key: cord-0304150-k77lcvcw
authors: Harrison, Freya; Furner-Pardoe, Jessica; Connelly, Erin
title: An assessment of the evidence for antibacterial activity of stinging nettle (Urtica dioica) extracts
date: 2021-11-17
journal: bioRxiv
DOI: 10.1101/2021.11.17.468953
sha: b59470df5db008e4e3c14338219621697913e204
doc_id: 304150
cord_uid: k77lcvcw

Stinging nettles (Urtica spp.) have been used in a diverse range of traditional and historical medicines from around the world for the treatment of skin diseases, wounds, urinary disorders, respiratory diseases, bone and joint pain, anaemia and other circulatory problems, as well as in cosmetic preparations for skin and haircare. As part of an interdisciplinary exploration of nettle-based remedies, we performed a systematic review of published evidence for the antimicrobial activity of Urtica spp. extracts against bacteria and fungi that commonly cause skin, soft tissue and respiratory infections. We focussed on studies in which minimum inhibitory concentration (MIC) assays of U. dioica were conducted on the common bacterial opportunistic pathogens Escherichia coli, Pseudomonas aeruginosa, Klebsiella pneumoniae and Staphylococcus aureus. No studies used fresh leaves (all were dried prior to use), and no studies prepared nettles in weak acid (corresponding to vinegar) or in fats/oils, which are common combinations in historical and traditional preparations. We addressed this gap by conducting new antibacterial tests of extracts of fresh U. dioica leaves prepared in vinegar, butter or olive oil against P. aeruginosa and S. aureus. Our systematic review and additional experimental data leads us to conclude that there is no strong evidence for nettles containing molecules with clinically useful antimicrobial activity. It seems most likely that the utility of nettles in traditional topical preparations for wounds may simply be as a “safe” absorbent medium for keeping antibacterial (vinegar) or emollient (oils) ingredients at the treatment site. Data Summary All data relating to the systematic review are supplied in the Datasets and Supplementary tables contained in the Supporting Data file.

7 of 45 µg.ml -1 . The plates were shaken for 5 minutes on an orbital microplate shaker to ensure the resazurin dispersed in the wells containing oil or butter, then wrapped in foil to exclude light and incubated at room temperature for two hours. The MIC of each test substance was scored as the lowest concentration at which reduction of blue resazurin to pink resorufin was not observed. The following day, MBC plates were inspected and the MBC of each treatment was determined as the lowest concentration at which no colonies were observed.

Neither the LB plates, nor the selective plates, used for MBC determination showed any visible growth of organisms other than either P. aeruginosa or S. aureus.

Pilot experiments were conducted prior to this work to assess the MICs and MBCs of several types of vinegar and olive oil to see if these varied (supplementary tables below). Red wine vinegar, white wine vinegar and cider vinegar all had low MICs/MBCs for the test species, and these were comparable between species and with MIC/MBC of 6% acetic acid, although the cider vinegar showed some variability in MIC/MBC between replicates (Table S1) .

We chose to work with red wine vinegar for further work. Three brands of olive oil, produced in Spain (La Española), Crete (Olive Branch) and Italy (Garofalo), all had MICs and MBC of >50%, which was the highest concentration tested (Table S2) . We randomly selected Olive Branch olive oil for use in further experiments.

U. dioica leaves were immediately transferred to the laboratory and chopped roughly using a sterilised knife and chopping board. 5 g aliquots of chopped leaves were placed into 2 sterile 50 ml screw-cap tubes. 5 ml red wine vinegar (Waitrose Essentials) was added to one tube, and 5 ml sterile water to the other. For this experiment, homogenisation was carried out by hand by pounding in a sterile pestle and mortar for two minutes. Mid-log phase (6h) cultures of S. aureus Newman and P. aeruginosa PA14, grown in caMHB at 37°C, were divided into 200 μl aliquots in a 96-well microplate (Corning Costar). Aliquots of approx. 100 µl (judged by eye) of soaked leaf material was removed from each preparation of pounded nettle leaves using sterile forceps, and transferred to triplicate wells of each bacterial culture. The material was fully submerged by the culture. Further triplicate cultures were treated by adding 100 µl of either water or vinegar as negative and positive controls, respectively. The culture plates were then incubated at 37°C for 18 hours, during which time untreated bacterial cultures are expected to reach stationary phase. At the end of the incubation period, aliquots of each culture were removed, serially diluted and plated on LB agar plates. These were incubated overnight at 37°C and the number of colony-forming units (CFU) present in each culture well was calculated.

A systematic review of published articles was performed in order to find primary experimental literature on the antimicrobial activity of Urtica extracts, as measured by the MIC or ZOI of the extract when applied to human-pathogenic bacteria or fungi, plus the model Gram-positive bacterium Bacillus subtilis as this is a commonly-used model organism for testing of antibacterial agents. The PRISMA diagram in Figure 1 and associated data in Datasets   1,2 and Tables S3-S5 summarise the systematic review process which led to the extraction of data from 42 publications with ZOI data and 38 publications with MIC data in our final database of eligible articles. The ZOI dataset contained many publications with missing data or low-quality data (e.g. no negative controls reported for assays; a lack of clarity about the solvent used for the final testing of extracts; unclear if assays were replicated) and a wide variety of test methods (disk diffusion and well diffusion, using different diameters of wells or disks and different thicknesses of agar; in many studies, information was lacking about well/disk diameter or agar thickness), which made it impossible to make any meaningful comparisons across studies. We thus performed no further analysis of ZOI data. Summary information and quality assessment for 38 studies containing MIC data is shown in Table 1 .

As detailed in Table 1 , there are clear gaps in the reporting of experiments and heterogeneity in the methods used.

One study did not specify the method used to calculate the MIC and another used a non-standard version of the microdilution method; these two studies were both of U. dioica extracts and were excluded from further consideration. Of the remaining 36 studies, 26 used broth microdilution, 3 used broth macrodilution, one used an unspecified broth dilution method, two used agar well diffusion and four used agar dilution. Broth microdilution was the most widely used method, and is a standard diagnostic assay around the world. Further, an issue with agarbased methods is that some compounds cannot diffuse through the agar and so these assays are more likely than broth dilutions to give false negatives. We therefore decided to include only broth microdilution studies for further consideration. Four of the 26 microdilution studies did not specify what growth medium was used (although one of these states that NCCLS/CLSI guidelines were followed, which implies the use of Müller-Hinton broth) (16). The rest of the studies employing microdilution used a range of media, which could influence the results obtained (17).

Studies were also excluded if any essential information was not reported (Urtica species, part of plant used, growth medium used for MIC assays, maximum concentration used in MIC assay where "no inhibition" was reported), or if the paper contained any conflicting or unclear data (examples included MIC values of zero, MIC values reported differently in data tables vs. text, and MICs given as masses not concentrations). In many cases, studies used an inappropriate negative control for MIC assays (i.e. not the vehicle in which the extract was dissolved), or did not report a negative control. However, we decided to include studies with this problem as long as they met other quality criteria, because if an extract has a MIC that is sufficiently low to be clinically meaningful, then any solvent present should be sufficiently dilute to have a negligible or no effect on bacterial growth. This process resulted in 17 studies being classified as of sufficient quality for inclusion in our review.

Of the 17 studies classified as of sufficient quality for inclusion, 15 tested extracts of U. dioica; as only four studies tested another species (two for U. urens and two for U. pilulifera; these last also tested U. dioica extracts) we focussed only on studies of U. dioica extracts. All of these used dried or freeze-dried plant material to make their extracts (or, in one case, seeds), but the studies used a variety of solvents and extraction techniques, plus a variety identified as being of sufficient quality for further analysis (see Table 1 ). Note that many studies tested more than one strain 2 or isolate of the same species. S. aureus, extracted from studies described in Table 1 ; raw data are provided in Table S4 . Each circle denotes the MIC value for 10 an individual bacterial strain or isolate, obtained from one study. Triangles denote where the MIC is greater than the highest 11 concentration tested, the MIC is therefore presented as the highest concentration tested. The x axis records the type of solvent (water, methanol, ethanol, dichloromethane, chloroform, hexane) used to make the extract (ext.), or the class of compounds 13 extracted (essential oils, oils). Note that exact extraction methods varied between studies. The x axis is ordered with more 14 hydrophilic/lipophobic extracts on the left, and more hydrophobic/lipophilic extracts on the right. The same data are presented 15 again with different colour coding in Figure S1 to show the different strains or isolates of each bacterial species used, and in 16 Figure S2 to show which data points came from which study. 

Of direct relevance to the pre-modern European wound preparations outlined in the Introduction, we noted a 56 relative dearth of studies looking at lipophilic extracts, and we did not see any studies which paralleled the use of 57 vinegar (aqueous extract but in a low pH solvent). The two studies which did look at lipophilic extracts did report 58 some of the lowest MICs in our dataset. These were seed oils (34) and essential oils (62); note "essential oils" are 59 volatile oils and the study which extracted these from nettle leaves reported a preponderance of fatty acids in the 60 extract. The lack of focus on fats/oils is even more interesting because of reports that lipophilic extracts of nettle on their own may synergise with an antibacterial agent like vinegar to produce a mixture with enhanced activity physical components of a remedy to facilitate the delivery of other ingredients to the treatment site.

Author statements team's project.

antimycobacterial, and antifungal activities of some plants used as folk remedies in Turkey against

Urtica spp.: Ordinary Plants with Extraordinary Properties

Comparing antibacterial potential and phytochemical constituents of 316 two species of genus Urtica International Journal of Life science and Pharma Research

Studies on antioxidant and antimicrobial activities of the aerial parts of Urtica 337 dioica L

Eastern Cape Province traditionally used for skincare and beauty