MSACL 2023 Abstract

INTRODUCTION

A sub-optimal vaginal microbiome is characterised by depletion of commensal Lactobacillus species and increased bacterial diversity. This is associated with common infections, like bacterial vaginosis (BV) and increased risk of developing sexually transmitted infections (e.g. HIV), gynaecological cancers, and in pregnancy, suffering miscarriage and preterm birth. However, a sub-optimal vaginal microbiome is often observed in asymptomatic women. Currently, vaginal infection is diagnosed by sending a vaginal swab to a laboratory for culture or microscopy testing. These tests are slow, inaccurate, and require highly trained experts to perform them. We have addressed these issues by developing Direct Swab Analysis by Desorption Electrospray Ionisation Mass Spectrometry (DESI-MS)1. This method works by measuring specific chemicals directly from the surface of routinely collected clinical swabs which can then be used to predict the microbiome composition and immune status of the sample within 2 minutes. The method was originally developed on a large-footprint instrument using an in-house, custom-built swab holder that limited the rotation of the swab to a fixed position. To facilitate the use of the method as a point-of-care device, we aim to transition the technique towards a smaller format, portable instrument and test the stability, reproducibility, and robustness needed to meet regulatory requirements.

OBJECTIVES

To transition our existing DESI-MS prototype from an LTQ Orbitrap mass analyser to a small footprint and robust time of flight (e.g. ACQUITY RDa Detector) instrument and perform analytical validation and assessment of technical variation introduced by sample collection and storage conditions, laboratory procedures or instrument performance.

METHODS

Vaginal swab samples were collected as part of a prospective observational cohort study from the early pregnancy clinic at Hammersmith Hospital, Imperial NHS Trust, London, UK. Swabs for DESI-MS, metataxonomic profiling of the vaginal microbiota, and immune profiling were immediately stored at −80 °C upon collection. An additional swab was collected for Gram staining and microscopic assessment. DESI-MS data were acquired using the RDa/BioAccord mass spectrometry platform (Waters Cor., UK). For swab analysis and as part of the DESI stage, a roboticized sampling device was used. Data were acquired in both positive and negative ion modes in the m/z range of 50-2000. For data acquisition parameters the capillary voltage was set to 0.8-1.2 kV, gas pressure was set at 4 bar, source temperature at 120°C, and gas desolvation at 450°C. Swabs were placed into a rotating holder positioned orthogonally in front of the MS inlet capillary with a swab–capillary distance of approximately 2 mm. The DESI sprayer tip was pointed to the swab centre with a tip-sample distance of 1.5–2 mm and a distance between the tip and the inlet capillary of 2 mm. The entire surface of the medical swabs was analysed by DESI-MS through clockwise rotation of the swab toward the MS capillary. A mixture of methanol/water (95:5, v/v) was used to produce the charged droplets and the desorption of analytes. Statistical and multivariate analysis was done in R and Python.

RESULTS

A swab holder capable of executing screw motion and simultaneous linear plane movement was designed and manufactured. The holder could be fitted onto the existing ionisation source (DESI, Waters RDa system), and seamlessly integrated with the current workflow. The device includes a magnetic, adjustable swab grip that facilitates the analysis of different swab types with differing diameters and lengths. Optimisation was achieved for several DESI geometric parameters and ergonomics, including distance of the swab tip from the swab-inlet capillary-solvent sprayer, facility of cleaning minimising contamination, rotation and linear speed and pause functionalities.

A quality control (QC) standard that provides a sample matrix mimicking the physical, biochemical, and microbiological properties of mucosal smears was developed and used to assess the instrumental stability and correction for drifts in signal intensity commonly observed during the analysis of samples. Twenty endogenous metabolites across the whole m/z range were identified and used to assess instrument performance.

Analysis of vaginal swab samples (n=20) by DESI-MS performed on an Orbitrap mass analyser or the ACQUITY RDa Detector showed comparable data including various small metabolites, polyunsaturated fatty acids, cholesterol sulphate and glycerophosphoserines. Differences in DESI-MS profiles acquired using the ACQUITY RDa Detector associated with vaginal microbial composition could be readily detected including differences in the relative abundances of oxypurinol and various lysophospholipids (p<0.05).

CONCLUSIONS

We have developed a robust, easy-to-use, high-throughput, automated swab-holding device that allows rapid metabolic profiling of vaginal swabs in less than two minutes. Incorporation of QC’s within the analytical workflow allows accurate and reproducible monitoring of instrument stability and performance. This integrated “all in one” benchtop vaginal swab analysis platform is currently being used in an ongoing large-scale analysis of vaginal swabs collected from 1000 women attending early pregnancy clinics. Future work includes the installation of an instrument in the clinical environment for data acquisition and the development of prediction models for vaginal microbiome status.

1. Pruski, P., Correia, G.D.S., Lewis, H.V. et al. Direct on-swab metabolic profiling of vaginal microbiome host interactions during pregnancy and preterm birth. Nat Commun 12, 5967 (2021).