Use of faecal volatile organic compound analysis for ante-mortem discrimination between CWD-positive, -negative exposed, and -known negative white-tailed deer (Odocoileus virginianus)

Figures & data

Table 1. Linear Discriminant Analysis Classification Models of CWD positive and negative deer.

Number of principal components	4	5	6	7
Total misclassification rate (all data combined) (%)	16	12	12	8
Total correct classification rate (all data combined) (%)	84	88	88	92
Misclassification rate (%) by cohort
●PP3 misclassified as PP4	0	0	0	0
●PP3 misclassified as NE3	10	0	10	10
●PP3 misclassified as NE4	0	0	0	0
●PP3 misclassified as NN1	0	0	0	0
●PP3 misclassified as NN2	0	0	0	0
●PP4 misclassified as PP3	0	0	0	0
●PP4 misclassified as NE3	0	0	0	0
●PP4 misclassified as NE4	25	25	0	0
●PP4 misclassified as NN1	0	0	0	0
●PP4 misclassified as NN2	0	0	0	0
●NE3 misclassified as PP3	10	10	0	0
●NE3 misclassified as PP4	0	0	0	0
●NE3 misclassified as NE4	0	0	0	0
●NE3 misclassified as NN1	20	10	10	10
●NE3 misclassified as NN2	0	0	0	0
●NE4 misclassified as PP3	0	0	0	0
●NE4 misclassified as PP4	14	14	29	20
●NE4 misclassified as NE3	0	0	0	0
●NE4 misclassified as NN1	0	0	0	0
●NE4 misclassified as NN2	0	0	0	0
●NN1 misclassified as PP3	0	0	0	0
●NN1 misclassified as PP4	0	0	0	0
●NN1 misclassified as NE3	0	0	0	0
●NN1 misclassified as NE4	0	0	0	0
●NN1 misclassified as NN2	0	0	0	0
●NN2 misclassified as PP3	0	0	0	0
●NN2 misclassified as PP4	20	20	20	0
●NN2 misclassified as NE3	0	0	0	0
●NN2 misclassified as NE4	0	0	0	0
●NN2 misclassified as NN2	0	0	0	0
Misclassification by group (%)
●(PP3 + PP4) misclassified as (NE3 + NE4)	14	7	7	7
●(PP3 + PP4) misclassified as (NN1 + NN2)	0	0	0	0
●(NE3 + NE4) misclassified as (PP3 + PP4)	12	12	12	12
●(NE3 + NE4) misclassified as (NN1 + NN2)	12	6	6	6
●(NN1 + NN2) misclassified as (PP3 + PP4)	10	10	10	0
●(NN1 + NN2) misclassified as (NE3 + NE4)	0	0	0	0
Correct classification by group (%)
●Positive (PP3 + PP4)	86	93	93	93
●Negative exposed (NE3 + NE4)	76	82	82	82
●Negative (NN1 + NN2)	90	90	90	100
Sensitivity when NE3 + NE4 are included with NN1 + NN2	86	93	93	93
Specificity when NE3 + NE4 are included with NN1 + NN2	89	89	89	95
Sensitivity when NE3 + NE4 are included with PP3 + PP4	93	97	97	97
Specificity when NE3 + NE4 are included with PP3 + PP4	90	90	90	100

Eight per cent of all animals in the study were misclassified in the LDA classification model constructed using seven principal component (PC) scores. By cohort, only CWD-positive deer from Herd 3 (PP3), and negative exposed deer from Herds 3 and 4 (NE3, NE4) were misclassified (10%, 10%, 33%, respectively). By group, positive (PP3, PP4) animals were misclassified as NE (7%), and NE individuals were misclassified as PP or known negative (NN) (12% and 6%, respectively). No NN animals were misclassified in the optimal model. Calculated SN:SP when NE3 and NE4 animals were classed as NN were 86%:89%; 93%:89%; 93%:89%; and 93%:95%, respectively. When NE3 and NE4 animals were classified as PP, calculated SN:SP are 100%:90%; 100%90%; 100%:90%; and 100%:100%, respectively.

Table 2. Tentative identification of seven peaks used to discriminate between CWD-positive, -negative exposed, and -negative deer.

Peak	Retention time	Tentative identification	Compound family
1	12.532	1-Butanol	Aldehyde
2	13.059	4-Methyl 1-pentanol (isohexanol)	Alcohol
3	17.376	6,6-Dimethoxy-2,5,5-trimethyl 2-hexene	Acetal
4	28.343	Hexanoic acid (caproic acid)	Fatty acid
5	28.535	p-Cresol (4-methyl phenol)	Phenol
6	31.403	2-Propyl phenol	Phenylpropane
7	34.737	3-Methyl 1H indole (skatole)	Indole

Compounds were tentatively identified using a ≥ 65% probability match with AMDIS and NIST software.

Figure 1. Three-dimensional PCA scatterplot of CWD-positive, -negative exposed, and -negative deer. All known negative animals from Herd 1 (NN1; green dots) and one from Herd 2 (NN2, black dot) are located in a cluster (green) closely associated with all other NN2 individuals (black cluster). Three Herd 3 CWD-positive (PP3, red dots) and one -negative exposed individual (NE3, blue dot) are not associated with clusters in the plot and represent outliers. Remaining PP3 and NE3 animals and Herd 4 (PP4, pink dots; NE4, aqua dots) animals form closely associated clusters, with the exception of four NE3 animals found within or in close association to the PP3 cluster.

Figure 2. Enlarged view of the area in the PCA scatterplot area containing Herd 3 and 4 CWD-positive and -negative exposed animals. CWD-positive animals from Herd 3 (PP3, red dots) form a distinct cluster containing three negative exposed animals (NE3) from that herd. Remaining NE3 animals form a distinct cluster with the exception of one individual found adjacent to the PP3 cluster. Herd 4 CWD-positive (PP4; pink dots) and -negative exposed (NE4; aqua dots) animals form separate clusters. The close approximation of these clusters indicates that there are some distinct similarities between the cohorts, yet differences between the groups do exist. The three NE3 animals located within the PP3 cluster, and the one NE3 individual located near that cluster may represent animals that were incorrectly classified by our analysis, or may be positive animals infected with a prion burden so low that prion was not detected in the post-mortem IHC analysis performed on the submitted tissues.

Table 3. Potential sources of tentatively identified volatile organic compounds used to discriminate between CWD positive and negative deer.

Tentative identification	Biological location	Biological function	Ruminant source	Disease association	Miscellaneous
1-Butanol	Mitochondria Muscle tissue	Butanoate metabolism [53,56]	Endogenous Food [53] Microbial metabolism [54] Rumen fermentation [55]	Crohn’s disease (CD) [53] Ulcerative colitis (UC) [53]	Lipids, proteins and DNA oxidative damage biomarker [53] Inhibits intercellular gap-junction communication [66]
Isohexanol (4-methyl 1-pentanol)		Lipid metabolism [53,56] Steroid hormone metabolism [53] Pregnenolone biosynthesis [57]	Rumen fermentation [52]		Exists in all eukaryotes [53]
6,6-Dimethoxy-2,5,5-trimethyl 2-hexene	Cell membrane	Cell signalling [53] Membrane stabilizer [53] Energy source and storage [53] Nutrient [53]	Endogenous [53] Food [53]
Hexanoic acid (caproic acid)	Cytoplasm Cell membrane Adiposome	Cell signalling, lipid transport, metabolism, and peroxidation. Β-oxidation of very long chain fatty acids and mitochondrial β-oxidation of short chain fatty acids [53] Energy source and storage [53] Fatty acid biosynthesis [53,67]	Endogenous [53] Food [53] Rumen fermentation [52,58,68] Detected in cattle breath [59]		Found in animal fats and oils; some plants. Abnormal concentrations in humans with autism, pervasive developmental disorder, CD, UC, Immunoglobulin A nephropathy [53]
p-Cresol (4-methyl phenol)	Adipose tissue Cell membrane Fibroblasts	Protein, tyrosine, toluene, phenylalanine metabolism [53,67]	Rumen and intestinal bacteria metabolism of protein [53,60,69]. Present in cattle faeces [49] Present in WTD faeces [50]	Inhibits phagocyte function; decreases leukocyte adhesion to cytokine-stimulated endothelial cells [53] Decreased in M. bovis infection [49,50]	Diminishes oxygen uptake in rat brain. Alters bacterial cell membrane permeability. Blocks cell potassium channels [53]
4-Propyl phenol	Cell membrane		Endogenous [53] Food [53] VOC in ox urine [62] VOC associated Iberian red deer [63]
Skatole (3-methyl-1H indole)	Cell membrane	Tryptophan metabolism [53,67] Bacterial intercellular signal molecule [70] Some derivatives function as neurotransmitters [71]	Endogenous [53] Food [53] Rumen, gastrointestinal metabolite [52] Present in cattle faeces [49] Present in WTD faeces [50]	Abnormal concentration in humans with autism, UC, CD [53] Increased in M. bovis infection [49,50]

Most tentatively identified VOCs serve as endogenous or exogenous nutrients, metabolites or are by-products of microbial or host metabolism. Several compounds (1-butanol, hexanoic acid, p-cresol, skatole) appear related to disease presence or immune system function.

Table 4. Comparison of CWD assay sensitivities (SN) and specificities (SP).

		Assay
Test application	Tissue	IHC	ELISA	PMCA	RT-QuIC^a	References
Post-mortem	Obex	99:100	92:100			[11,75,76]
Post-mortem	MRLN		99:99		100:100	[11,69,75,76]
Ante-mortem	MRPLN	99:100				[11,67,68]
Ante-mortem	RAMALT	68:99			70:94	[11,69]
Ante-mortem	Nasal Brushing				16:91	[11,69]
Ante-mortem	Blood				93:100	[11,70]
Ante-mortem	CSF			19:100	50:96	[11,71]
Ante-mortem	Saliva				78:98	[11,72]
Ante-mortem	Urine				39:100	[11,72]
Ante-mortem	Faeces			NA:NA	53:100	[11,73,74]

Published results of studies evaluating the capability of various assays to detect CWD prion in a variety of samples are presented. Post-mortem (i.e., gold standard) IHC and ELISA SN: SP are included for comparison.

IHC: immunohistochemistry; ELISA: enzyme-linked immunosorbent assay; PAMP: protein misfolding cyclic amplification; RT-QuIC: recombinant PrPc quaking induced conversion assay; MRPLN: medial retropharyngeal lymph node; RAMALT: rectoanal mucosal associated lymphoid tissue; CSF: cerebrospinal fluid.

^aPotential ante-mortem assay.

Table 5. Herd identification, animal classification, CWD status, and the number of animals from each sample group used in the study.

Herd	Animal classification	CWD status	Number of animals
1	Known Negative (NN1)	Non-Detect	10
2	Known Negative (NN2)	Non-Detect	10
3	Negative Exposed (NE3)	Non-Detect	10
3	Positive (PP3)	Positive	10
4	Negative Exposed (NE4)	Non-Detect	7
4	Positive (PP4)	Positive	4

Known Negative (NN): deer not exposed to CWD. Negative Exposed (NE): deer resided with known CWD-positive cohorts but no prion detected by IHC post-mortem. Positive (PP): deer were identified as CWD-positive by post-mortem immunohistochemistry (IHC) in both the medial retropharyngeal lymph nodes (MRPLN) and the obex.

Wishart DS, Jewison T, Guo AC, et al. HMDB 3.0—the human metabolome database in 2013. Nucleic Acids Res. 2012;41(D1):D801–D807.

 PubMed Web of Science ®Google Scholar

Kanehisa M, Goto S. KEGG: kyoto encyclopedia of genes and genomes. Nucleic Acids Res. 2000;28(1):27–30.

 PubMed Web of Science ®Google Scholar

Sauer M. Industrial production of acetone and butanol by fermentation—100 years later. FEMS Microbiol Lett. 2016;363(13):fnw134.

 PubMed Web of Science ®Google Scholar

Parmar NR, Kumar JN, Joshi CG. Deep insights into carbohydrate metabolism in the rumen of Mehsani buffalo at different diet treatments. Genom Data. 2015;6:59–62.

 PubMedGoogle Scholar

DeHaan R, Chen Y. Multiple connexins and asymmetric currents in embryonic cardiac gap junctions. In: Y. Kanno, Y. Kataoka, Y. Shibata, T. Shimazu, editors. Intercellular Communication through Gap Junctions. Progress in Cell Research, (4). Amsterdam: Elsevier; 1995. p. 187–200.

 Google Scholar

Bolton EE, Wang Y, Thiessen PA, et al. PubChem: integrated platform of small molecules and biological activities. Annu Rep Comput Chem. 2008;4:217–241, Elsevier

 Google Scholar

Cai L, Koziel JA, Davis J, et al. Characterization of volatile organic compounds and odors by in-vivo sampling of beef cattle rumen gas, by solid-phase microextraction, and gas chromatography–mass spectrometry–olfactometry. Anal Bioanal Chem. 2006;386(6):1791–1802.

 PubMed Web of Science ®Google Scholar

Aoki KF, Kanehisa M. Using the KEGG database resource. Curr Protoc Bioinform. 2005;11(1) 1–12.

 PubMedGoogle Scholar

Ørskov E, Oltjen R. Influence of carbohydrate and nitrogen sources on the rumen volatile fatty acids and ethanol of cattle fed purified diets. J Nutr. 1967;93(2):222–228.

 PubMedGoogle Scholar

Jeon BS, Choi O, Um Y, et al. Production of medium-chain carboxylic acids by Megasphaera sp. MH with supplemental electron acceptors. Biotechnol Biofuels. 2016;9(1):129.

 PubMedGoogle Scholar

Spinhirne JP, Koziel JA, Chirase NK. Sampling and analysis of volatile organic compounds in bovine breath by solid-phase microextraction and gas chromatography–mass spectrometry. J Chromatogr A. 2004;1025(1):63–69.

 PubMed Web of Science ®Google Scholar

Hamer HM, De Preter V, Windey K, et al. Functional analysis of colonic bacterial metabolism: relevant to health?. Am J Physiol Gastrointest Liver Physiol. 2011;302(1):G1–G9.

 PubMed Web of Science ®Google Scholar

Brunet P, Dou L, Cerini C, et al. Protein-bound uremic retention solutes. Adv Renal Replace Ther. 2003;10(4):310–320.

 PubMedGoogle Scholar

Ellis CK, Rice S, Maurer D, et al. Use of fecal volatile organic compound analysis to discriminate between non-vaccinated and BCG—vaccinated cattle prior to and after Mycobacterium bovis challenge. PloS one. 2017;12(7):e0179914.

 PubMed Web of Science ®Google Scholar

Stahl RS, Ellis CK, Nol P, et al. Fecal Volatile Organic Ccompound Profiles from White-Tailed Deer (Odocoileus virginianus) as Indicators of Mycobacterium bovis Exposure or Mycobacterium bovis Bacille Calmette-Guerin (BCG) Vaccination. PloS One. 2015;10(6):e0129740.

 PubMed Web of Science ®Google Scholar

Bursell E, Gough A, Beevor P, et al. Identification of components of cattle urine attractive to tsetse flies, Glossina spp. (Diptera: glossinidae). Bull Entomol Res. 1988;78(2):281–291.

 Web of Science ®Google Scholar

Martín J, Carranza J, López P, et al. A new sexual signal in rutting male red deer: age related chemical scent constituents in the belly black spot. Mamm Biol-Z Säugetierkunde. 2014;79(6):362–368.

 Web of Science ®Google Scholar

Lee J-H LJ. Indole as an intercellular signal in microbial communities. FEMS Microbiol Rev. 2010;34(4):426–444.

 PubMed Web of Science ®Google Scholar

Nelson DL, Cox MM. Lehninger Princ Biochem Lect Notebook. 4th ed. New York: W. H. Freeman; 2004.

 Google Scholar

Haley NJ, Richt JA. Evolution of Diagnostic Tests for Chronic Wasting Disease, a Naturally Occurring Prion Disease of Cervids. Pathogens. 2017;6(3):35.

 Web of Science ®Google Scholar

Haley NJ, Siepker C, Walter WD, et al. Antemortem detection of chronic wasting disease prions in nasal brush collections and rectal biopsies from white-tailed deer by real time quaking-induced conversion. J Clin Microbiol. 2016;54(4) :1108–1116.

 Web of Science ®Google Scholar

Hibler CP, Wilson KL, Spraker TR, et al. Field validation and assessment of an enzyme-linked immunosorbent assay for detecting chronic wasting disease in mule deer (Odocoileus hemionus), white-tailed deer (Odocoileus virginianus), and Rocky Mountain elk (Cervus elaphus nelsoni). J V Diag Invest. 2003;15(4):311–319.

 PubMed Web of Science ®Google Scholar

Mochalski P, King J, Haas M, et al. Blood and breath profiles of volatile organic compounds in patients with end-stage renal disease. BMC Nephrol. 2014;15(1):43.

 PubMedGoogle Scholar

Spraker T, O‘Rourke KI, Balachandran A, et al. Validation of monoclonal antibody F99/97.6. 1 for immunohistochemical staining of brain and tonsil in mule deer (Odocoileus hemionus) with chronic wasting disease. J Vet Diag Invest. 2002;14(1):3–7.

 PubMed Web of Science ®Google Scholar

Haley NJ, Mathiason CK, Carver S, et al. Sensitivity of protein misfolding cyclic amplification versus immunohistochemistry in ante-mortem detection of chronic wasting disease. J General Virol. 2012;93(5):1141–1150.

 PubMedGoogle Scholar