The effectiveness of a nonalcoholic disinfectant containing metal ions, with broad antimicrobial activity

Disinfectants have different efficacies depending on their use and the target microorganism. This study aimed to evaluate the efficacy and safety of our new nonalcoholic disinfectant, which consists mainly of metal ions. According to the 17th revised Japanese Pharmacopoeia and ASTM international E1052 method, the bactericidal and virucidal efficacy of this new disinfectant against 13 microorganisms was evaluated by the in vitro quantitative suspension test. Additionally, the disinfectant cytotoxicity against multiple cell lines was examined. Then, a safety test using a human open patch test was performed with 26 healthy volunteers. This disinfectant showed strong bactericidal and virucidal activities: all microorganisms except enterovirus were inactivated very quickly. The infectivity of 12 microbial strains was eliminated within 5 min of disinfectant exposure. Additionally, this disinfectant showed little acute cytotoxicity in vitro. All volunteers were negative in the human open patch test. Our new disinfectant has a broad spectrum of microbial targets, is safe for human skin, and demonstrates no cytotoxicity. This disinfectant could prevent common microbial infections.


Results
The results of experiment 1 are presented in Table 1. The number of bacteria after 5 and 30 min of disinfectant exposure was less than the lower limit of quantification or < 100 CFU/mL for the 9 bacteria. Additionally, the log10 reduction value (LRV) was 3.3-4.8, indicating bactericidal efficacy, and the disinfectant is fast-acting and persistent and has a broad antibacterial spectrum. Next, the viral infection titers of influenza A virus and human adenovirus after 15 s and 5 min of exposure to the disinfectant were less than the lower limit of quantification (< 1.3 × 10 1 the 50% tissue culture infectious dose (TCID 50 )/mL), and those of feline calicivirus were 1.9 × 10 2 TCID 50 /mL, also less than the lower limit of quantification. Additionally, the LRV was 4.6 to 5.9, indicating virucidal efficacy for the above 3 viruses; however, the viral infection titers of human enterovirus after 15 s and 5 min were 1.5 × 10 4 TCID 50 /mL and 2.7 × 10 3 TCID 50/ mL, respectively, and the LRV was 0-0.8, indicating no virucidal efficacy for human enterovirus. Then, the ratios of the bacterial number after the action of the inactivators to that of the control were 92-127%, which were within the criteria in the JP17 (Supplementary Table S1). As a result, the inactivator/neutralizer was validated as being effective in quenching the activity of the biocide, as bacterial viability remained within one-half to 2 times the control value.
The results of experiment 2 are presented in Table 2. The viable rate of cells cultured using the cytotoxicity confirmation sample was 50% or greater, and the disinfectant had no cytotoxicity against the cells used for the infectivity titer determination. For A549 cells, the stock solution of the cytotoxicity confirmation sample had a viable cell rate of less than 50% by absorbance measurements, but the disinfectant was determined to have no cytotoxicity against these cells because a cytopathic effect (CPE) was observed under a microscope. Additionally, Table 1. Time taken to completely inactivate microorganisms following contact with the disinfectant. CFU/ mL colony-forming units per milliliter, lower limit of quantification 10 CFU/mL (except for Campylobacter, with a lower limit of quantification of 100 CFU/mL), min minutes, LRV log reduction value, TCID 50 /mL 50% tissue culture infectious dose. www.nature.com/scientificreports/ it was determined that no cytotoxicity occurred in the stock solution of the sample used for cytotoxicity confirmation, indicating that the detection limit was 1.3 × 10 1 TCID 50 /mL. Finally, the results of experiment 3 are presented in Table 3. The skin reaction after both 30 min and 24 h in the human open patch test was completely negative for all healthy volunteers, indicating that the disinfectant is safe for human skin.

Discussion
Disinfection is an important method to prevent infection and cuts off infection routes. The selection of disinfectants according to the situation and environment often requires expertise and experience. Rutala and Weber 14 listed 14 properties of an ideal disinfectant and 5 key considerations for selecting the optimal disinfectant. To date, no product that meets all of the properties for healthcare disinfection has been introduced; however, products with these properties and meeting these key considerations as much as possible are recommended as disinfectants. Recently, HAIs as well as viral pandemics, such as influenza virus and coronavirus, have caused social disturbances worldwide [4][5][6][7][8][9] . The EPA also recommends the use of disinfectants that cover the vast majority of HAIs (79.1%) 15 . We examined the bactericidal and virucidal efficacy of a newly developed disinfectant against microorganisms that can cause HAIs and pandemics and confirmed its effectiveness.
Currently, alcohol is most often used as an antiseptic, but it is not EPA-registered and is classified as a low-level disinfectant. Alcohol has a relatively broad spectrum for microorganisms but is slow to act against nonenveloped viruses. Additionally, it is difficult for alcohol to complete the required contact time with microorganisms because of its rapid evaporation, and it is not recommended for use on large surfaces because of its flammability. Rutala and Weber 2 summarized the advantages and disadvantages of other low-level disinfectants. Based on their properties, the disinfectants not only require proper contact time with microorganisms but also must have a short kill time to achieve complete disinfection. Generally, the contact time should be longer than or equal to the kill time. Most alcohol-based solutions dry quickly, while aqueous-based disinfectants such as hypochlorite and phenolics will maintain a wet-contact time of approximately 1-2 min 14 . Our disinfectant is a nonalcoholic product made by mixing three types of metal ions (iron, zinc, and nickel) with an amino acid (L-cysteine), a surfactant (sodium lauryl sulfate), vitamin C (ascorbic acid) and an organic compound (potassium sorbate). This disinfectant rapidly inactivated almost all of the microorganisms tested. Additionally, this is a nonalcoholic liquid and contains a surfactant; hence, this may prolong the contact time with microorganisms. The metal ions are expected to be involved in sterilization. Several studies have reported that iron chloride exhibits antiviral activity in a concentration-dependent manner against herpes simplex virus type 1 (HSV-1) and bovine viral diarrhea virus and decreases the number of RNA or DNA replicates 16 . Zinc was shown to play a complementary role in enhancing bactericidal activity against the common mastitis causative pathogens Streptococcus uberis, S. aureus, and E. coli. Additionally, zinc has been reported to enhance virucidal activity against HSV-1 up to fourfold and mediate antiviral effects through inhibition of viral penetration or egress or progression of the intracellular phase of the viral life-cycle of transmissible gastroenteritis virus. Consequently, zinc has been shown to mediate antibacterial and antiviral effects against certain bacteria and viruses [17][18][19] . Some antimicrobial effects of other additives have also been reported. L-cysteine is an important amino acid and has been reported to inhibit the growth of various strains of E. coli 20 . Additionally, cysteine thiol groups often interact with metal ions and are involved in binding, transport, and storage of these ions in the cell 21 . Sodium dodecyl sulfate (SDS) is widely known as a representative anionic surfactant and is one of the main components of hand soaps. SDS was shown  www.nature.com/scientificreports/ to have anti-influenza virus effects on a human influenza virus strain (H3N2). It is believed that the mechanism of inactivation of the influenza virus is an electrical interaction between the surfactant and HA proteins 22 . A former disinfectant of ours, reported by Takakuwa et al. 13 , includes a small amount of surfactant and has very strong anti-influenza viral activity (including avian, swine, and human). Ascorbic acid, also called vitamin C (VC), has been widely known for its antioxidant properties, immunomodulatory and anti-infectious effects since the 1930s and can scavenge damaging reactive oxygen species. Numerous reports have indicated that VC has antibacterial effects against distinct bacteria in vitro. VC concentrations of 0.31 mg/mL, 0.15 mg/mL, and 0.5 mg/ mL effectively inhibited Pseudomonas aeruginosa (P. aeruginosa), S. aureus/Enterococcus faecalis, and Campylobacter jejuni growth in vitro, respectively. Furthermore, 8-16 μg/mL (i.e., low-level) VC effectively counteracted biofilm formation by MRSA. VC alone could even more effectively inhibit Salmonella growth. Several studies have reported that VC inhibited the replication of HSV-1, poliovirus type 1, and influenza virus type A, and it exhibited low-level fungistatic activity against Candida albicans (C. albicans). Thus, VC possesses antimicrobial activities that reduce the pathogenicity of bacteria, viruses, and fungi 23 . Therefore, we hypothesize that this disinfectant exerts an antimicrobial effect via the coordinated action of these ingredients. However, in this study, our disinfectant had different virucidal efficacies between norovirus (feline calicivirus) and enterovirus, which belong to a family of nonenveloped RNA viruses. It is well known that alcohol has a minimal virucidal effect and that sodium hypochlorite is effective for disinfecting these nonenveloped viruses. Sato et al. reported that low-pH alcohol (acid-alcohol) had virucidal efficacy against human norovirus, indicating enhancement of the virucidal effect of alcohol by acidification 24 . On the other hand, enteroviruses are well recognized to be stable under acidic conditions. The virucidal efficacy against enterovirus 71 has been reported to be higher under alkaline conditions (pH 8.2) 25 . The differences in virucidal efficacy among nonenveloped viruses may be pH dependent.
The proper selection of a disinfectant requires taking into account not only antimicrobial activity and contact time but also a lack of harmful effects on the human body. Safety involves several components, including the toxicity, flammability and compatibility of the substance and the use of personal protective equipment. The disinfectant should be nontoxic and harmless to users. The least toxic product should be selected. It is generally known that nickel, which is contained in our disinfectant, is a common allergen responsible for allergic contact dermatitis 26 . However, the human skin patch test confirmed safety in all subjects. Additionally, this disinfectant showed little acute cytotoxicity in vitro. In the future, we plan to evaluate effects on the mucous membrane and eyes as a further safety assessment.
Our new disinfectant is a next-generation product, and its in vitro antimicrobial effects and safety were confirmed in this study. However, its mechanism of action, antimicrobial effects under other conditions (in the presence of organic substances, etc.), and clinical efficacy remain unclear. Therefore, it is necessary to further study the antimicrobial effects and safety of this disinfectant, but this is a product with many possibilities.

Disinfectant.
To prepare the disinfectant, the bacteriostatic action was verified for each component, and the optimum concentration was confirmed. These components were mixed to adjust the concentration having the strongest bactericidal action. Additionally, it was confirmed that the bactericidal effect was further enhanced by adjusting the pH to approximately 3. Therefore, the new disinfectant (FEION) was prepared as follows. First, solution A was made by dissolving 0.96 g of FeCl 3 6H 2 O, 0.25 g of ZnSO 4 and 0.18 g of NiSO 4 7H 2 O in 200 mL of distilled water (Otsuka Pharmaceutical). Second, solution B was prepared by dissolving 1 g of l-cysteine, 0.1 g of ascorbic acid, 0.05 g of potassium sorbate, and 0.1 g of sodium lauryl sulfate in 800 mL of distilled water. Finally, solutions A and B were mixed, and 3 N HCl was added to this mixture to adjust it to pH 2.3-2.5. This new disinfectant was a colorless and transparent liquid and had less metallic odor than the previous ones. Additionally, a patent application for this disinfectant has been submitted to the Japan Patent Office (patent number: JP 5327218).  Table 1. First, the cryopreserved bacterial strains #1-#9 were inoculated and cultured under the conditions shown in Supplementary Table S2. Then, the grown colonies were scraped, suspended in sterile ion-exchanged water (#1-#6, #9), 3% NaCl (#7) or KH 2 PO 4 (3.4% solution mixed with water at 1:800, Wako) (#8) to prepare a mixture at approximately 10 7 colony-forming units per milliliter (CFU/mL), which was used as the bacterial test solution. The initial inoculum size of E. coli, E. coli O157, P. aeruginosa, S. enterica, S. aureus, MRSA, V. parahaemolyticus, C. jejuni, and C. albicans was 5.2 × 10 7 CFU/mL, 4.6 × 10 7 CFU/mL, 6.5 × 10 7 CFU/mL, 7.0 × 10 7 CFU/mL, 5.7 × 10 7 CFU/mL, 4.8 × 10 7 CFU/mL, 5.0 × 10 7 CFU/ mL, 2.5 × 10 7 CFU/mL, and 7.0 × 10 7 CFU/mL, respectively.

Experiment 1. Preparation of bacterial test solutions
Preparation of viral test solutions. The viruses tested are presented in Table 1. First, #10 was inoculated into embryonated specific-pathogen-free hen eggs and cultured at 35.5 °C for 2 days. Then, the allantoic fluid was collected and concentrated with an ultrafiltration membrane, which was followed by sucrose density gradient centrifugation (centrifugation condition: 108,000×g, 4 °C, for 3 h) to obtain a viral solution. Viruses #11-#13 were infected into Crandell-Rees feline kidney (CRFK), A549 and Vero cells, respectively, and when approximately 90% or more of the cell-cultured area showed a CPE, the cells were cryopreserved in a freezer at − 30 °C. Thereafter, a freeze-thaw operation was performed, and the supernatant obtained by centrifugation at 2380×g www.nature.com/scientificreports/ for 10 min (min) was collected and concentrated by the ultrafiltration membrane. Viruses #12 and #13 were preserved as virus solutions at this stage. The collected solution (#11) was further concentrated by a sucrose cushion method (centrifugation conditions: 108,000×g, 4 °C, for 3 h) and finally preserved as a virus solution. For the test, all of the virus solutions (#10-13) were diluted tenfold with Dulbecco's phosphate-buffered saline (-) (PBS, Nissui Pharmaceutical) and used.  27 . A test bacterial solution (0.1 mL) was added to disinfectant (10 mL), and mixed with a tube mixer, and incubated at 25 ± 2 °C for 0 (initial), 5 min, and 30 min. After the action for a predetermined time, the mixture (1 mL) was added to inactivator (9 mL) (SCDLP bouillon medium (SCDLP), Eiken Chemical Co., Ltd., or SCDLP containing 0.25% NaCl, data in the Supplementary Table S1) to stop the bactericidal action, and this solution was prepared as a sample solution for measuring the number of bacteria. Distilled 0.85% NaCl (Wako) was used in place of the disinfectant for the initial time and the control, and 3% NaCl was used for Vibrio parahaemolyticus.

Evaluation of virucidal efficacy (virus inactivation test).
The virus inactivation test was performed according to ASTM international E1052 method (Standard Test Method to Assess the Activity of Microbicides against Viruses in Suspension). After dispensing 0.9 mL of the disinfectant into a test tube, 0.1 mL of a virus solution was added and mixed. Each mixture was allowed to act at room temperature for a predetermined time. To stop the action of the disinfectant, we added 0.1 mL of the mixture to 9.9 mL of the inactivator (SCDLP) to dilute the sample. This solution was used as a stock solution for measuring viral infection titers. In addition, PBS was used in place of the disinfectant for the action at time 0 (initial) and in the control.
Inactivator validation. Since our disinfectant is at a low pH, it was judged that the action cannot be stopped in MEM medium, which has almost no pH-buffering capacity. Therefore, the effectiveness of the inactivator SCDLP used to stop the bactericidal and virucidal activities of the disinfectant was verified. The methods and results are presented in Supplementary Table S1.

Measurements of bacterial counts and viral infection titers.
The medium and reagents used are shown in Supplementary Table S3. A set of tenfold serial dilutions was prepared for the sample solution with each medium and reagent. Then, each of the sample stock solutions and the diluent (1 mL) was transferred to a Petri dish; after mixing with each medium, these were solidified and cultured under the conditions shown in Supplementary  Table S3. The sample stock solution of Campylobacter (#8) was also diluted, smeared and cultured as shown in Supplementary Table S3. Finally, the number of colonies that grew was counted, and the number of test bacteria per 1 mL of the test product was determined. Bacterial count measurements were performed using the platecount method. The choice of method was determined based on the bacterial species. Bacteria #1-#7/#9 and #8 were counted using the pour-plate method and surface-spread method, respectively. The lower limits of quantification were 10 CFU/mL and 100 CFU/mL, respectively. The viral infection titers were measured using the 50% tissue culture infectious dose (TCID 50 ) method. Cells for measuring the viral infection titers were seeded in a 96-well plate and cultured in a CO 2 incubator (MCO-20AIC, SANYO) for 4 days. Next, a stock solution of the viral infection titer sample was serially diluted tenfold with PBS. Each well without the cultured solution was inoculated with 25 μL of a stock solution for measuring infectious titers or a sample diluted tenfold in PBS, and the cells were infected with the virus at 37 °C for an hour. Thereafter, the inoculated virus solution was removed, 0.1 mL of medium for virus culture (see Supplementary  Table S2 for details) were added per well, and the cells were cultured in a CO 2 incubator at 37 °C (culture periods, see Supplementary Table S3 for details). After culturing, the CPE generated by virus multiplication was observed under a microscope, and the viral infection titers were determined using the Reed-Muench method and expressed as TCID 50 /mL.

Calculation of microbial log reduction values.
A log reduction value (LRV) was used as an evaluation parameter for judging the bactericidal and virucidal activities according to the JP17 and European Norm (EN) 14,476: 2013 + A1: 2015 28 . The formula for calculating LRVs is log 10 (initial bacterial count of a control divided by the bacterial count after the action of the disinfectant or initial viral infection titer of a control divided by the viral infection titer after the action of the disinfectant). Note that LRVs are represented to one decimal place (rounded down). Here, an LRV of 3 or greater calculated using the numbers of bacteria before and after the action of the disinfectant (usually for 5-15 min) was used to define the disinfectant as "effective" against bacteria. Then, LRVs of 4 or greater calculated using the viral titers before and after the action of the disinfectant (usually for 15 s to 5 min) indicated that the disinfectant was "effective" for viruses. Experiment 2. The cytotoxicity of this disinfectant for each cultured cell line (for the cell lines used for determining infectious titer, see Table 1 for details) was investigated. The cytotoxicity test was generally conducted according to ASTM E1052. After adding PBS (0.1 mL) to the disinfectant (0.9 mL), a solution diluted 100-fold with SCDLP was used for measurements of cytotoxicity. This stock solution and a solution diluted tenfold with PBS were inoculated with 25 μL per well of cell lines previously cultured in a monolayer in a 96-well plate and then incubated at 37 °C for an hour in a CO 2 incubator. Then, the inoculum was removed, and an infectious titer measurement medium (0.1 mL per well) was added, followed by culturing in a CO 2 incubator.