Silica Particles with Encapsulated DNA
The synthesis and characterization of SPED was conducted according to Paunescu et al.(23) (Figure 1) First, per batch, 4x4ml of silica nanoparticles (50mg/ml in isopropanol) were surface functionalised in 4 separate falcon tubes by adding 40µg of N-trimethoxysilylpropyl-N,N,N- trimethylammonium chloride (TMAPS; 50% wt in methanol) and stirred at 900 rounds per minute (rpm). for 12 hours at room temperature. The trimethylammonium on the surface gave the silica particles a positive surface charge. For DNA adsorption on the surface a 2 ml batch of corresponding annealed DNA molecules was added to 200ml ultrapure MilliQ (mQ) water (150ng dsDNA/µl, Nanodrop). Of the previously TMAPS functionalised particles, 0.4g were added to the solution and the batch was shaken for 10 seconds. Subsequently, 4µl TMAPS were added, the batch was shaken and then sonicated for 20 seconds. Next, 62.5µl of tetraethyl orthosilicate (TEOS) (≥99.0%, Aldrich) were added and the batch was shaken for 5 hours. In a next step, 10ml isopropanol and 5.9ml TEOS were mixed with 484.1ml mQ water before adding it to the previous mixture. The batch was then stirred at 600rpm for 4 days. To demonstrate the potential of multi-tracing, three different batches were produced, each containing a unique DNA sequence. The three particle batches named SPED1, SPED2 and SPED3 had a hydrodynamic size of 218±80nm, 144.6±46nm and 173.4±82nm and a DNA loading of 21µg, 23µg, and 26µg dsDNA/mg of particles, respectively. The suspension used for our experiments was prepared by diluting the particles to 0.1mg/ml or 1mg/ml in mQ water + 10% glycerine. We used glycerine to increase the probability of transfers in our experiments according to preliminary data showing higher transfer rates in the presence of glycerine (data not shown). Glycerine is viscous and therefore increases the stickiness of the test suspension, simulating the adhesive function of bacterial appendages called pili.
Escherichia coli bacterial culture
Standard procedures were used to cultivate bacterial cultures. Escherichia coli was cultured overnight in tryptic soy broth (TSB; Sigma-Aldrich, St. Louis, Missouri, USA). An aliquot of 2ml of the bacterial cell culture was triple washed by spinning down the E. coli culture in a 2ml microcentrifugation tube (Eppendorf AG, Hamburg, Germany) for 5 minutes at 3500rpm. After each cycle, the supernatant TSB was removed, and the bacterial culture resuspended in 1ml phosphate-buffered saline (PBS). The obtained bacterial concentration was 108 CFU/ml. To prevent further bacterial growth during transfer experiments, microcentrifugation tubes were stored temporarily on ice.
Study procedures
We conducted three experiments to investigate the behaviour of SPED in a simulated care scenario (Experiment 1), while performing hand hygiene (Experiment 2), and comparing SPED transfer dynamics with those of E. coli (Experiment 3).
Experiment 1 - SPED transmission characteristics in a simulated care scenario
We defined a simulated patient care scenario featuring an examination stretcher with bed rails, a privacy screen, and an infusion stand with two infusion bags. A patient actor (henceforth patient) and a HCW trained as a physician (MS; henceforth HCW) conducted the following actions in chronological order: closing a door using the door handle (stainless steel), moving the privacy screen (polyvinylchloride), elevating the bed rail (polypropylene), auscultating the patient’s chest with a stethoscope (polyvinylchloride, PVC), taking the patient’s radial pulse (bare skin), checking pupil reactivity while pulling up the patient’s upper eyelid (bare skin), and changing the infusion bag (polyolefin) (Figure 2). Two video cameras positioned in a 90° angle on adjacent walls (GoPro® Hero 4 Black edition, GoPro Inc., San Mateo, CA) recorded the scene simultaneously to monitor the sequence.
The scenario featured 13 experimentally pre-defined hand-to-surface exposures, of which 10 were fomites and 3 were patient skin. The sites were chosen according to our observational experience in healthcare settings as being habitual.(8) We marked swabbing areas of 1x3cm on each of the 13 sites. The sites and both hands and fingers of the HCW were swabbed prior to every simulation to demonstrate they were free of SPED. The swabbing technique included with 20% glycerine solution premoistened sterile cotton swabs (Naturaline Wattestäbchen, Steinfels Swiss, Winterthur, Switzerland) that were rolled three times over the area and then stored in a 2ml microcentrifugation tube at room temperature. This swabbing procedure applied to all following swabs in our experiments.
We inoculated three of the 13 sites with 0.2ml of the SPED test suspensions (1mg/ml) and left to air dry for 5 minutes, namely the door handle with SPED1, the bed rail with SPED2, and the left forearm of the patient with SPED3 (Figure 2). We decided to use three different initial deposition sites with three SPED with distinct DNA codes to test the parallel application and recovery of distinguishable SPED batches in the same healthcare scenario. Throughout the entire simulated care scenario, the patient was laying on his back on the examination stretcher with his bare arms placed beside him without contact to his body or the examination stretcher. Subsequently, the HCW actor performed the care scenario while exclusively touching the 13 marked sites in a natural way. Once the simulation scenario was concluded, we swabbed the 13 marked test areas as well as both of the HCW’s palms and the fingers of her dominant right hand, resulting in 48 specimens overall. After the experiment, the microcentrifugation tubes were immediately transported to the laboratory for processing.
The experiment was repeated three times on different days and different simulation rooms, each, referred to as ‘runs’. The repetitions were to add validity to the findings.
Experiment 2 - Effect of hand hygiene action on SPED concentration
We tested the impact of hand disinfection with alcohol-based handrub and hand washing using soap and water on SPED skin concentration. Three groups of five participants, each, were formed: a control group (CTRL), a hand disinfection group (DISINF), and a hand washing group (WASH). First, swabs were taken from the right thenar of each participant to demonstrate no SPED were present. Subsequently, a previously marked area of 2x2cm on the palmar side of the right thenar of each participant was inoculated with 0.2ml of the SPED1 test solution (0.1mg/ml). The fluid was dispensed over the palms of both hands by rubbing both hands against each other for 10 seconds. Once dry, swabs were taken from the previously marked area of the right thenar of each person, and additionally from the dorsal hypothenar of DISINF participants to test for dispersal by rubbing hands with alcohol-based handrub.
CTRL participants held both their hands in the air without any contact for 1 minute. DISINF participants performed hand hygiene according to the norm EN1500 using 3ml of a commercially available alcohol-based handrub (80% Ethanol with 1% Glycerine, B. Braun Medical AG, Sempach, Luzern, Switzerland). WASH participants washed their hands for 30 seconds with common soap (Liquid Soap®, Sterisol AB, Vadstena, Sweden) and water using the norm EN1499 procedure. The right thenar of all participants – and additionally the dorsal hypothenar in DISINF participants – were swabbed again after the non-contact waiting time in the CTRL group and hand hygiene actions in the DISINF and WASH group. Thereafter the test tubes were immediately transported to the laboratory for analysis.
Experiment 3 – Transmission dynamics of SPED compared with Escherichia coli
We analysed transmission tendencies of E. coli and SPED. Prior to the experiment, a piece of laminate (melamine resin) was rinsed with 70% ethanol, wiped with a sterile cloth and allowed to air dry. Forty series of four squares of 2x2cm were defined. The four squares were marked as origin surface (ORIGS), 1st transfer surface (1TS), 2nd transfer surface (2TS) and control surface (CTRLS), respectively. Next, samples of the surface and gloves (GLOVES) were taken by either rinsing the surface with 100µl of PBS or by cutting off the glove’s index fingertip and adding it to a 1.5ml micro-centrifugation tube filled with 100µl of PBS to prove sterility. Then, 20µl of the E. coli solution was spread onto ORIGS, and with intervals of 2 minutes, onto CTRLS. The 2-minute interval allowed for continuous sampling of surfaces with identical times between surface inoculation and subsequent sampling. A two-step transmission was performed from ORIGS à 1TS à 2TS by pressing the gloved index finger onto the corresponding areas in that order. Samples were taken as described above. The CTRLS served as a reference for the number of E. coli or nanoparticles recovered from an inoculated surface in the absence of a transfer event, and so not part of the transfer. This procedure was repeated until there were 20 replicates. The same procedure was reproduced for 20 replicates with 20µl of the SPED solution (0.1mg/ml).
Quantification of SPED
To quantify the amount of DNA particles from each swab after having performed the experiments, the cotton swab was placed in a 2ml microcentrifugation tube (Eppendorf AG, Hamburg, Germany) with 200µl of mQ water (Figure 1). Each sample was ultrasonicated for 1 minute and vortexed for 10 seconds. To release the DNA from the silica coating, 1%vol of a highly diluted solution of buffered oxide etch (BOE) with 0.03wt% ammonium hydrogen difluorid (NH4FHF, pure, Merck) and 0.02wt% ammonium fluoride (NH4F, puriss., Sigma-Aldrich, St. Louis, Missouri, USA) were added to the sample. The resulting suspension was then analysed by quantitative PCR (LightCycler® 96) in a multiplex setup with three sequence-specific fluorescent probes (SPED1: Hex, SPED2: Texas Red, SPED3: FAM). All three fluorescent probes could be detected simultaneously and allowed to quantify each particle separately. The qPCR total reaction volume was 12.5µl consisting of 2.5µl sample solution, 0.2µl of each primer (3x forward and reverse),0.125µl of each probe (Microsynth AG) and Mastermix 2x (GoTaq® Probe, Promega). The qPCR program consisted of a preincubation step for 600 seconds at 95°C, followed by a 2-step cycling for 15 seconds at 95°C and for 60 seconds at 56°C. For quantification, a dilution series with known concentration of particles was performed.
Sampling of Escherichia coli
Once the transfer procedures were completed, the samples were processed within 2 hours. For sampling, the areas contaminated with E. coli were rinsed with 100µl of PBS, which was then added to a further quantity of 100µl of PBS in a 1.5ml micro-centrifugation tube, and vortexed. As for gloves (Nitrile latex free, LLG Labware, Meckenheim, Germany), the index fingertip was cut off after performing the transmissions, placed in a 1.5ml micro-centrifugation tube containing 200µl of PBS and subsequently vortexed. Aliquots of 100µl of undiluted specimen solution and 100µl from 1:10–1:100’000 dilutions were plated onto TBX Agar (Sigma-Aldrich, St. Louis, Missouri, USA) and incubated at 36±1°C for 24 hours.
Statistical analysis
Descriptive analysis of our data in Experiment 1 and 2 was performed with Microsoft® Excel® 2016. To estimate the detection limit of the nanoparticles in our experiments, we used the Method Detection Limit (MDL). The MDL is defined as “the minimum measured concentration of a substance that can be reported with 99% confidence that the measured concentration is distinguishable from method blank results.” (28) Values that are below MDL are considered to be negative, values higher than the limit as positive. The MDL was calculated as the mean concentration of the blank samples plus three times the standard deviation.
For Experiment 3, test statistics including t-test were conducted with R Statistical Software (version 3.6.0, R foundation, Vienna, Austria). We defined transfer percent as the fraction of E. coli or SPED recovered from 1TS, 2TS, and glove as a percentage of the seeded inoculum recovered from the surface by swabbing, defined by CTRLS.