METHOD AND APPARATUS FOR MONITORING THE STATE OF HEALTH OF DAIRY COWS

Abstract

The present invention relates to methods and apparatuses for monitoring the state of health of dairy cows, in particular of entire dairy herds. The method is based on analysing the haptoglobin (HP) biomarker and part of the polymeric immunoglobulin receptor (PIGR), the secretory component (Secretory Component, SC), in a milk sample. In particular, the claimed method and apparatus of the invention make it possible to diagnose mastitis or systemic diseases which occur outside the udder on the basis of the protein biomarker described here. The invention therefore makes it possible to regularly monitor the general state of health of a dairy herd. The present invention relates to non-invasive diagnostic methods and to apparatuses and diagnostic kits for carrying out these methods.

Claims

1-4. (canceled)

15. A non-invasive system for monitoring the state of health of a dairy cow, comprising: a means to measure concentrations of at least two biomarkers including one of the following biomarker combinations (i) to (vi): (i) haptoglobin and Vascular Endothelial Growth Factor; (ii) haptoglobin and Lactoferrin; (iii) Vascular Endothelial Growth Factor and polymeric immunoglobulin receptor; (iv) Lactoferrin and polymeric immunoglobulin receptor; (v) Lactoferrin and Vascular Endothelial Growth Factor; (vi) haptoglobin and polymeric immunoglobulin receptor; in a milk sample, and a processor configured to compare the measured concentrations of the at least two biomarkers with reference values for the at least two biomarkers, wherein a deviation of the measured concentration from the reference values indicates an unhealthy condition in the dairy cow; and an automated or semi-automated milking system configured to obtain the milk sample from the dairy cow during a milking process.

16. The non-invasive system according to claim 15, further comprising at least one of: a memory and a display.

17. The non-invasive system according to claim 15, wherein the means to measure concentrations of the at least two biomarkers is selected from the group consisting of chemiluminescent immunoassay apparatus, Fast Protein Liquid Chromatography apparatus, Enzyme immunoassay apparatus, Enzyme-linked Immunosorbent Assay apparatus, Electrospray Ionization Mass Spectrometer, Fluorescence immunoassay apparatus, High Performance Liquid Chromatography apparatus, Matrix-assisted Laser Desorption/Ionization Mass Spectrometer, Radio immunoassay apparatus, Sodium dodecyl sulfate polyacrylamide gel electrophoresis apparatus, Surface plasmon resonance apparatus.

18. The non-invasive system of claim 15, wherein the means to measure the concentrations of the at least two biomarkers comprises one of the following (i) to (vi): (i) an antibody specific to haptoglobin and an antibody specific to vascular endothelial growth factor; (ii) an antibody specific to haptoglobin and an antibody specific to Lactoferrin; (iii) an antibody specific to Vascular Endothelial Growth Factor and an antibody specific to polymeric immunoglobulin receptor; (iv) an antibody specific to Lactoferrin and an antibody specific to polymeric immunoglobulin receptor; (v) an antibody specific to Lactoferrin and an antibody specific to Vascular Endothelial Growth Factor; (vi) an antibody specific to haptoglobin and an antibody specific to polymeric immunoglobulin receptor.

19. A diagnostic kit to monitor the state of health of a dairy cow, comprising a means to measure concentrations of at least two biomarkers including one of the following biomarker combinations (i) to (vi): (i) haptoglobin and Vascular Endothelial Growth Factor; (ii) haptoglobin and Lactoferrin; (iii) Vascular Endothelial Growth Factor and polymeric immunoglobulin receptor; (iv) Lactoferrin and polymeric immunoglobulin receptor; (v) Lactoferrin and Vascular Endothelial Growth Factor; (vi) haptoglobin and polymeric immunoglobulin receptor;

20. The diagnostic kit of claim 19, wherein the means to measure the concentrations of the at least two biomarkers comprises one of the following (i) to (vi): (i) an antibody specific to haptoglobin and an antibody specific to vascular endothelial growth factor; (ii) an antibody specific to haptoglobin and an antibody specific to Lactoferrin; (iii) an antibody specific to Vascular Endothelial Growth Factor and an antibody specific to polymeric immunoglobulin receptor; (iv) an antibody specific to Lactoferrin and an antibody specific to polymeric immunoglobulin receptor; (v) an antibody specific to Lactoferrin and an antibody specific to Vascular Endothelial Growth Factor; (vi) an antibody specific to haptoglobin and an antibody specific to polymeric immunoglobulin receptor.

Description

[0045] THE FIGURES SHOW

[0046] FIG. 1: mRNA expression of selected markers in milk cells (MZ) and leucocytes (BL) from cows in various states of disease. The concentration of the markers was determined with qPCR and is given as a percentage of the expression of the reference gene cyclophilin B (PPIB) and ubiquitously expressed transcript (UXT). system.: systemic; Erkrank.: disease; MZ: milk cells; BL: leucocytes; * 0.05>p>0.01, and ** p≤0.01.

[0047] FIG. 2: Concentrations of potential protein biomarkers in milk. The concentrations were determined using commercially available ELISA kits. n.d.=non-detectable; system.: systemic; Erkrank.: disease; * 0.05>p>0.01, and ** p≤0.01.

[0048] FIG. 3: Concentration correlations for the biomarkers HP and LTF in milk and plasma. The concentrations were determined using commercially available ELISA kits. Positive correlations are indicated by the regression lines.

[0049] FIG. 4A: ROC curves from selected milk biomarkers. A: ROC analysis of the individual markers in various states of disease.

[0050] FIG. 4B: Summarized ROC analysis of all sick animals. system.: systemic; Erkrank.: disease

EXAMPLES

[0051] Material and Methods:

[0052] Quantification of Protein Biomarkers in Milk and Plasma

[0053] Selected proteins in milk and plasma were quantified using commercially available ELISA kits. All HP measurements were done based on undiluted samples since this is sufficient to detect fluctuations of the HP marker at various stages of disease. Precoated plates were incubated with 100 μl of sample (30 min, room temperature (RT)). Purified HP (LeeBioSolutions, St. Louis, Mo., USA) was used as the standard in a range from 8 to 0.125 μg/ml. The plate was washed 3 times in assay wash buffer, then incubated with 100 μl of 1:40 diluted peroxidase-conjugated anti-HP antibodies (30 min, RT). After 3 washings, 100 μL of ready-made tetramethylbenzidine substrate solution (Moss Inc., Pasadena, Md., USA) was added, and incubated for 10 to 30 minutes at RT. The reaction was stopped with 50 μl 9.9% H.sub.3PO.sub.4.

[0054] PIGR (SC) was quantified with an ELISA kit to detect bovine PIGR (Life Science USCN Inc.) according to the manufacturer's information. In each case, milk was diluted at a ratio of 1:300 to 1:1,000 for the control samples and 1:5,000 to 1:10,0000 for samples from sick cows. Plasma samples were diluted 1:100,000.

[0055] Statistical Analysis

[0056] Analysis of the differences between the groups was performed by means of Spearman rank correlations, Receiver Operating Characteristic (ROC) analysis and visualization of the results using SigmaPlot11 Software (Systat Software, Erkrath, Germany). To avoid undesired statistical tendencies, animal samples were randomly selected for analysis with quantitative real-time RT-PCR (qPCR) or ELISA. Data sets were analyzed for standard distribution. If the Shapiro-Wilk test returned a positive result, a t test was performed. The Mann-Whitney Rank Sum test was performed for data without standard distribution. All sick groups were compared to the control group. The data for various diseases outside the udder were combined if a small number of samples had been tested. P values are defined as follows: * 0.05>p>0.01, and ** p≤0.01.

[0057] Selection and Evaluation of Potential Biomarkers

[0058] The ROC analysis was used to evaluate the discriminatory ability of the biomarkers. An area under the curve (AUC) >0.9 was regarded as highly discriminating and an AUC value <0.6 as non-discriminating. Biomarkers were selected based on the best distinction between minor systemic disease and the control group. Statistical evaluation of biomarkers and marker combinations was performed using TANAGRA open source data mining software. To avoid potential overfitting, cross-validation (CV) was performed (10-fold, 1 repetition). The values for sensitivity, specificity and resubstitution error rate were taken over from the CV. The various diseases were collected into one group. The biomarkers or their combinations were evaluated on the basis of their ability to discriminate sick cows.

EXAMPLE 1

Differential Gene Expression of Biomarkers in Milk

[0059] The mRNA expression of individual biomarkers in milk cells was analyzed with qPCR. To confirm the systemic significance of potential biomarkers from the local environment of the mammary gland, the expression pattern of the biomarkers in peripheral leucocytes was examined. Data from groups with minor and serious systemic diseases was combined and tested in the case of a small number of samples in a systemic disease group. FIG. 1 shows the results for the most relevant biomarkers.

EXAMPLE 2

Quantification and Selection of Biomarkers

[0060] Based on the results of the previous experiments (microarray, qPCR, etc.), potential biomarkers were selected and quantified at the protein level using commercial ELISA kits. Elevated concentrations of IL-18, LTF, PIGR (SC), TNF-alpha and VEGF were detected in milk in the presence of abomasal displacement, serious systemic disease, mastitis and combinations of the diseases. HP and S100A9, however, showed increased values in the presence of minor systemic disease (FIG. 2). The expression patterns of HP, IL-18 and LTF were also confirmed in plasma in order to determine the validity of the markers for systemic diseases. The correlations of milk and plasma HP and LTF concentrations are shown in FIG. 3. The positive Spearman correlation coefficients (Spearman p) show the relationship between milk and plasma protein concentrations. In addition, the correlation of the strongest biomarkers in the milk was examined. All proteins showed positive correlation of concentrations in the milk in the presence of diseases (Table 1). The best markers underwent further statistical evaluation.

TABLE-US-00001 TABLE 1 Correlations of Protein biomarkers in milk and plasma Spearman Correlation of correlation coefficient p n Correlation in milk Milk HP and milk PIGR (SC) 0.67 0.001 71 Milk LTF and milk PIGR (SC) 0.61 0.001 79 Milk HP and milk LTF 0.59 0.001 142 Milk HP and milk VEGF 0.58 0.001 120 Milk LTF and milk VEGF 0.54 0.001 132 Milk VEGF and milk PIGR (SC) 0.41 0.001 79 Correlation in milk and plasma Milk HP and plasma HP 0.78 0.001 121 Milk IL-18 and plasma IL-18 0.38 0.088 21 Milk LTF and plasma LTF 0.33 0.005 69 Correlation in Plasma Plasma HP and plasma LTF 0.59 0.001 63

EXAMPLE 3

Statistical Evaluation of the Biomarkers

[0061] The heavily regulated and highly concentrated milk biomarkers HP, PIGR (SC), LTF and VEGF were selected for statistical evaluation. A subgroup of samples in which all four markers had been determined was used for a direct comparison of the results.

[0062] Each biomarker alone and combinations of two biomarkers were evaluated. In this regard, 17 control samples and 49 samples from sick cows were used. The discriminatory ability for each disease group was determined by ROC analysis (FIG. 4A, Table 2). HP and PIGR (SC) showed the best distinction of minor systemic disease with an AUC of 0.69 and 0.68. All proteins were highly discriminating for serious systemic diseases and mastitis (AUC>0.9).

TABLE-US-00002 TABLE 2 Discriminatory ability of milk biomarkers for various diseases. The data was generated by means of ROC analysis. (Control: n = 17, minor systemic (system.) disease (Erkrank.): n = 17, Abomasal displacement (LMV) (+metabolic disorder): n = 8, serious systemic disease: n = 5, serious systemic disease + abomasal displacement: n = 8, mastitis: n = 11) 95% Confidence Control vs. Sick group AUC interval p HP Minor systemic disease 0.69 0.48-0.89 0.065 LMV (+metabolic disorder) 0.96 0.89-1.03 <0.001 Serious systemic disease 0.99 0.95-1.03 0.001 Serious systemic disease + LMV 0.99 0.95-1.02 <0.001 Mastitis 1.00 1.00-1.00 <0.001 PIGR (SC) Minor systemic disease 0.68 0.49-0.87 0.071 LMV (+metabolic disorder) 0.84 0.64-1.04 <0.05 Serious systemic disease 0.95 0.87-1.04 <0.05 Serious systemic disease + LMV 0.80 0.61-0.99 <0.05 Mastitis 0.99 0.98-1.01 <0.001 LTF Minor systemic disease 0.67 0.48-0.86 0.088 LMV (+metabolic disorder) 0.82 0.62-1.03 <0.05 Serious systemic disease 0.95 0.86-1.05 <0.05 Serious systemic disease + LMV 0.93 0.84-1.03 <0.001 Mastitis 0.98 0.95-1.02 <0.001 VEGF Minor systemic disease 0.57 0.38-0.77 0.459 LMV (+metabolic disorder) 0.99 0.96-1.02 <0.001 Serious systemic disease 0.84 0.58-1.08 <0.05 Serious systemic disease + LMV 0.96 0.90-1.03 <0.001 Mastitis 0.97 0.91-1.03 <0.001

[0063] To discriminate between sick and control animals, marker combinations were evaluated using two statistical classification methods, namely multinomial logistic regression (MLR) and k-nearest neighbor classification (K-NN) (Table 4). A second statistical model was applied to avoid potential distortions of the results. HP is the best choice for use as a single biomarker. In combination with PIGR (SC) or LTF, a minor increase in sensitivity or specificity can be achieved. These combinations showed the best results for detecting sick animals.

[0064] Practical application of biomarkers requires that the tests have high specificity in order not to overestimate the occurrence of diseases in large dairy cattle herds. An ROC analysis was therefore combined for all sick groups vs. control in order to evaluate the sensitivity (“correct positive”), specificity (“correct negative”), 1-sensitivity (“false negative”) and 1-specificity (“false positive”) of the biomarker determination in milk using various threshold value (cut-off) concentrations. Table 3 shows the values for possible cut-off concentrations with a high specificity of 94%. The corresponding ROC curves are shown in FIG. 4B. At a specificity of 94%, 6% of actually healthy animals would be identified as sick. In the case of determination of HP, PIGR (SC), LTF and VEGF, 18%, 41%, 45% and 33%, respectively, of sick animals would be classified as healthy.

[0065] On the basis of this analysis, it could therefore be demonstrated that the determination of HP is suitable for detecting diseases in dairy cattle. A combined measurement with PIGR (SC) or LTF is also possible in order to increase the sensitivity or specificity.

TABLE-US-00003 TABLE 3 Disciminatory ability of milk biomarkers for sick animals. The data was generated through ROC analysis. (Control: n = 17, sick: n = 49) 95% Confidence- Cut-Off at 94% Sensiivity at 94% AUC interval P specificity specificity % HP 0.88 0.80-0.96 <0.001 0.58 μg/ml 82 PIGR (SC) 0.82 0.72-0.93 <0.001 8.20 μg/ml 59 LTF 0.84 0.74-0.94 <0.001 120.7 μg/ml 55 VEGF 0.82 0.72-0.92 <0.001 9.50 ng/ml 67

TABLE-US-00004 TABLE 4 Evaluation of milk biomarkers and their combinations. The classification was performed by using MLR and K-NN: Control (n = 17) vs. sick (n = 49). Sensitivity, specificity and resubstitution error rates were taken over from the CV (10-fold, 1 repetition). Multinomial logistic regression k-nearest neighbor classification (cross-validation)/% (cross-validation)/% Marker (Combination) Sensitivity Specificity Error rate Sensitivity Specificity Error rate Single marker HP 86 88 13 91 69 15 LTF 84 44 27 82 63 23 VEGF 84 38 28 73 31 38 PIGR (SC) 86 25 30 77 19 38 Marker combinations HP & VEGF 86 88 13 80 94 17 HP & PIGR (SC) 89 81 13 84 75 18 HP & LTF 89 69 17 86 81 15 VEGF & PIGR (SC) 86 63 20 82 56 25 LTF & PIGR (SC) 84 56 23 86 31 28 LTF& VEGF 82 56 25 84 44 27

LIST OF ABBREVIATIONS

[0066] AUC Area Under the Curve

[0067] BL Leucocytes

[0068] CIA Chemiluminescent immunoassay

[0069] CV Cross-validation

[0070] EIA Enzyme immunoassay

[0071] ELISA Enzyme-linked Immunosorbent Assay

[0072] Erkrank. Disease

[0073] ESI Electrospray Ionization

[0074] FIA Fluorescence immunoassay

[0075] FPLC Fast Protein Liquid Chromatography

[0076] HP Haptoglobin

[0077] HPLC High Performance Liquid Chromatography

[0078] Ig Immunoglobulin

[0079] IL Interleukin

[0080] K-NN k-nearest neighbor classification

[0081] LMV Abomasal displacement

[0082] LTF Lactoferrin

[0083] MALDI Matrix-assisted Laser Desorption/Ionization

[0084] MLR Multinomial logistic regression

[0085] mRNA Messenger ribonucleic acid

[0086] MZ Milk cells

[0087] PIGR Polymeric immunoglobulin receptor

[0088] PPIB Cyclophilin B (reference gene)

[0089] RIA Radio immunoassay

[0090] ROC Receiver Operating Characteristic

[0091] S100A9 S100 calcium-binding protein A9

[0092] SC Secretory Component, secretory component of the PIGR

[0093] SDS-PAGE Sodium dodecyl sulfate polyacrylamide gel electrophoresis

[0094] SPR Surface plasmon resonance

[0095] system. Systemic

[0096] TNF-alpha Tumor necrosis factor alpha

[0097] UXT Ubiquitously-Expressed Transcript (reference gene)

[0098] VEGF Vascular Endothelial Growth Factor