Developing liquid AP-MALDI MS as a rapid large-scale classification method for determining farm animal health

Lead Research Organisation: University of Reading
Department Name: Chemistry

Abstract

Good farm animal health is paramount for food safety and security. It also reduces the burden on the environment and public health by a reduction in disease treatments using antimicrobials or other drugs. Greater knowledge and earlier detection of agriculturally important diseases will result in better farm disease management and welfare of farm animals as well as improved food produce yield and safety. In addition, recent food adulteration scandals have highlighted the need for faster and more detailed knowledge of our food and its production line to determine and guarantee food authenticity.

Information gathering for all of the above is still limited by the analytical methods available. Some of these are not sufficiently cost-effective and/or too cumbersome and slow for rapid action. Others can be fast but are often very targeted, allowing the detection of only one condition, and limited in their practical use, particularly in combination with other tests.

Fortunately, advances in modern mass spectrometry now allow highly sensitive recordings of the molecular profiles of biological samples such as milk. In combination with advanced bioinformatics these recordings can be used to teach computer algorithms for future classification and thus detection of agriculturally important conditions such as mastitis. As modern mass spectrometry is one of the most sensitive molecular detection and characterisation techniques, allowing the analysis of many molecular biomarkers simultaneously, there is now a great opportunity to make a step change in the early detection of such markers for many diseases in a cost-effective and fast way.
However, cost-effectiveness and speed of analysis can still be limited due to the sample preparation necessary for analysis by mass spectrometry. Our lab has recently demonstrated that biological samples such as milk can be effectively analysed by mass spectrometry with no or minimal sample preparation by using liquid MALDI mass spectrometry.

In collaboration with the University of Reading's Department of Agriculture and Centre of Dairy Research (CEDAR) with its large research herd of around 600 cows, we now aim to explore and fully develop a rapid analysis workflow for biological fluids from farm animals based on modern liquid MALDI mass spectrometry. This workflow should allow for early and specific detection of agriculturally important diseases and conditions, exploiting to the fullest mass spectrometry's great potential in highly sensitive, fast and inexpensive characterisation of the health of our farm animals and their produce.

Technical Summary

Good farm animal health is paramount for food safety and security. It also reduces the burden on the environment and public health by a reduction in disease treatments using antimicrobial and other drugs. Greater knowledge and earlier detection of agriculturally important diseases will result in better farm disease management and welfare of farm animals as well as improved food produce yield and safety.
Information gathering for the above is still limited by the analytical methods available, often due to cost or a lack of robustness, accuracy or speed. Some methods such as ELISAs can be fast but are targeted, allowing the detection of only one condition.

We have recently developed liquid AP-MALDI MS as a new rapid analysis method using the well-established MALDI technique but allowing for the analysis of crude liquids such as milk with no or minimal ('one-pot') sample preparation. We have also used this new method in atmospheric pressure (AP) ionisation sources of three commercial instruments with great success, thus circumventing the requirement for time-consuming sample introduction to a high-vacuum ion source. Both changes have enabled us to substantially reduce the time needed per sample to be analysed, now allowing highly sensitive, accurate and rapid recordings of the molecular profiles of biofluids such as milk. In combination with advanced bioinformatics these recordings can be used to teach computer algorithms for the detection and classification of agriculturally important conditions.

In collaboration with the University of Reading's Department of Agriculture and Centre of Dairy Research (CEDAR) with its large research herd of ~600 cows, we will develop and validate this new methodology. The developed workflow should allow for early and specific detection of agriculturally important diseases and conditions, exploiting to the fullest mass spectrometry's great potential in highly sensitive, fast and inexpensive characterisation of farm animal health.

Planned Impact

Farm animal and plant health is widely discussed in the public and arguably an area where great improvements can still be made. Through accurate (and if possible earlier) diagnosis of diseases on our farms substantial improvements will be achievable in many areas such as the reduction of antimicrobial treatments and pest control agents. Advances in these are a crucial component in combating antimicrobial resistance and environmental pollution. They also provide improved animal health and welfare and an environment that is greener, more robust and diverse. These are desirable outcomes that naturally attract interest and benefit the public. In addition, farming can be more cost-effective if diagnostic testing is faster, cheaper (but more accurate) and multiplexed.

Consequently, the impact of the proposed work of advancing farm diagnostics to the next level of analytical capabilities using novel and more powerful methods can be extremely rewarding.

Scientists and farmers will benefit from the direct and indirect scientific and agricultural impacts of the proposed research, which include:
- More accurate and potentially earlier (pre-symptomatic) analysis of farm animal and crop diseases.
- Less testing (through the potential of multiplexing) but more diagnostic information
- Reduction in the application of disease treatments, including antimicrobial treatments
- Less environmental pollution
- Discovery of new research targets and leads for improved animal and crop health and their identification and characterization
- Further input for the development of a comprehensive model for pathways in agriculturally important diseases

However, the public (and farmers and the food industry) will also benefit through the direct and indirect societal and economic impacts of the proposal, which include:
- Cost-reduction in farming by reducing treatments and pest control and ultimately lower costs of diagnostic testing
- Greener farming with all its benefits to the environment
- Improved consumer confidence and public opinion towards farming
- Population health benefits through a reduction in the risk of antimicrobial resistance using 'precision medicine' for farm animals (lowering the number of unnecessary treatments)
 
Description Highly accurate identification of milk mixture by liquid MALDI MS profiles;
Early data shows great potential for presymptomatic detection of mastitis.
Exploitation Route We have been awarded a contract with DEFRA to apply the technology to bovine TB. Currently, the translation of the developed methods to nasal swabs for the detection of bTB looks very promising. We will further discuss this data with DEFRA and explore how this can be taken forward. We will also explore our contacts in the farming community to see how this new diagnostic tool can be further exploited. In addition, it is timely to explore its clinical application, which is an on-going project funded by the EPSRC.
Sectors Agriculture, Food and Drink,Environment,Healthcare,Pharmaceuticals and Medical Biotechnology,Other

 
Description We are exploring its wider impact in animal health through the above-mentioned DEFRA grant and in clinical diagnostics through the above-mentioned EPSRC grant.
First Year Of Impact 2021
Sector Agriculture, Food and Drink,Healthcare
Impact Types Policy & public services

 
Description A Cost-Effective High-Speed Clinical Diagnostics Instrument for Large Population Screening Based on Novel Liquid AP-MALDI MS Technology
Amount £905,728 (GBP)
Funding ID EP/V047485/1 
Organisation Engineering and Physical Sciences Research Council (EPSRC) 
Sector Public
Country United Kingdom
Start 09/2021 
End 09/2024
 
Description DEFRA tender 26952: Developing liquid AP-MALDI MS as a rapid diagnostics method for bovine tuberculosis
Amount £99,613 (GBP)
Organisation Department For Environment, Food And Rural Affairs (DEFRA) 
Sector Public
Country United Kingdom
Start 02/2021 
End 03/2022
 
Description Equipment loan for HTS by AP-MALDI MS
Amount £20,000 (GBP)
Organisation Waters Corporation 
Sector Private
Country United States
Start 11/2019 
 
Title Analysis of adulteration in milk by liquid AP-MALDI MS profiling 
Description This research tool allows a fast identification of milk from different species and detection of up to 5% of cow milk added to goat. It is based on the application of the newly developed AP-MALDI MS ion source capable to rapidly identify the specific molecular signatures of a given sample. 
Type Of Material Technology assay or reagent 
Year Produced 2019 
Provided To Others? No  
Impact The developed method could be easily applied to the routine analysis for food adulteration and traceability. In the field of animal production it could be applied for the early detection of diseases relevant for animal welfare (bovine mastitis). 
 
Title COW MILK BIOREPOSITORY 
Description Diagnostic and screening tools are becoming more and more important in the field of animal productions. This topic is particularly relevant, not only for business-related issues linked to dairy industry, but with respect to the One Health concept. Being able to rapidly characterize easily bioavailable specimens, such as milk, for the detection of cow-husbandry-related conditions means providing the tools for a prompt and effective intervention. This is particularly true, for example, in the case of mastitis and antibiotic resistance. Consequently, a longitudinal well-characterized sample collection from a controlled research environment provides the basis for analyzing health conditions/pathologies in relation to time. Such biobank is available at the Chemistry Department of the University of Reading. 
Type Of Material Biological samples 
Year Produced 2019 
Provided To Others? Yes  
Impact The purpose of this new biobank tool is to provide the possibility to study the evolution of diseases and health conditions over time in an easily collected and biologically informative specimen, such as milk. The dataset obtained so far confirms that an SCC number higher than 200,000/ml is related to a higher probability of mastitis infection in at least one quarter. However, as shown in the results section, an average number of 150 cows presented with an SCC number higher than 200,000 but only 22 were diagnosed with clinical mastitis. This underlines the necessity to develop more accurate screening methodologies. This biobank offers the possibility to study each condition before clinical events and, in the case of mastitis, provides the basis to study around 100 cases (with adequate controls) during a period of four months in their pre-clinical phase. 
 
Title Detection of clinical and preclinical mastitis 
Description Introduction: Liquid AP-MALDI as a modified ion source were used to analyse homogenous liquid sample droplets on a commercial high-performing Q-TOF mass spectrometer with ion mobility separation. This system has the advantage of being able to produce and analyze MALDI-generated protein ions with high charge states. As the arrangement allows for the detection of high-mass proteoforms in an m/z range below 2000, the full power of modern hybrid mass analyzers can be utililzed for MALDI, in the same way as for ESI-based analyses. Thus, it is possible to detect diagnostically relevant proteins, lipids and metabolites in the same MALDI-MS profile and achieve structural analysis of these by high-quality MS/MS and from the same sample by just performing a simple, one-pot sample preparation. Methods: A milk biobank of 12,000 samples was established by collecting weekly samples from around 500 cows at the CEDAR research farm at the University of Reading, UK. Milk aliquots from this biobank (20µL) and from milk bought at local groceries (50µL) were precipitated with trichloroacetic acid and re-suspended in water/acetonitrile/isopropanol. The analyte extraction solution (0.65uL) was added to a pre-spotted liquid support matrix (LSM) droplet (0.65uL; CHCA 30 mg/mL in acetonitrile:water, 70:30 (v/v), mixed with ethylene glycol in a ratio of 10:7). Automated MS data acquisition was achieved with a Waters Synapt G2-Si and an in-house developed AP-MALDI ion source. Data analysis was undertaken with AMX (Waters) software. Preliminary data: Using milk bought at local groceries, goat milk adulterated with 5% cow milk was detected with 92.45% sensitivity and 94.53% specificity. Goat milk adulterated with 10% cow milk was detected with 99.15% sensitivity and 99.10% specificity. Using aliquots from the milk biobank, clinical bovine mastitis (n=60) was detected with a classification accuray of 98.52% using a large control group (n=327) that included a subgroup with high somatic cell counts (hSCC, n=106). Pre-clinical mastitis (PCM) was detected with 85% correct classification rate for up to four days before the clinical mastitis event and with 95% correct classification rate two days before the clinical mastitis event. These results demonstrate the potential of liquid AP-MALDI MS profiling coupled with multivariate analysis for robust and highly accurate profiling using lipid and protein MS profiles in biological specimens. This was achieved with stable ion yields and low sample consumption. Its simplicity and applicability to high-performing hybrid MS instrumentation manufactured for the use of AP ion sources makes it a MALDI MS method that can uniquely take advantage of the performance of such instrumentation, including speed in changing samples and various analysis modes in a very short time. Examples of MS/MS structural analysis of relevant biomarkers from the same liquid MALDI sample will also be presented. Novel aspect: High-throughput AP-MALDI MS as a novel biotyping approach combining high-sensitivity detection and MS/MS analysis of ESI-like multiply charged analytes. Oral Sessions: Food Safety & Chemistry: Innovations; Food Safety & Chemistry: Foodomics, Allergens, Bacteria, Foods, and Supplements Poster Sessions: Food Safety: General; Ambient Ionization: Applications; MALDI: Applications 
Type Of Material Technology assay or reagent 
Year Produced 2020 
Provided To Others? Yes  
Impact These results demonstrate the potential of liquid AP-MALDI MS profiling coupled with multivariate analysis for robust and highly accurate profiling using lipid and protein MS profiles in biological specimens. This was achieved with stable ion yields and low sample consumption. Its simplicity and applicability to high-performing hybrid MS instrumentation manufactured for the use of AP ion sources makes it a MALDI MS method that can uniquely take advantage of the performance of such instrumentation, including speed in changing samples and various analysis modes in a very short time. Examples of MS/MS structural analysis of relevant biomarkers from the same liquid MALDI sample will also be presented. 
 
Title Liquid AP-MALDI MS spectral fingerprint for detection of food fraud/adulteration 
Description The database is composed by a set of liquid AP-MALDI MS scans (60 seconds) necessary to build the model for recognition and a set of scans necessary to test the model. 
Type Of Material Database/Collection of data 
Year Produced 2019 
Provided To Others? No  
Impact The model will be used for the detection of food fraud in milk. More precisely, it is capable to detect 5% of interspecific milk mixtures. 
 
Title MALDI MS-based analysis of Puccinia komarovii var. glanduliferae-infected Impatiens glandulifera leaves 
Description Mass spectrometric methods such as MALDI MS, including the newly developed liquid AP-MALDI MS, are able to discriminate accurately between P. komarovii var. glanduliferae-infected and uninfected control I. glandulifera leaves. 
Type Of Material Computer model/algorithm 
Year Produced 2020 
Provided To Others? Yes  
Impact MALDI MS biotyping methods have the potential to be applied to rapid and large-scale detection of plant diseases, using simple extraction protocols. However, for liquid AP-MALDI-Q-TOF MS other extraction protocols will need to be developed before protein profiles can be used for biotyping. Once these are developed, there will be great potential for this new MALDI methodology in advancing MALDI MS biotyping due to its superior MS/MS capabilities using high-performing hybrid mass analyzers. Nevertheless, the current protocol provided sufficient amounts of small biomarkers in profiles for a classification accuracy of 100% by liquid AP-MALDI-Q-TOF MS. Small biomarkers are virtually impossible to record on commercial MALDI MS biotyping instruments that typically use linear axial-TOF mass analyzers. 
 
Title Rapid liquid AP-MALDI MS profiling of lipids and proteins from goat and sheep milk for speciation and colostrum analysis 
Description Dataset of liquid AP-MALDI MS profiling of lipids and proteins from goat and sheep milk for speciation and colostrum analysis. The dataset supports the article by Piras et al. 2020, 'Rapid machine learning-liquid AP-MALDI MS profiling of lipids and proteins from goat and sheep milk for speciation and colostrum analysis'. The dataset contains mass spectrometry data, multivariate models and validation reports. 
Type Of Material Database/Collection of data 
Year Produced 2020 
Provided To Others? Yes  
URL https://researchdata.reading.ac.uk/id/eprint/247
 
Description Analysis of goat and sheep milk samples for traceability and welfare assessment 
Organisation Magna Græcia University
Country Italy 
Sector Academic/University 
PI Contribution Analysis of goat and sheep samples through liquid AP-MALDI MS and generation of a database of mass spectrometry profiles.
Collaborator Contribution Samples and metadata collection. Analysis of the same samples through Fourier-transform infrared spectroscopy (FTIR) and of somatic cell count (SCC).
Impact Analysis of correlation indexes between liquid AP-MALDI MS profiling and parameters related to quality/welfare/traceability.
Start Year 2019
 
Description Developing liquid AP-MALDI MS as a rapid large-scale classification method for determining farm animal health (VEERU) 
Organisation University of Reading
Department Veterinary Epidemiology and Economics Research Unit
Country United Kingdom 
Sector Academic/University 
PI Contribution As described in BBSRC proposal
Collaborator Contribution As described in BBSRC proposal
Impact Multidisciplinary involving veterinary epidemiology, agriculture and analytical chemistry
Start Year 2017
 
Description Developing liquid AP-MALDI MS as a rapid large-scale classification method for determining farm animal health (Waters) 
Organisation Waters Corporation
Department Waters Corporation Centre of Mass Spectrometry Excellence
Country United Kingdom 
Sector Private 
PI Contribution As described in BBSRC proposal
Collaborator Contribution As described in BBSRC proposal
Impact None yet.
Start Year 2018
 
Description Liquid AP spectral profiling of milk from cows (single quarters) with different inflammatory/infectious profiles. 
Organisation University of Glasgow
Country United Kingdom 
Sector Academic/University 
PI Contribution Our team will perform the full spectral profiling (metabolites, lipids and proteins) of the milk of single quarters from cows controlled for bacterial growth and haptoglobin levels.
Collaborator Contribution Our partners will provide the milk of the individual quarters of cows collected from 2 to 10 days post calving.
Impact This collaboration will improve the knowledge of the quarter to quarter variability in terms of timing, inflammatory and infectious conditions. The possibility to analyse these samples with our spectral profiling approach will provide an important data-set for the interpretation of the data related to the BBSRC research project.
Start Year 2019
 
Description Vet school talk 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach Local
Primary Audience Professional Practitioners
Results and Impact Presentation of our work to the vet school for discussion. The talk was entitled "Liquid AP-MALDI MS for HTP bioprofiling of dairy milk".
Year(s) Of Engagement Activity 2018