cytotoxicity of the treatment.  We used these cell culture assays as an initial screen and followed up with more robust models as seen in the paper.

Have your say

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Sign up to customise the site to your preferences and to receive alerts

Register here

Also in the microbiology category:

Characterization of natural product inhibitors of quorum sensing in Pseudomonas aeruginosa reveals competitive inhibition of RhlR by ortho-vanillin

Kathryn E. Woods, Sana Akhter, Blanca Rodriguez, et al.

Selected by 22 May 2024

UofA IMB565 et al.


Feedback loop regulation between viperin and viral hemorrhagic septicemia virus through competing protein degradation pathways

Xiaobing Lu, Meisheng Yi, Zhe Hu, et al.

Selected by 21 May 2024

UofA IMB565 et al.


Lytic bacteriophages interact with respiratory epithelial cells and induce the secretion of antiviral and proinflammatory cytokines

Paula F. Zamora, Thomas G. Reidy, Catherine R. Armbruster, et al.

Selected by 20 May 2024

UofA IMB565 et al.