The study had three phases: (1) laboratory experiments to identify the optimal LNP composition (types of LNPs) and formulation (LNP-matrix combination), (2) testing the selected formulation in a laboratory-based patient care experiment with touching inanimate surfaces with bare hands, and (3) a clinical patient care experiment including touching skin of participants representing patients³⁵.

Synthesis of LNP compositions

Based on dedicated literature^15,16 and our experience, we selected two LNP compositions for the laboratory experiments, one composition containing the lipidoid 306O_i10, which has been previously investigated^16,36, and another composition containing SM-102 as an ionisable lipid, which is known for its increased stability and is close to the formulation used for the Moderna COVID-19 vaccine Spikevax¹⁵.

First, an organic phase containing the lipids and an aqueous phase containing the DNA were prepared. This procedure was the same for both selected LNP compositions. For particles containing the lipid 306O_i10¹⁶, DOPE (1,2-dioleoyl-snglycero-3-phosphoethanolamine, Avanti), cholesterol (Sigma-Aldrich), C14-PEG1000 (1,2-dimyristoyl-snglycero-3-phosphoethanolamine-N-[methoxy(polyethylene glycol)-1000] (ammonium salt), Avanti), and 306O_i10 (tetrakis(8-methylnonyl) 3,3′,3″,3‴-(((methylazanediyl) bis(propane-3,1-diyl))bis(azanetriyl))tetrapropionate) were dissolved in 90% ethanol and 10% 10 nM citrate buffer at molar ratios of 16:46.5:2.5:35. For LNPs with the lipid SM-102, DSPC (1,2-Distearoyl-sn-glycero-3-phosphocholine, Adipogen), cholesterol (Sigma-Aldrich), C14-PEG1000 (1,2-dimyristoyl-snglycero-3-phosphoethanolamine-N-[methoxy(polyethylene glycol)-1000] (ammonium salt), Avanti) and SM-102 (9-Heptadecanyl (8-(2-hydroxyethyl)[6-oxo-6-(undecyloxy)hexyl]amino) octanoate, Cayman) were dissolved in ethanol at molar ratios 10:38.5:1.5:50.

The aqueous phase was prepared to comprise of an annealed DNA amplicon (Microsynth AG, for sequences see Supplementary Table S1), 10 mM Tris base buffer (pH = 7), and 10 nM citrate buffer, with a final DNA concentration of 1 g/l. Both phases were preheated to 37 °C, and the ethanol phase was added dropwise to the aqueous phase in a 3:1 volumetric ratio and mixed by rapid pipette mixing. After mixing, twice the volume of 300 Mm NaCl buffer (pH 4) was added and the LNPs were incubated at 37 °C for 1 h, shaking at 450 rpm. The particles (2 g/l) were stored at 4 °C.

Sample analysis

Samples were collected from surfaces by rolling a pre-moistened (milliQ water) swab (PCR-Nasal swab, MFS-96000BQ, Meidke Gene) over an area of approximately 3 cm². The swab was placed in a clean Eppendorf tube, cut to fit the cap, and 200 µl milliQ water was added. Analysis was conducted as described elsewhere¹⁶.

Two parameters were determined from each sample: (1) total DNA and (2) LNP integrity (Fig. 5). Thereby, LNP integrity represents viability qPCR in viruses^37,38.

Schematic representation of different measurable parameters using LNPs, namely LNP transfer efficiency and LNP transfer integrity. LNP transfer efficiency quantifies the proportion of DNA successfully moved from the donor to the receiver surface by the transfer process and LNP transfer integrity is a measure of degradation of LNPs when transferred from the donor to the receiver surface with or without intermittent hand hygiene. LNP transfer integrity is expected to decrease with intermittent hand hygiene, because LNPs are known to disintegrate when in contact with ABHR.

LNP transfer efficiency quantifies the proportion of DNA successfully moved from the donor to the receiver surface by the transfer process. It is calculated from the difference in cycle threshold between the total DNA measured on the receiver surface and the total DNA remaining on the donor surface measured after transfer, expressed as:

Here, the DNA quantity measured on the receiver surface is \(total \;DNA_{receiver}\) and \(total \;DNA_{donor}\) is the DNA quantity measured on the donor surface. The values are given in cycle thresholds \(Ct\), representing the number of PCR amplification cycles necessary to obtain a signal.

To assess the LNP integrity on a surface, we measured the proportion of DNA encapsulated within intact LNPs. The sample was divided into two portions: one was treated with Benzonase, an enzyme that degrades unencapsulated DNA while leaving DNA within intact LNPs unaffected, and the other portion remained untreated, containing both free and encapsulated DNA. LNP integrity was then calculated from the difference in cycle thresholds between encapsulated DNA (from the treated sample) and the total DNA (from the untreated sample), expressed as:

The resulting value represents the fraction of DNA that remained protected within LNPs expressed as percentages. To evaluate \(LNP\; transfer\; integrity\), i.e., a measure of degradation of LNPs when transferred from the donor to the receiver surce with or without intermittent hand hygiene, we determined the change in \(LNP\; integrity\; \) between the donor and receiver surfaces, indicating how well the LNPs preserved their structural integrity during the transfer. \(LNP\; transfer \;integrity\) is expected to decrease with intermittent hand hygiene, because LNPs are known to disintegrate when in contact with alcohol-based hand rub (ABHR)¹⁶.

In all transfer experiments, the measurements on every surface, including the donor surfaces, were taken after transfer, as swabbing the donor surface before contact would have removed a significant amount of LNPs and, consequently, compromised the results.

Identification of a suitable LNP formulation

To ensure enduring resistance to desiccation on dry surfaces, LNP integrity was assessed. Each LNP composition (1 µl, 2 g/l) was applied to a surface and dried either alone or within synthetic matrices: glycerine (10 µl, 100%), sucrose solution (20 µl, 50 wt% in water), or a sucrose-glycerine combination (20 µl + 10 µl), at 1000 times the LNP mass, yielding eight formulations. External DNA (10 µl, 1 g/l) with a distinct sequence (see Supplementary Table S1)was added to control for LNP reformation. Additional components (chitosan, dextran, agarose, methyl cellulose) had no further effect.

To serve as surrogates for microbial pathogens, LNPs must transfer efficiently between surfaces while remaining intact. LNP transfer efficiency and integrity were assessed by applying LNPs in synthetic matrices (glycerine, sucrose, or both) to an initial coverslip (borosilicate glass, Roth), then transferring them via a gloved fingertip (nitrile, semper care) to a second clean coverslip. High values for both parameters indicate effective transfer without disintegration.

The susceptibility to disinfection is defined as LNP disintegration upon exposure to ABHR. To assess this, gloved hands were disinfected with Sterillium med (85 g ethanol/100 g, Hartmann Group, Heidenheim) using a standardised three-step method²⁹ between touching the first and second coverslip.

Laboratory-based patient care experiment

To assess LNPs as pathogen surrogates, a typical patient examination was modelled using two distinct LNP barcodes. An experimenter (LP) acted as the HCP, while clean coverslips (borosilicate glass, Roth) represented patient body sites. The door handle was contaminated with LNP-1, and the coverslip representing the patient’s left wrist with LNP-2.

The experiment began with the HCP opening the door, contaminating the bare right hand with LNP-1., then touching the coverslip representing the patient’s right hand with right thenar, followed by touching the coverslip representing the left wrist (LNP-2) with right index and middle fingers. Next, the coverslip representing the patient’s submandibular lymph nodes with both index and middle fingers was touched before interacting with a computer mouse to simulate data entry.

This experiment was repeated four times, with and without hand hygiene performed in distinct instances following the WHO My Five Moments for Hand Hygiene (Fig. 4) resulting in four different scenarios³⁹. Hand hygiene was executed using a palmful of Sterillium med (85 g ethanol/100 g, Hartmann Group, Heidenheim) for 15 s according to a validated three-step method⁴⁰. After each scenario, samples were collected from the door handle, ‘patient hand’, ‘wrist’ and ‘lymph node’ coverslips, and computer mouse before cleaning.

Clinical patient care experiment

To assess the impact of human skin, the transmission experiment was repeated using actual patient body sites instead of coverslips but following the same procedure. Again, door handle and patient pulse were contaminated with LNP-1 and LNP-2, respectively. An experienced medical doctor (HS) acted as the HCP, while lab members served as patients. The experiment was conducted four times over two days, with different patient actors, with and without hand hygiene, resulting in four different scenarios. After each scenario, samples were collected from the door handle, patient’s right hand, left wrist, lymph nodes, and computer mouse before cleaning.

Effect of hand washing

After each scenario of the clinical patient care experiment, the HCP washed hands following the WHO ‘How to handwash’ poster²⁷ using liquid soap (sodium laureth sulfate, fresh soap, MBudget). Both hands were then sampled for LNP contamination.