Florida State University Libraries

Electronic Theses, Treatises and Dissertations The Graduate School

2017 Immunodetection of Porcine Blood in Foods Xingyi Jiang

Follow this and additional works at the DigiNole: FSU's Digital Repository. For more information, please contact [email protected] FLORIDA STATE UNIVERSITY

COLLEGE OF HUMAN SCIENCES

IMMUNODETECTION OF PORCINE BLOOD IN FOODS

By

XINGYI JIANG

A Thesis submitted to the Department of Nutrition, Food and Exercise Sciences in partial fulfillment of the requirements for the degree of Master of Science

2017

Xingyi Jiang defended this thesis on March 27, 2017. The members of the supervisory committee were:

Qinchun Rao Professor Directing Thesis

Shridhar K. Sathe Committee Member

Timothy M. Logan Committee Member

The Graduate School has verified and approved the above-named committee members, and certifies that the thesis has been approved in accordance with university requirements.

ii ACKNOWLEDGMENTS

Firstly, I would like to express my sincere gratitude to my major professor Dr. Qinchun

Rao for the continuous support of my master study. I could not finish the thesis without him. As my major professor, Dr. Rao taught me all the knowledge and lab techniques. I am very grateful for giving me research assistantship in my master study. I also deeply thank Dr. Timothy Logan and Dr. Shridhar Sathe who served as my committee members. They gave me valuable advice when I was conducting my thesis.

Secondly, I want to thank the entire faculty and staff at College of Human Sciences. They are all very helpful. I sincerely thank the Department of Nutrition, Food and Exercise Sciences for providing the opportunity and financial support to my master study. Particularly, I am grateful for Dr. Shridhar K. Sathe, Dr. Jeong-Su Kim and Dr. Gloria Salazar for letting me use the equipment in their labs. Also, I also thank all the lab mates, Behnam Keshavarz who helped me a lot during researches, Han Mu, Yuyun Wu and Mustafa Samiwala who were always willing support me.

Finally, I would like to show my deepest gratitude to my family members including my parents, Jianhua Jiang and Jing Wang, grandparents, Zhichun Wang and Guiying Xu, my cousin,

Judy Cai. They are supportive of every decision I have made. Without their understanding, I could not get through the difficulties. Also, my friends, although we are not even in the same time zone, they always show up immediately whenever I need them. I believe I will never come to his point without the help of these supportive people.

iii TABLE OF CONTENTS

LIST OF TABLES ...... vi LIST OF FIGURES ...... vii ABSTRACT ...... ix CHAPTER 1 INTRODUCTION ...... 1 1.1 Background ...... 1 1.2 Rationale and Significance ...... 2 1.2.1 Rationale 1: The importance of developing immunoassays for porcine blood detection ...... 3 1.2.2 Rationale 2: The importance of developing and characterizing different antibodies....5 1.2.3 Rationale 3: The importance of studying the stability of target analytes ...... 6 1.3 Objectives and Hypothesis ...... 7 1.3.1 Objective 1: To characterize three antibodies ...... 7 1.3.2 Objective 2: To develop an icELISA for porcine blood detection via porcine hemoglobin ...... 8 1.3.3 Objective 3: To improve a reported sELISA for porcine blood detection via plasma protein ...... 9 CHAPTER 2 REVIEW OF LITERATURE ...... 10 2.1 Porcine Blood ...... 10 2.1.1 Porcine hemoglobin ...... 11 2.2 Porcine Blood Application ...... 15 2.2.1 Animal feed supplement ...... 16 2.2.2 Emulsifier, binder, and gelling agent ...... 17 2.2.3 Nutritional fortification ...... 19 2.2.4 Natural color enhancer ...... 20 2.2.5 Bioactive compounds ...... 21 2.3 Needs for Porcine Blood Detection ...... 22 2.3.1 Religious issues ...... 23 2.3.2 Food safety concerns: Microbial safety ...... 25 2.3.3 Food safety concerns: Diseased pork ...... 26 2.3.4 Food quality concerns: Fraudulent cases ...... 27 2.3.5 Health and environmental concerns ...... 28 2.4 Porcine Blood Detection Methods ...... 29 2.4.1 Chromatography ...... 29 2.4.2 Polymerase chain reaction (PCR)...... 30 2.4.3 Spectrophotometry ...... 31 2.4.4 Immunoassay ...... 32 CHAPTER 3 INDIRECT COMPETITIVE ELISA FOR THE DETECTION OF PORCINE BLOOD IN FOODS ...... 34 3.1 Materials ...... 34 3.2 Methods ...... 36 3.2.1 Production of mAbs ...... 36 3.2.2 Protein extracts preparation ...... 38 3.2.3 Indirect non-competitive ELISA (inELISA) ...... 39 3.2.4 SDS-PAGE and Western blot ...... 40

iv 3.2.5 Aqueous two-phase system (ATPS) ...... 42 3.2.6 Gel filtration chromatography ...... 42 3.2.7 Gradient gel electrophoresis ...... 43 3.2.8 Two-dimensional gel electrophoresis ...... 43 3.2.9 icELISA ...... 44 3.2.10 Data analysis ...... 45 3.3 Results and Discussion ...... 45 3.3.1 Antibody screening ...... 45 3.3.2 Antibody purification ...... 46 3.3.3 Antibody characterization ...... 46 3.3.4 Immunoaffinity ...... 50 3.3.5 Hemoglobin stability ...... 54 3.3.6 Assay development ...... 59 3.4 Conclusions ...... 66 3.5 Limitations ...... 68 CHAPTER 4 . CHARACTERIZATION OF MONOCLONAL ANTIBODIES SPECIFIC FOR PORCINE PLASMA ...... 70 4.1 Materials ...... 70 4.2 Methods ...... 72 4.2.1 Production of mAbs ...... 72 4.2.2 Protein extracts preparation ...... 74 4.2.3 SDS-PAGE and Western blot ...... 74 4.2.4 Two-dimensional gel electrophoresis ...... 76 4.2.5 Target protein isolation ...... 77 4.2.6 Target protein sequence ...... 78 4.2.7 Data analysis ...... 79 4.3 Results and Discussion ...... 79 4.3.1 Target analyte characterization ...... 79 4.4 Conclusions ...... 83 APPENDICES ...... 84 A. TABLES...... 84 B. FIGURES ...... 102 REFERENCES ...... 131 BIOGRAPHICAL SKETCH ...... 155

v LIST OF TABLES

Table 1. List of worldwide incidents worldwide related to pork/porcine blood...... 84

Table 2. Characteristics of mAbs that are specific to heat-treated porcine blood...... 85

Table 3. Characteristics of mAb19C5 and mAb16F9...... 86

Table 4. Major porcine blood proteins...... 87

Table 5. Comparison of hemoglobin between different species...... 88

Table 6. Porcine blood application...... 90

Table 7. Porcine blood detection methods...... 92

Table 8. Commercial ELISA kits for the porcine blood detection...... 94

Table 9. Reagents formula...... 96

Table 10. The water content of porcine blood...... 98

Table 11. The affinity constant (M-1) of mAb13F7/C8...... 98

Table 12. Coefficient of variations (CVs) of standard curve established by icELISA using mAb13F7/C8...... 99

Table 13. Recovery of porcine hemoglobin extracted by extraction buffer...... 99

Table 14. Summary of extraction buffers...... 100

Table 15. Summary of commercial antibodies used in Western blot...... 101

vi LIST OF FIGURES

Figure 1. Illustration of pig farm and pork...... 102

Figure 2. The process of pig slaughtering...... 103

Figure 3. Three-dimensional structure of porcine hemoglobin...... 104

Figure 4. Two-dimensional structure of heme moiety...... 105

Figure 5. Illustration of research objective and methodology...... 106

Figure 6. Screening test for mAb13F7/C8 using indirect non-competitive ELISA...... 107

Figure 7. SDS-PAGE of purified mAb13F7/C8 IgG...... 108

Figure 8. SDS-PAGE and Western blot to study the selectivity of the mAb...... 109

Figure 9. SDS-PAGE and Western blot to study the target analyte of mAb13F7...... 110

Figure 10. Western blot to study the cross-reactivity of porcine blood IgG with anti-IgG- HRP...... 111

Figure 11. Gradient gel electrophoresis and Western blot...... 112

Figure 12. Isolation of porcine hemoglobin using aqueous two-phase system (APTS)...... 113

Figure 13. Column chromatography purification and SDS-PAGE...... 114

Figure 14. Two-dimensional gel electrophoresis of heated whole porcine blood...... 115

Figure 15. Comparison of immunoaffinity of mAb13F7 to porcine and bovine hemoglobin using inELISA and icELISA...... 116

Figure 16. Two-dimensional titration using indirect non-competitive ELISA...... 117

Figure 17. Effect of pH on thermostability and storage stability of porcine hemoglobin...... 118

Figure 18. The effect of pH on molecular integrity and storage stability of porcine hemoglobin...... 119

Figure 19. SDS-PAGE and Western blot to study the effect of pH on thermostability and storage stability of isolated porcine hemoglobin...... 120

Figure 20. A standard curve of indirect competitive ELISA using mAb13F7...... 121

vii Figure 21. A415 of 0% inhibition of different extraction buffers...... 122

Figure 22. Image of laboratory adulterated samples...... 123

Figure 23. Western blot to determine porcine hemoglobin concentration in laboratory spiked samples...... 124

Figure 24. Western blot using different commercial antibodies...... 125

Figure 25. Reducing and non-reducing SDS-PAGE...... 126

Figure 26. Two-dimensional gel electrophoresis of heated porcine plasma proteins...... 127

Figure 27. SDS-PAGE and Western blot to verify isolated target analyte purity and immunoreactivity...... 128

Figure 28. Western blot using commercial anti-haptoglobin mAb...... 129

Figure β9. Western blot with protein treated with -mercaptoethanol at different concentrations...... 130

viii ABSTRACT

Different porcine blood proteins have been widely used as emulsifier, binder, and colorant in processed foods. However, misusage of porcine blood ingredients, such as mislabeling and substitution, can cause religious objections, law violation, food safety and food quality decrease. These issues highlight the need for detecting unfavorable porcine blood in foods to fight food fraud. Porcine blood plasma and blood cells can be applied individually or in combination as food additives. Therefore, the study was divided into two parts. The objectives of

Part 1 were (1) to develop and characterize a monoclonal antibody (mAb) that is specific for porcine hemoglobin; and (2) to develop an indirect competitive enzyme-linked immunosorbent assay (icELISA) that can detect porcine blood adulteration in foods. The objective of Part 2 was to characterize two mAbs which have the target protein in porcine blood plasma.

In Part 1, mAbs were developed using hybridoma technique and purified using immunoaffinity. Western blot was applied to verify the target protein; to study the mAb selectivity; and to study the effect of pH on the thermostability (50 ºC, 100 °C and 121 ºC for 15 min, respectively) and storage stability (29 days at 4 °C) of the target protein. Indirect non- competitive ELISA (inELISA) was performed to study antibody affinity and storage stability of target protein, and to choose the optimized conditions for icELISA. Finally, an extraction method and an optimized icELISA were developed. The assay was validated by the US FDA’s guidance.

In Part 2, immunoaffinity column was applied to isolate the target protein. The isolated proteins were sequenced using mass spectrometry. The immunoreactivity of the target protein was verified by Western blot using four commercial antibodies (anti-transferrin-IgG, anti- haptoglobin-IgG, anti-plasminogen-IgY and anti-C7-IgG). Two methods (two-dimensional gel

ix electrophoresis and non-reducing SDS-PAGE) were performed to further investigate the isoelectric point (pI) and disulfide information of target protein, respectively.

As to the results of this study, in Part 1, mAb13F7 was chosen after screening test because it had the best selectivity to porcine blood. The target protein of the mAb was porcine hemoglobin (PHb) subunit (14 kDa). Although this mAb could cross-react with hemoglobin from bovine, horse and sheep, their hemoglobin subunit band color intensity was much less than that of PHb according to Western blot. From inELISA and icELISA, this mAb showed a high immunoaffinity to PHb compared with bovine hemoglobin. The affinity constant of this mAb is in a nanomolar range, which can be considered as a high-affinity antibody. As for thermostability,

PHb can retain the best molecular integrity and immunoreactivity at alkaline pH compared to acidic and neutral pHs. During storage at 4 ºC up to 29 days, PHb retained intact without any degraded peptides observed, and its immunoreactivity did not change significantly (P > 0.05).

Finally, a sample extraction buffer (12.5 mM NaHCO3 and 25 Mm NaCl, pH 8.3) and an anti-PHb icELISA were developed. After assay validation, the optimized icELISA was PHb-specific and had a working range from 0.5 ppm to 1000 ppm. This assay was sensitive (limit of detection: 0.5 ppm) and reproducible with low intra- and inter-assay coefficient of variances (CVs < 15%).

The established icELISA assay in Part 1 can be used to detect trace amount of porcine blood in foods to fight food fraud. It also has the potential to be used in identifying diseased pork through determining residual hemoglobin concentration in pork. It is suitable for (1) government to enhance food regulation; (2) food industry to surveillance product quality; and (3) third-party authority to certify halal/kosher foods or evaluate food authenticity.

In Part 2, the target protein was successfully isolated, and its purity and immunoreactivity has been confirmed. Target protein amino acid sequence was obtained. In total, four commercial

x antibodies were tested, neither of them showed similar band pattern as mAb19C5 and mAb16F9.

The target analyte of these two mAbs is still under investigation.

xi CHAPTER 1

INTRODUCTION

1.1 Background

Porcine blood can be made into foods such as curd, cakes, sausages, pudding and soup. It is also used as an additive in the food industry. It is composed of plasma (~ 56%) and blood cells

(~ 44%) [1]. Plasma proteins have excellent water holding capacity, foaming, emulsifying and gelling properties [2]; they can be used as binder, emulsifier and gelling agent in many meat products such as sausages, meatballs, and meat pies. For blood cells, hemoglobin accounts for approximate 90% of the total protein in cells [3], it can be applied as color enhancer and nutritional supplement due to the unique color and the heme iron [4].

However, blood can be harmful to humans and environment. First, the addition of porcine blood in foods hurt certain people’s interests. For example, Jews and Muslims are forbidden to consume porcine blood-contained products due to their religious beliefs. Second, consumption of raw or under heated porcine blood can cause foodborne illness such as Streptococcus suis [5].

Finally, random disposal of blood will cause water and land pollution [6].

These issues highlight the need for porcine blood detection. Currently, polymerase chain reaction (PCR), spectrometry, and immunoassay are developed to detect blood adulteration in foods. For spectrometry, they have great sensitivity but requires operational skills. Immunoassays have been successfully applied in blood detection in foods [7; 8]. Immunodetection of porcine blood lacks until several specific antibodies are developed [9; 10]. However, all of them have target analytes in plasma, making blood detection via cells impossible. Therefore, efficient, reliable and sensitive immunoassays are in need because they can (1) detect both plasma and cells;

(2) protect customers; (3) improve food safety and quality.

1 1.2 Rationale and Significance

The significance of this research is based on different customers’ needs. Certain people like Muslims and Jews are forbidden to consume porcine blood-derived foods according to halal and kosher food laws [11]. They are increasingly concerned about the undeclared blood components in foods. This is because current food labeling law on blood additives is incomplete.

For example, it is voluntary to illustrate animal origin on the label in the US [12; 13]. Therefore, religious consumers are under risks of consuming porcine blood.

Food safety and quality are always a top concern for consumers. From the view of food safety, the high nutritive components, favorable pH, and temperature make blood an ideal medium for microorganism growth [14]. Foods made with infected porcine blood can lead to foodborne illness such as infection of Streptococcus suis [5] and cholera [15]. In addition, circulation of diseased pork (meat from pigs died due to illness before slaughtering) on the market is a major concern in developing countries such as China and Vietnam [16]. The pig mortality during raising is above 20% worldwide [17]. To reduce loss manufactures either sell meat directly on the market or use into foods such as meatballs, jerky, pasty (meat pie), etc. This behavior not only violates food safety laws but also can cause foodborne illness. The current method to differentiate diseased pork is sensory evaluation [18]. Fresh pork (Figure 1c) is pinkish while diseased one is dark with blood remained in tissues (Figure 1d). However, the color difference is difficult to tell when the pork has been processed. The color enhancer and heat treatment might eliminate the color difference between healthy and diseased pork. For these reasons, the regulatory departments need a reliable and robust detection method to identify diseased pork. Hemoglobin can be used as an index to indicate blood residual in pork tissues

[19]. This is because pigs die from natural causes has a much higher level of hemoglobin

2 compared to the normally slaughtered pigs [20]. This high concentration is caused by the difficulty in the bleeding process. Therefore, it is viable to identify diseased pork via hemoglobin concentration in pork.

From food quality view, porcine blood proteins can cause fraudulent cases. They can function as protein substitutions which increase the economic gain to manufacturers but decrease meat nutritional value for the customers. For example, transglutaminase from porcine blood works as a binder in the meat industry. Its usage makes people consuming scraps of meat

“glued” together instead of a genuine piece of meat [21; 22; 23]. The significance of the proposed research is: (1) to protect customers; (2) to improve food safety and quality. The proposed research can be applied (1) to help government regulate meat products on the market and to monitor food safety and quality; (2) to be used by industry as quality control; (3) to be used by third-party to monitor food safety and quality on the market. The significance is based on the following rationale.

1.2.1 Rationale 1: The importance of developing immunoassays for porcine blood detection

Currently, adulterated porcine blood detection methods are developed using chromatography plus detector [24], polymerase chain reaction (PCR) [25], spectrometry [26] and immunoassays [8]. For chromatography plua detector and spectrometry, there are several drawbacks. First, they require expensive and sensitive instruments, such as high-performance liquid chromatography (HPLC) and mass spectrophotometer (MS). Therefore, conducting the experiments requires professional technicians. Second, chromatography involves the ingredient separation between two phases while spectrometry involves the absorbance at a specific

3 wavelength for certain substance. The procedures suffer from difficulties in sample preparation, laborious work, and usage of expensive reagent. Based on these shortcomings, their application is restricted when it comes to on-site detection. PCR is used in the detection of pork adulteration

[27] or porcine blood virus [28; 29]. It can be applied in identifying the pathogens in diseased pork. However, there is one concern that when blood is adulterated with other food ingredients, the trace amount of virus requires a great sensitivity of the assay. According to Ramahefarisoa,

Rakotondrazaka, Jambou, et al. (2010), compared with ELISA assays, PCR showed high specificity but a low sensitivity. A minimum of 10-13 g of DNA was detectable. Similarly, for identifying diseased pork, a rapid peroxidase-based detection kit was developed recently, but its application in processed foods is limited because temperature and pH can inactivate peroxidase

[31].

Immunochemical techniques especially enzyme-linked immunosorbent assays (ELISAs)

[7] have been successfully applied to animal blood detection in foods. It is easy to operate, sensitive and reliable. Therefore, ELISA will be chosen in this study for the detection of porcine blood in foods based on following advantages. First, ELISAs have different formats including sandwich ELISA (sELISA), indirect/direct competitive ELISA (icELISA/dcELISA) and indirect/direct non-competitive ELISA (inELISA/dnELISA). Based on the available conditions

(antibodies, assay objectives, etc.), the suitable format can be chosen. Second, ELISAs can be developed into commercial kits which are easy to operate and can finish detection in less than 6 hours. The kits have the potential to be applied to on-site detection. Third, ELISA is a sensitive, accurate and reliable method which can be applied to a large-scale screening test [32].

4 1.2.2 Rationale 2: The importance of developing and characterizing different antibodies

mAbs are superior to polyclonal antibodies (pAbs) in its homogeneity and consistency.

Although producing pAbs is cheap and fast, large amounts of non-specific antibodies might be generated. Moreover, pAbs are different from batch to batch. Compared to pAbs, mAbs have the advantages of specificity, long-time usage and uniform quality [33]. mAbs are useful in evaluating changes in molecular conformation, protein-protein interactions, and phosphorylation states and in identifying single members of protein families [34]. Therefore, it is important to develop mAbs.

Several mAbs have been developed which are specific to porcine blood (Table 2 and

Table 3). Two of these have been applied in a sELISA to detect porcine blood/spray-dried porcine plasma in animal meat [9; 10]. However, those antibodies have not been characterized yet. Knowing the target analyte is beneficial to the assay development in the following aspects.

First, it determines the application range of the assay. Porcine blood cells and plasma can be used individually or in combination in foods. Choosing the right antibodies can detect porcine whole blood, plasma, and blood cells. Second, it provides variable ways to establish an assay. sELISA can be established using two antibodies targeted at same protein with different epitopes. icELISA can be performed using a single antibody. For both formats, a standard curve constructed by target protein will be established so that target protein concentration in foods can be determined using the curve. Third, knowing target analytes provides further information about antibody characteristics. Antibody affinity can be obtained via ELISA [35; 36], isothermal titration calorimetry (ITC) [37], surface plasmon resonance (SPR) [38]. Antibodies can be categorized into high- or low- affinity antibody based on their affinity constants. The high-affinity antibody

5 is proved to be superior in a variety of biological reactions [39]. Besides, the sensitivity and working range of the assay are also related to the antibody affinity [40].

1.2.3 Rationale 3: The importance of studying the stability of target analytes

The application of the assay is based on a reliable and sensitive detection of porcine blood in foods, more specifically, target analyte in porcine blood. The target analyte should be stable and detectable after processing. Stability is defined as protein’s resistance to adverse effects such as heating or denaturant. The protein should retain its molecular integrity or immunoreactivity after such treatment [41]. The stability of the proteins can be categorized into thermal stability, pH stability and storage stability. First, many foods sold on the market have a trend in developing into

“ready to eat” format, they have been heated before packaging and entering the market [42]. In addition, food additives such as plasma powder and hemoglobin powder are produced through high-temperature spray drying [43]. Therefore, the target analytes are expected to be thermal stable so that they are detectable in final products. Second, food systems are complex with some factors to cater customers’ sensory and flavor requirement; pH is one of the major factors. For example, fermented sausages prefer an acid pH due to lactic acid bacteria, while iron-fortified cereal usually has a neutral pH [44]. pH can affect the solubility and the shape of target proteins, which may lead to changes in immunoreactivity. Third, some foods especially meat are required to be frozen before cooking. However, proteins might degrade or denature during storage, making target proteins undetectable [41]. It is, therefore, necessary for the target proteins to have strong storage ability so that they can be detectable after long-term storage.

6 1.3 Objectives and Hypothesis

The overall goal of this research is immunodetection of porcine blood in foods. More specifically, the objective can be divided into two parts. Part 1, a mAb which has porcine hemoglobin as its target analyte will be developed and selected. An icELISA will be developed to detect porcine blood in foods through porcine hemoglobin; this assay will be applied to detect blood adulteration and have the potential to identify diseased pork. In part 2, an established sELISA will be improved to detect adulterated porcine blood in foods via plasma protein. The optimization will be focused on characterizing the target protein and improving extraction buffer.

There are three specific objectives and with their hypotheses were illustrated below.

1.3.1 Objective 1: To characterize three antibodies

Specific aim 1: To know the target analytes of three antibodies. The molecular weights and pIs of target proteins will be obtained through sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE), Western blot and two-dimensional gel electrophoresis. For Part 1 study, a mAb has porcine hemoglobin as its target analyte will be selected and verified. For Part

2, target protein of mAb19C5 and mAb16F9 can be isolated using immunoaffinity. Briefly, activated agarose contains N-hydroxysuccinimide (NHS) ester functional groups will form stable linkages with the antibody. After antibody immobilization and blocking, the protein sample

(porcine plasma proteins) will be loaded, and target protein will be isolated by immunoaffinity.

The isolated protein will be sequenced at the Translational Laboratory in College of Medicine,

Florida State University.

Specific aim 2: To study the stability of target analyte. The immunoreactivity and molecular integrity of target protein of mAb should be retained after heat treatment at different

7 temperatures. However, they can be affected by pH and storage conditions. For storage stability study, target analyte will be prepared in 50 mM Tris-HCl at different pHs (acidic, neutral and alkaline) and stored at 4 ºC up to 29 days in this study. To study thermostability, target analyte will be heated at different temperatures (50 ºC, 100 ºC and 121 ºC). The molecular integrity, immunoreactivity, and antigenicity will be studied via SDS-PAGE, ELISA and Western blot. Also, extraction buffer will be studied by itss ability to extract adulterated porcine blood in the meat.

Specific aim 3: To study the affinity constants of three mAbs. The affinity constants can be obtained via inELISA and icELISA. For inELISA, a microplate will be serially coated with the target protein and then incubated with mAb serially diluted in antibody buffer. The affinity constant will be obtained according to Beatty, Beatty, and Vlahos (1987). For icELISA, it will be performed as described by Friguet, Chaffotte, Djavadiohaniance, et al. (1985). Affinity constants from two experiments will be compared. Currently, no research has been done to study the affinity constant of the same antibody using two formats of ELISA to the best of our knowledge.

It is expected to see the affinity constants from two methods are not significantly different.

1.3.2 Objective 2: To develop an icELISA for porcine blood detection via porcine hemoglobin

An icELISA will be developed using previously described mAb to detect porcine blood adulteration in foods. The selectivity of this mAb will be studied using different animal meat/blood proteins via inELISA and Western blot. inELISA will also be performed using a two- dimensional titration (antigen and antibody) to select the optimum concentration of immobilized antigen and primary antibody concentration in the icELISA. The optimal coating amount and

8 immunoglobulin (IgG) concentration will be chosen. The optimized icELISA will be used to detect trace amount of porcine blood in foods and have the potential to identify diseased pork.

1.3.3 Objective 3: To improve a reported sELISA for porcine blood detection via plasma protein

mAb19C5 and mAb16F9 have been successfully applied in a sELISA to detect porcine whole blood/plasma in animal meat [9] (Table 3). Although this assay has low limit of detection, the target analyte of these two antibodies is unknown. In addition, this assay did not cover blood detection in other widely consumed meat such as pork, beef, and lamb. Therefore, sELISA will be improved based on target protein’s properties and buffer extraction ability. The optimized sELISA will be able to detect porcine blood adulterated in pork and beef with low detection limit.

9 CHAPTER 2

REVIEW OF LITERATURE

2.1 Porcine Blood

In 2015, 115.5 million heads of pig were slaughtered in the US, a 8% higher compared to

2014 [45]. Figure 2 illustrates the diagram starting from pig slaughtering to by-products collection.

Considering blood represents up to 4% of the live animal weight, a huge amount of porcine blood is produced each year [47]. Porcine blood is a red fluid which contains approximate 80% water,

17% to 20% protein, 0.1% to 0.3% fat and 0.8% ash [48]. The constituents of porcine blood can be classified into plasma and cellular elements, which accounts for around 56% and 44% of whole blood, respectively. Blood fractions can be separated by centrifugation or allowing the blood to clot [47].

In blood cells, the most important elements are red corpuscles (erythrocytes, red blood cells), white corpuscles (leukocytes, white blood cells), and platelets. One liter of porcine blood contains 6 to 8 million red blood cells. The composition, shape, and functions of red blood cells are different among animal species. For example, they are round in porcine blood but are oval- shaped in deer and camel blood. In addition, porcine blood cells can be separated faster than that of cattle and sheep [47]. Among blood cells, hemoglobin accounts for more than 90% of total protein [3]. As one of the most important blood proteins, hemoglobin will be discussed in details.

Blood is easy to clot; the clotting can be prevented by adding anticoagulants such as sodium citrate, ethylenediaminetetraacetic acid (EDTA) or heparin. Plasma, the part of the blood left after removal of cells from unclotted blood consists of 6% to 8% proteins depending on the animal. Major porcine plasma proteins include albumin (3.8%), globulins (2.1%) and fibrinogen

(0.7%). Globulins can be divided into alpha-, beta-, and gamma-globulins, which account for

10 approximate 0.6%, 0.5% and 1.0% in whole porcine blood, respectively. Among all the globulins, IgGs take up the most significant amount [1]. Serum differs from plasma in that fibrinogen has been removed. The removal is completed by converting fibrinogen into fibrin, and then deplete the clots or aggregates. Several other proteins are also removed via specific or non- specific interactions [49]. Major porcine blood proteins are summarized in Table 4, and it should be noted that more than 100 other proteins have been characterized in porcine blood but not illustrated here [50].

2.1.1 Porcine hemoglobin

Hemoglobin is the most important protein in red blood cells which represents more than half of the total blood proteins [4]. Porcine hemoglobin has a molecular weight approximately of

64,500 Da, accounts for 14.2% of whole porcine blood [47]. Its structure, functions and sequence comparison between other species will be illustrated below.

Structure. Normal porcine hemoglobin is a tetramer composed of two identical chains named α-chain and -chain, with a molecular weight of 16,166 Da and 15,039 Da respectively

(Figure 3). α-chains and -chains are linked by non-covalent bonds [51; 52]. Each α-chain contains seven helical segments (A, B, C, E, F, G, H) and seven non-helical segments (NA, AB,

CE, EF, FG, GH, HC); while each -chain contains eight helical segments (A, B, C, D, E, F, G,

H) and six non-helical segments (NA, CD, EF, FG, GH, HC) [53]. Each chain also contains a heme group (Figure 4). When hemoglobin is heated in water, the protein denatures, and the heme is released [46]. The heme unit in hemoglobin from different animal species is the same. It contains an organic component named protoporphyrin and a central iron atom bound to the four pyrrole nitrogen atoms [54]. In the structure of porcine hemoglobin, all four heme irons lie in the

11 heme plane. However, the heme group of the -chains lies slightly further away from the plane than the heme groups of α-chains. The atoms of each heme group interact not only with the

“local environment” but also with surrounding proteins with van der Waals force [53]. The heme groups of the α1, α2 subunits contact 17 and 11 residuals respectively while the 1 and 2 subunits both contact 16 residuals [53].

Once equivalent amounts of α and subunits are obtained, and heme group is inserted, the functional α22-tetramer begins to assemble. In normal circumstances, one -chain will first non-covalently combine with one α-chain to form a α-dimer:

Two α-dimers will form a tetramer. +It should → be noted that the formation of the dimer is irreversible while the tetramer can be reversed easily. This tetramer is named hemoglobin A

(HbA) which accounts for more than 90% of total hemoglobin. Subunit interaction is also the reason of different mutations and variations of hemoglobin which take up the remained 10%.

Fetal hemoglobin (HbF, α22) is composed of two α chains and two chains, which is the main oxygen transport protein in fetus stage [55]. In addition, if -chains aggregate to form 4- tetramer, hemoglobin H (HbH) will be formed.

Functions. Hemoglobin is responsible for the red color of blood, more specifically, when heme binds to iron, the iron interacts with oxygen to form the redness. This interaction leads to another function of hemoglobin: to carry oxygen from the lungs to all the other tissues in the body and returns carbon dioxide from the tissues back to the lung. The ferrous atom in the middle has four of its six coordination sites occupied by bonds with porphyrin. The fifth and the sixth coordination positions are available for binding with O2 or other small molecules [54]. Based on the state of iron, hemoglobin can be classified into ferrohemoglobin and ferrihemoglobin.

12 Ferrohemoglobin, or reduced hemoglobin, refers to the one which contains reduced Fe (Fe2+).

Oxyhemoglobin (HbO2), carboxyhemoglobin (HbCO) and carbominohemoglobin (HbCO2) all belong to ferrohemoglobin. On the other hand, ferrihemoglobin such as methemoglobin (MetHb) refers to the hemoglobin which contains iron in the +3 oxidation state (Fe3+) [54]. This oxidation state makes it unable to bind with oxygen or other small molecules. In this case, heme unit oxidizes into haemin which is a dark brownish pigment. Besides providing color and transporting oxygen, hemoglobin has other biological and physiological functions including (1) keep animal internal temperature stable; (2) work as a modulator of erythrocyte metabolism; (3) interaction with drugs;

(4) as a source of physiological active catabolites [56].

Hemoglobin comparison between different species. Hemoglobin is found in all vertebrates and several invertebrates, however, its concentration is different among animal species. Hall and Gray (1929) found that higher hemoglobin concentration was observed in more active animals. This is also reported by Larimer (1959) that human, dog, and cow have higher hemoglobin concentration than that of cat, sheep, and goat.

HbA is composed of four polypeptides (two α-chains and two -chains). The amino acid sequence of the polypeptides is determined by genes. In porcine HbA, it is coded by two different genes, HBA and HBB. In other animals, more hemoglobin phenotypes can be found.

For example, HbC is found in sheep whose -chain of is coded by HBBC [59]. In addition, a minimum of five hemoglobins is found in chicken and at least four genes (HBB, HBAA, HBAD,

HBE) are involved in coding [60]. The genes cause the differences in amino acid sequence, further lead to the differences in protein functions and properties. The comparison of hemoglobin between different animals and its sequence similarity are summarized in Table 5.

13 Techniques for separating hemoglobin subunits. Electrophoretic technology is widely used to differentiate hemoglobin subunits. First, isoelectric focusing is chosen regards to the different charges of subunits. This method can resolve proteins whose pIs differ by as little as 0.02 pH unit

[61]. The pIs of porcine hemoglobin α-chain and -chain are 7.10 and 8.76, respectively [50].

Researchers have used this method to separate subunits from human, cat, rabbit, dog, etc. [62; 63;

64]. The resolution of bands can be further improved by adding Triton X-100 [65]. The subunits can even be recovered by immersing the band in 2% (v/v) acetic acid [66].

Second, a 15%-20% gradient SDS polyacrylamide gel with specific running conditions (5

V for 1-2 h, 35 V for 5 h) can produce good separation [67]. In addition, a high concentration of urea can be applied to sample and gel preparation. Urea is a denaturant which dissociates non- covalently associated subunits [66]. When hemoglobin is incubated in a denaturing solution of 6

M urea and 1 M -mercaptoethanol and analyzed in an acrylamide slab gel containing 6 M urea

[68], the two subunits can be successfully separated.

Third, hemoglobin subunits can be separated using two types of capillary electrophoresis.

The first one is capillary zone electrophoresis (CZE), in which hemoglobin moves via electroosmotic flow and separates into subunits because of different electrophoretic. It can also be run under denaturing conditions [69]. The effect of separation is influenced by buffer salts, their concentration, and pH [70]. Generally, Tris or borate buffer is chosen to create a pH level higher than the pI of hemoglobin. The second one is capillary isoelectric focusing (CIEF), which arranges hemoglobin in a pH gradient. Hemoglobin subunits have been successfully separated by different researchers using this method [71; 72]. It should be noted that all the electrophoretic methods can also be applied to identify variants and mutations.

14 Chromatography is another category of the methods used to fractionate hemoglobin subunits. Three kinds of chromatography will be illustrated as follows. First, hemoglobin subunits can be separated by gel filtration or size exclusion chromatography. Different types of

Sephadex are loaded to column and subunits come out successively based on their molecular weights [73; 74]. Typically, Sephadex G-50 and Sephadex G-75 are chosen because their fractionation range covers the molecular weight of hemoglobin [75]. This method usually requires a long column so that the time of different subunits come out can be spaced to the greatest extent. Second, reversed-phase chromatography (RP-HPLC) is applied by immobilizing a hydrophobic phase on the column, and the separation is based on the hydrophilicity of each chain. This technology usually gives excellent separation, high efficiency and requires simple sample preparation [76; 77; 78]. Third, ion-exchange chromatography, typically, cation- exchange HPLC is used to identify different hemoglobin fractions [79; 80; 81].

2.2 Porcine Blood Application

Porcine blood has been widely used as an ingredient in foods [82]. In western countries, it is used in preparing blood pudding and blood sausages [83]. In Asian countries, porcine blood can be applied in foods with multiple formats. In Vietnam, traditional family celebrations usually involve the consumption of tiet canh, which contains coagulated, fresh and uncooked porcine blood mixed with chopped pork tissues [5]. Porcine blood curd, also called blood tofu is widely consumed in Taiwan, China. It is composed of fresh porcine blood and water coagulated by heat treatment [84]. Another well-known porcine blood dish is rice cake, which is made by steaming a mixture of blood and water-soaked glutinous rice [85]. Dinuguan (pork blood stew, or chocolate meat) is a dish popular in Philippines. It is made from porcine blood and other innards

15 which are stewed together in vinegar [86]. Aside from being made into foods, it can also be applied as food/feed additives in industry or be applied in medical/clinical/pharmaceutical area

(Table 6).

2.2.1 Animal feed supplement

Spray-dried porcine plasma is used as animal feed to enhance growth rate and meat production. Researchers have discovered that porcine plasma works as a protein supplement in the diets of pigs [87; 88; 89]. Hansen, Nelssen, Goodband, et al. (1993) found that porcine plasma is better compared to skim milk and it gives a positive effect on pigs growing performance. Pierce, Cromwell, Lindemann, et al. (2005) also showed that porcine plasma is advantageous to young pig performance during the first seven days after weaning, and the IgG fraction is responsible for the enhancement in growth rate and feed intake [87]. It should be noted that all researches were conducted at weaning pigs. It makes sense because animals have the most significant growing rate during the weaning period.

Hemoglobin can also function as a supplement in the aquatic feed. SONAC BV

(Netherland) is a manufacturer of pet food and feed made from proteins of animal origin. They discover that the spray dried porcine hemoglobin powder has a higher protein content and is easier for fish/shrimp to digest. It is recommended that porcine hemoglobin powder is included in the fish meal to obtain optimal growth results [91]. However, contradictory findings are obtained from other researchers [92; 93; 94]. In a six-week trial on growth performance of

Litopenaeus vannamei (Pacific white shrimp), as hemoglobin concentration increased in the meal, the shrimp growth became slower. Even a 12.5% substitution led to lower final weight, less feed intake, smaller protein efficiency ratio and productive protein value [92]. Another study

16 conducted on sea bream also showed hemoglobin inclusion in diets reduced growth performance

[93]. The completely different results suggest that growth performance depends on the animal’s ability to absorb, digest and utilize hemoglobin. A comprehensive study on the inclusion of porcine hemoglobin in aquatic feed and other animal feed is in need in the future.

2.2.2 Emulsifier, binder, and gelling agent

Animal proteins are good emulsifiers due to their ability to bind protein, fat, and water in a stable state. However, the high costs render people to utilize blood proteins. They have comparable or even superior emulsifying abilities in some cases. Ramos-Clamont, Fernandez-

Michel, Carrillo-Vargas, et al. (2003) found that porcine serum proteins have a higher emulsifying activity index than that of whey protein and egg albumin. The emulsion stability, which is measured by % of oil released after a certain time of storage, also indicated that porcine blood emulsions are more stable compared to other frequently used proteins (casein, egg albumin powder, whey powder, soy flour, etc.).

Heat treatment leads to protein denaturation and under specific conditions, the polypeptides may form a three-dimensional network (gel). Porcine plasma proteins have abilities to form a gel with strength. It is probably because plasma proteins contain both proteinase inhibitors and transglutaminase that can enhance gelation [96]. The gel made from freeze-dried porcine plasma (70%)/serum (84%) has the stronger strength after heating compared to spray- dried egg albumin (80%) [97]. In another study on properties of frankfurters, replace polyphosphate and caseinate with porcine plasma did not affect moisture content and water holding capacity of final products significantly, but it increased the hardness and chewiness [98].

This change of texture is welcomed by the customers. The gel texture is affected by the format of

17 plasma, the pressure during production and the pH of the system. First, blood plasma can be applied in foods as liquid or as spray-dried powder. These two forms provide no difference in water-holding capacity. However, the penetration force from spray-dried porcine plasma is significantly less than that from liquid plasma. It might be caused by structure changes during the spray-drying process [99]. Second, high hydrostatic pressure (HHP) treatment is applied to plasma production due to its ability to kill microorganisms in a non-thermal way. However, the hardness of heat-induced gels significantly decreased when the pressure increased over 400 MPa

[100]. Third, pH will change gel properties and texture significantly. In acidic conditions, the addition of porcine plasma can increase water holding capacity [101]. However, the hardness, elasticity and cohesiveness increase when pH increases from acidic to alkaline. In conclusion, porcine plasma can be a gelling agent which provides similar or even better properties under specific conditions. As a by-product during pig slaughtering, it also reduces the costs of production significantly.

Porcine plasma proteins can also be applied as a binder in meat industry [102].

Transglutaminase in blood not only catalyze the covalent cross-linking of fibrin monomers but also can form bonding between the muscle protein side chains. It has the highest activity at neutral pH, temperature 35 - 45 °C in the presence of Ca2+ [103]. Using up to 15% porcine blood plasma transglutaminase combined with thrombin and fibrinogen provides stronger binding and better sensory scores in reconstructed meat. This addition did not significantly affect pH value, thiobarbituric acid (TCA) value and total plate count of final products. However, higher level addition might lead to a higher cooking loss and shrinkage [104].

Red blood cell fraction also has functional properties such as foaming ability, emulsifying property and water holding capacity [105; 106; 107]. Similarly, those properties are affected by

18 processing method and pH. Spray-dried porcine red cell protein concentrate has better emulsifying and foaming properties than the fresh red cell protein concentrate [108]. This is because spray- drying changes the conformation of porcine hemoglobin and the denaturation can improve its function [109]. pH affects protein functionality by changing the electronic state and interaction force of amino acid side chains [110]. The foaming properties of porcine red cells are greatly enhanced under acidic or neutral environment [108]. Despite the great properties, even with a few red blood cells added to food might produce an undesirable blackish brown color. Therefore, decolorize blood cells is in need. As mentioned before, the red color comes from heme unit, which can be removed by adding acetone containing 1% HCl [111]. Alternatively, it can be absorbed on certain agents such as carboxymethyl cellulose [112], sodium carboxymethyl cellulose [113] and sodium alginate [114]. Cation exchange chromatography can also be applied to decolor hemoglobin [115]. Each method can provide porcine blood cells with different properties and functionalities at different pHs [116]. However, food-grade decolorization method is still under investigation at the current stage.

2.2.3 Nutritional fortification

Iron deficiency anemia is a common illness worldwide that especially affects infants and pregnant females [117]. It is recognized as one of the most common nutritional problems [118].

Irons can be classified into heme iron which is derived from hemoglobin and non-heme iron which is also called inorganic iron. It is desirable to include hemoglobin in foods because of its higher bioavailability. It is recognized that heme iron can be absorbed more efficiently compared to inorganic iron [119] [120]. Mechanisms are still under discussion, two major opinions including the porphyrin ring or the peptides of globin chains are responsible for the increased

19 bioavailability [121; 122]. Overall, consumption of heme iron fortified foods can improve hemoglobin and serum ferritin, reduce iron deficiency and provide other beneficial biological outcomes [123; 124; 125]. Currently, hemoglobin fortified biscuits [126], cereal [127], beverages

[128] are available on the market.

Besides heme moiety, porcine blood proteins also possess incredible nutritional value due to their amino acids. When their amino acids are compared with FAO whole egg reference protein, the plasma proteins contain an excess amount of threonine (Thr), leucine (Leu), tyrosine

(Tyr) and lysine (Lys) [129]. Foods such as wheat, maize, and bread fortified with blood showed a higher protein utilization and digestibility [130; 131]. The high Lys concentration also makes blood proteins suitable for producing fortified cereals [68]. However, isoleucine (Ile) is the first limiting amino acid; this limitation can be solved with a complement of wheat, corn, soy or other vegetable proteins in the diet.

2.2.4 Natural color enhancer

Color gives consumers visual recognition of foods. It is one of the main factors that customers use to evaluate the freshness and quality. Instead of artificial color enhancer, hemoglobin obtained from porcine blood is a good source of natural red colorant. It has been used in sausage production [83]. However, it should be noted that the red color depends on the state of heme iron. HbO2 provides a bright red while MetHb shows an undesirable dark brown color. In normal red blood cells, HbO2 accounts for 97% while MetHb only accounts for approximate 3%. However, during storage, HbO2 gradually oxidizes into MetHb [132]. Different methods are developed to stabilize hemoglobin and to prevent oxidization. Adding a chelating agent (nicotinamide, nicotinic acid) and reducing agent (glucose) is proved to be effective in

20 preventing oxidation during storage of spray-dried porcine hemoglobin [133]. The chelating agents can form complex with heme moiety, therefore preventing oxidation. Another approach involves treating fresh blood with carbon monoxide (CO). Bubbling CO into the blood to reach saturation and eventually, HbCO is formed to provide a stable and attractive red color [134]. The principle is high affinity and low dissociation constant between CO and hemoglobin [135].

2.2.5 Bioactive compounds

Porcine blood contains some bioactive compounds, their activity including angiotensin I- converting (ACE)-inhibitory activity, antioxidant activity, antimicrobial capability, mineral- binding ability and opioid activity, etc. [47]. First, ACE plays a role in regulating blood pressure.

The isolation and characterization of ACE inhibitory peptides derived from porcine blood have been reported. Several peptides such as LGFPTTKTYFPHF, VVYPWT [136], PGLVVA,

GLLVLG [137], WVPSV, YTVF, VVYPW [138] are proved to possess strong inhibitory activity and antihypertensive effect. Their activity exists even after incubation with protease

[136]. There is potential to include those bioactive peptides in medicine production.

Second, antimicrobial compounds can also be purified from porcine blood. Two major groups of cationic antimicrobial components are cathelicidins and defensins in domesticated animals [47]. Researchers have successfully isolated both categories from the porcine blood. For cathelicidins, PR39 [139], prophenin [140], protegrin [141] have been isolated. Those compounds are targeted at a range of Gram-positive, Gram-negative bacteria and fungi. For defensins, -defensins are the only defensin found in pigs [142]. They display wide antimicrobial activity against intestinal bacterial (Salmonella typhimurium, Listeria monocytogenes and

21 Erysipelothrix rhusiopathiae) [143]. The antimicrobial peptides have the potential to be applied as biopreservatives in the meat industry, pharmaceutical, and nutraceutical.

Third, porcine plasma proteins have antioxidant properties. Porcine plasma hydrolysates prepared with pepsin and papain exhibited antioxidant activity [144]. This is probably because histidine (His), tyrosine (Tyr), methionine (Met) and cysteine (Cys) are rich in porcine blood, and those amino acids are reported to have antioxidant activity. Liu, Kong, Xiong, et al. (2010) also indicated that His-Asn-Gly-Asn provides the strongest antioxidant activity. However, the activity is influenced by the degree of hydrolysis [145]. Moreover, this activity makes porcine blood proteins as a potent source in Maillard reaction [146]. Benjakul, Lertittikul, and Bauer

(2005) suggested that 2% porcine plasma proteins with 2% of galactose showed the highest antioxidant activity of final products. However, along with that, the browning intensity increased while amino acid content decreased [147]. The antioxidant activity induced by plasma protein- sugar interaction is also dependent on the sugar molecular weight, plasma protein concentration and reaction conditions [146; 148].

2.3 Needs for Porcine Blood Detection

Despite the applications above of porcine blood, there are still concerns including religious issues, microbial safety, fraudulent cases, food safety and quality concerns, health concerns and environmental threatens. These highlighted the needs for porcine blood detection in foods.

22 2.3.1 Religious issues

Certain people including Muslims and Jewish condemn the consumption of blood-based foods. They follow rigorous dietary laws such as kosher and halal rules. It is reported that

Muslims and Jews accounts for 23.2% and 0.2% of world population in 2010, respectively, and these numbers are estimated to increase in the future [149]. Despite the growing needs for kosher and halal foods, issues with food authenticity can happen. First, manufacturers replace beef/mutton with pork due to the availability and low costs [27]. Also, pork by-products including gelatin, collagen, offal and fat can be adulterated in final products. Those products are inedible for religious people.

Second, food processors tend to incorporate porcine blood and its derivatives as food additives because of its biological and functional properties. Once blood is used, irrespective of the origin, is prohibited for Muslims. Meat processors are advised to ask their suppliers to provide halal certificates and clear labels [150]. However, the undemanding labeling regulations fail to eliminate the risks of consumption [4]. For example, Proferrin® is a brand producing iron supplement, which the iron is extracted from porcine red blood cells. However, it only mentions that the ingredients contain heme iron polypeptides without clarifying that they are extracted from porcine [151]. Immunolin® is another dietary product made of IgGs and other immunoglobulins [152]. Although it mentions that the components are animal derived, no further information on which animal is provided.

Currently, the established regulations on blood addition in foods are incomplete in the

US. First, USDA mentions that beef blood is acceptable in beef patties if a qualified product name such as “Beef and Blood Patties” or “Beef Patties with Blood” is provided [13], no further information on porcine blood as a component or as additives are provided. Second, it is

23 voluntary to declare the source of ingredients on the label [12]. The term “fibrin” can be directly used without clarifying “beef fibrin” or “porcine fibrin” [13]. The blank in labeling laws put potential risks of blood consumption by religious people.

Third, improper slaughtering. Kashrut is the Jewish law instructing on what foods are edible and how those foods must be prepared. One category of non-Kashrut foods is to kill the animal in a manner that prevents their blood from being completely drained from their bodies

[153]. After slaughtering, most of the blood will be rapidly drained. Bleeding efficiency depends on slaughter methods such as blood vessels that are served, size and patency of the sticking wound, orientation of the carcass-positioned [154; 155]. The remaining blood must be removed by broiling or soaking and salting [156]. If pigs are not slaughtered in such a way, around 2 – 9 mL of whole blood per kg of muscle will be remained, producing the risks of blood consumption

[157].

In general, kosher and halal foods are important parts of food systems. In the US, there are around 75,000 kosher products on the market each year [156]. Around $635 billion of halal food is produced each year worldwide, and this number is going to increase in the future [83].

However, those foods do not always meet the standards due to the three reasons mentioned before. Meat adulteration can be easily detected via PCR [158], Fourier transform infrared

(FTIR) spectroscopy [159], immunoassays [160; 161], chromatography [162], etc. However, there are limitations in porcine blood detection at current stage. To enhance food authenticity and to protect religious customers, effective porcine blood detection methods need to be developed.

24 2.3.2 Food safety concerns: Microbial safety

Blood is easy to be contaminated with pathogens via improper handling and processing

[163]. The high nutritive components, favorable pH, and temperature make blood an ideal medium for microorganism growth [14]. Bleeding and drainage system during collection are the two reasons for microorganism contamination [164]. Streptococcus suis is a gram-positive bacterium found in porcine blood [165]. The consumption of raw porcine blood poses the potential risk of its infection [5]. Since the first human case in Denmark, 1954, other severe outbreaks in Vietnam and China subsequently occurred [166; 167]. The reported infectious cases have increased since 2007. Patients experience symptoms such as meningitis, arthritis, endocarditis and endophthalmitis [168]. The case-fatality rate (CFR) of S. suis infection is around 12.8%. However, this value varies by countries [168]. Cholera is a disease caused by

Vibrio cholera that is transmitted by food or water. In a study of possible vehicles of cholera infection in Manila, dinuguan consumption was found to be associated with cholera [15].

Improper handling of blood will also threaten animal feed safety. It is possible that raw plasma collected from slaughterhouse contains a trace amount of harmful microorganism. For example, porcine epidemic diarrhea virus (PEDV) first appeared in North America in 2013; then it rapidly spread through the United States, Canada and Mexico [169]. Torque teno virus (TTV)

[170]¸ Mycoplasma suis (Eperythrozoon suis) [29], circovirus [171] and porcine reproductive and respiratory syndrome virus (PRRSV) are also detected in porcine blood. There are concerns over porcine plasma as a source of infection and lead to discontinuing usage in some countries such as UK [172]. In 2014, North American Spray Dried Blood and Plasma Producers developed manufacturing standards to ensure producing safe commercial products for animal feed [173].

25 Considering the feed is spray-dried via heat treatment and high level of antibodies in plasma, the final spray-dried porcine plasma should possess biosafety [174; 175].

2.3.3 Food safety concerns: Diseased pork

Circulation of diseased pork on the market is a food safety issue frequently happen in developing countries such as China, Vietnam, etc. Before pig slaughtering, especially during raising and transportation, they can die from diseases. The direct financial losses are borne by pig producers and slaughterhouse. It is estimated that dead and non-ambulatory pigs cost $50 to

$100 million loss in US pork industry annually [176]. In China, around 60 million heads of pigs die due to illness each year [177]. To reduce loss, dishonest manufacturers find different ways to resell the meat on the market. In 2014, the dead pigs from Gao’an were exported to Guangzhou,

Jiangsu, Shandong and other four provinces in China. Raw meat was made into sausages, jerky, meatball or even directly sold to customers. This behavior made the farmer approximate 3 million economic gain. Similarly, in 2015, around 2,000 kilograms of diseased pork were intentionally bought from slaughterhouses in Fujian, China and then sold on the market. Those foods not only violate food safety laws but also increase the risks of foodborne illness to customers. Currently, sensory evaluation is applied to differentiate the diseased pork. Pork produced from healthy pigs are usually in a bright pink color (Figure 1c), while diseased pork has a dark purplish color (Figure 1d). Recently, a peroxidase-based detection kit is also developed. Normally, healthy pork contains a high level of peroxidase, but its content will significantly decrease even disappear when pigs die from the illness. The peroxidase will decompose hydrogen peroxide to produce oxygen which will oxidize benzamine into a bluish- green compound [178]. However, those two methods have limitations in detection of processed

26 diseased pork: the additives might change the color and texture of final products, and peroxidase loses its activity after heat treatment. Therefore, it is desired to develop a simple, reliable and robust way to identify diseased pork.

Generally, the color of pork is directly related to myoglobin and hemoglobin concentration in the muscle [10; 179]. Hemoglobin is frequently used as an index of residual blood [19]. In a review paper by Paul D. Warriss (1977), several methods of residual blood measurement in muscle were illustrated and compared. Around 15-20% of blood will be found in the lean, fat and bone comprising the edible meat, while less than 10% will be found in the muscle [20]. When converted to hemoglobin concentration, 0.44 to 1.57 mg/g muscles of hemoglobin was measured according to animal species and bleeding methods [180]. However, it is a different case for diseased pork; blood will remain in all tissues because heart stops pumping the blood through the body. Therefore, it is feasible to detect diseased pork via monitoring hemoglobin concentration in pork.

2.3.4 Food quality concerns: Fraudulent cases

Economic adulteration is defined as fraudulently substitute an authentic component in foods with a cheaper and non-authentic ingredient for economic gain [181]. Blood plasma can mendaciously enhance food nutritional value because of its high protein content and superior functionalities. Only 2% of blood plasma added to a meat product can boost protein yield by 4%-

5% and substitute for almost 10% of the lean meat content [182]. Manufacturers have a financial incentive to do so. In some countries like UK, France, Germany, and Denmark, etc., blood cannot count towards the declared meat content, but it can contribute to the total nitrogen content using the traditional analytical method. In this case, an increased level of protein concentration

27 does not equal to the higher level of meat protein intake. Moreover, Fibroma®, a product that contains fibrinogen and thrombin extracted from porcine blood/bovine blood is used as binder in the meat industry. Some customers are worried that those substances negatively affect the real meat content and that they will end up consuming “glued meat.” Although European Food Safety

Authority (EFSA) assures its safety to use, the usage of Fibroma® still needs a country-by- country approval in EU [183]. In the US, the maximum amount can be used is 10% [13].

Interestingly, there are no dose requirement on ActivaTM Transglutaminase and Plasma Powder

FG dose, which are also composed of fibrinogen and be applied as binder in the food industry.

Since those products are flavorless and colorless, a sensitive and reliable detection method is in need.

2.3.5 Health and environmental concerns

People may avoid blood products intake due to allergy to blood proteins such as serum albumin and IgG [83]. Although it is rare, porcine serum albumin (PSA) represents the most significant amount of plasma protein, and it is a major allergen [184]. IgG is also proved to be a cross-reacting allergen that can cause allergy [185]. In addition, other customers such as the

Seventh Day Adventists and vegetarians refuse to intake blood products due to their personal preferences. They believe that blood contains toxic metabolites and other substances that will be harmful to their health. On the contrary, those blood-free foods are of safety and high hygiene.

Animal blood is the most problematic by-products after slaughtering [186]. The significant solids content and its high chemical oxygen demand (COD, 500,000 mg O2/L) can create a huge burden on the environment [144; 187]. In 2013, 16,000 dead pigs and their by- products were found in Jiapingtang River, China. The circovirus carried by pigs were detected in

28 tap water. Although it is non-transmissible and non-infectious to human, tap water were polluted, and a huge amount of money and effort were put to treat the pollution [6]. There are different methods such as drying and incineration used to deal with blood waste. To the best of my knowledge, there is still blank in blood regulation after animal slaughtering.

2.4 Porcine Blood Detection Methods

To solve the problems above and protect consumers, methods have been developed to detect porcine blood in foods. They can be classified into chromatography, polymerase chain reaction (PCR), spectrometry and immunoassay (Table 7).

2.4.1 Chromatography

Chromatographic techniques including gas chromatography (GC), liquid chromatography

(LC), and mass spectrometry (MS) have the ability to separate and detect porcine blood proteins in foods. Grundy, Reece, Sykes, et al. (2008) used liquid chromatography/tandem mass spectrometry (LC/MS/MS) to screen for the porcine blood-based products. The principle is based on the detection of fibrinogen, which produces fibrinopeptide A and B when it is cleaved by thrombin [189]. These two peptides are different among animal species, which allows for the detection and differentiation of porcine blood [190].

Hemoglobin can also be used as a marker. The HPLC method described by Espinoza,

Kirms, and Filipek (1996) enabled the identification of hemoglobin from different animal species. Porcine hemoglobin subunits have a unique retention time of 1β.9 min (α chain) and 5.5 min ( chain) [191]. Although it applied a diode-array detector (DAD) to have a great sensitivity

(hemoglobin content downs to 1.2 µg) and selectivity, it did not apply to the adulteration

29 samples. This method can also be applied to separate hemoglobin concentration in meat. Lyon,

Davis, and Lyon (1986) put cation exchange HPLC in isolating hemoglobin in poultry muscle.

After isolation, the isoelectrofocusing (IEF) was performed to determine contamination concentration. The results showed that 1% hemoglobin contaminated poultry meat is detectable.

Han, Mcmillin, and Godber (1994) also applied size-exclusion HPLC and hydrophobic interaction HPLC with a UV-VIS detector to determine hemoglobin concentration in beef and chicken meat.

Chromatographic methods can be applied to separate blood proteins from a mixture, and its detection sensitivity is dependent on the detector. However, they have several drawbacks.

First, the whole process is tedious, time-consuming, and costly. It requires precise equipment such as HPLC, gas chromatography, etc. Therefore, it usually requires professional technicians to conduct the experiments and analyze the data. Second, there are strict procedures related to sample preparation. Samples have to be pretreated before analysis. Therefore, not all samples are suitable for chromatographic technology. Third, the results are inconsistent by different researchers especially come to determine hemoglobin concentration in meat [20].

2.4.2 Polymerase chain reaction (PCR)

This method is based on DNA extraction and PCR amplification of a specific porcine fragment. According to Alaraidh (2008), the amplification of porcine leptin gene fragment proved successful at detecting pork contamination in food products. This detection method is merely applied in pork detection to the best of my knowledge [193] [194]. In addition, there are researches based on PCR detection of microbial infections in the porcine blood. Hoelzle, Adelt,

Hoelzle, et al. (2003) described a method of identifying a novel M. suis-specific DNA fragment

30 in porcine blood. Shibata, Okuda, Yazawa, et al. (2003) also applied PCR to detect porcine circovirus type 2 in infected pigs [28]. It has the potential to identify the disease pork. However, from the standpoint of food safety, it is more urgent of knowing the safety of the pork instead of knowing the exact pathogens. Therefore, this technique has not been applied to blood adulteration detection and diseased pork identification.

2.4.3 Spectrophotometry

This method can be used in detecting blood adulteration in meat. According to Maxstadt

J. J and Pollman M. R (1980), a standard curve is constructed using 0, 170, 340, 510, 680 mg hemoglobin/100 g ground meat. The absorbance was determined at 422 nm and 500 nm separately. Then the net absorbance (A422-A500) was plotted against mg hemoglobin/100 g sample. Unknown adulterated samples usually need to be diluted twice with cyanide reagent A and cyanide reagent B at each time, then subjected to the absorbance reading. Using absorbance and the standard curve, the added blood could be calculated by the following equation:

mg hemoglobin found per 100 g sample-a mg hemoglobin per 100 g meat Added blood%= b mg hemoglobin per g blood

Where “a mg hemoglobin/100 g meat” is the average residual hemoglobin content of unadulterated ground meat, and “b mg hemoglobin /1 g blood” is the average amount of hemoglobin recovered from 1 g added blood [195].

However, this method suffers following problems. First, the values of “a” and “b” specific for porcine hemoglobin still need evaluation. Second, based on the results from spiked samples, the standard deviations were quite large in all the samples. In addition, the reproducibility is also weak; even the same researcher would not be able to repeat their own results accurately. So far, no reasons can explain the situation [195]. Third, the range of the assay is limited. Any adulteration

31 level above 0.68% (g/g) cannot be determined via the standard curve. In addition, this assay can be applied to porcine hemoglobin or porcine whole blood detection, but it cannot be applied to porcine plasma adulteration detection. Last, the assay involves the usage of cyanide, which is poisonous.

Spectrometry can also be applied to detect hemoglobin level in meat. As described by

Gantner Jr, Sturner, and Caffrey (1962) and Lawrie (1950) that pork will be extracted with water and treated with sodium hydroxide, the absorbance will be measured at 470 nm. However, this method measured the total pigment content (i.e. hemoglobin and myoglobin) in the pork and assumed that myoglobin concentration is constant, therefore, the higher the absorbance, the more hemoglobin content. The blood level reported by this method is inconsistent with that from radioiodinated albumen [20; 198].

2.4.4 Immunoassay

Based on the interaction of a specific antibody with the antigen, immunoassay has been used in detecting trace amount of target analyte in mixtures quantitatively and qualitatively.

ELISA is one of the most powerful and useful techniques for detecting blood in foods because of its sensitivity, selectivity, reliability, robustness, easiness to operate and low costs. sELISA and icELISA are developed to detect bovine blood in feed and meat products [7; 8]. However, effective methods for porcine blood detection is inadequate until several antibodies specific to porcine blood are developed. mAb19C5-E10 and mAb16F9-C11 were chosen and developed in a sELISA to detect porcine whole blood or spray-dried porcine plasma in animal meat. This assay is porcine-blood specific and has a detection limit lower than all the reported methods (Table 3)

[9]. However, those two antibodies have target protein in porcine plasma so that porcine blood

32 cells adulteration is not detectable. That makes sense because plasma-derived products are more commonly used in comparison with proteins derived from the cellular fraction of blood [9]. In addition, the target analyte has not been characterized yet, making it limited in the commercial application. Table 8 summarizes a list of available commercial ELISA kits for detection of porcine blood in the market. There are less porcine blood cells based detection kit available at the current stage. As the functions of hemoglobin are studied further, its usage in the industry will continue to increase in the future. To the best of our knowledge, effective ELISAs to monitor porcine blood via blood cells are still lacking. In addition, currently, there is no immunoassay applied to detect blood residual in meat. Therefore, this study will characterize three antibodies (mAb, mAb19C5-E10, and mAb16F9-C11) which are specific for porcine blood. mAb is blood cells specific and has the target protein as porcine hemoglobin, while mAb19C5-E10 and mAb16F9-C11 have target proteins in porcine plasma. Their target analytes will be verified and studied. These antibodies have the potential to be used individually or in combination to detect porcine blood in foods. Briefly, mAb will be applied in an icELISA to detect blood adulteration and identify diseased pork via blood cell protein. An established sELISA using mAb19C5-E10 and mAb16F9-C11 will be improved to detect porcine blood in foods through plasma protein.

33 CHAPTER 3

INDIRECT COMPETITIVE ELISA FOR THE DETECTION OF PORCINE

BLOOD IN FOODS

3.1 Materials

Bromophenol blue (343) was purchased from Allied Chemical Corporation (Morristown,

NJ). 30% Acrylamide/bis solution (161-0158), Coomassie Brilliant Blue R-250 (BP101-50),

N,N,N’,N’-tetra-methylethylenediamine (TEMED, 161-0800), 0.5 M Tris-HCl buffer (pH 6.8,

161-0799), 1.5 M Tris-HCl buffer (pH 8.8, 161-0798), 7 cm pH 3 to 10 immobilized pH gradient

(IPG) strips (163-2007), Precision Protein StrepTactin-HRP conjugate (161-0380), Precision

Plus Protein Unstained Standards (161-0363), Precision Plus Protein WesternC Standards (161-

0376), Protein A MAPSII binding (151-6161) and elution (153-6162) buffer solids, rehydration/sample buffer (161-2306), ReadyPrep 2-D Starter Kit Equilibration Buffer I (163-

2107) and II (163-2108), supported nitrocellulose membrane (162-0115), 10× Tris/Glycine buffer (161-0734), 10× Tris/Glycine/SDS electrophoresis buffer (161-0732) were purchased from Bio-Rad Laboratories, Inc. (Hercules, CA). Luminol (21864) was purchased from Chem-

Impex International (Wood Dale, IL). Acetic acid (AX0073-9), ammonium persulfate (APS,

BP179-100), bovine serum albumin (BP1605-100), citric acid (A110-3), dimethyl sulfoxide

(DMSO, BP231-1), dithiothreitol (DTT, BP172), hydrochloric acid (A508), Pierce BCA Protein

Assay Kit (23225), sodium bicarbonate (NaHCO3, BP233-500), sodium carbonate (Na2CO3,

S495-500), sodium chloride (NaCl, S271-3), sodium citrate dehydrate (S279), sodium dodecyl sulfate (SDS, BP166-500), sodium phosphate dibasic anhydrous (Na2HPO4, S374-500), sodium phosphate monobasic anhydrous (NaH2PO4, S397-500), Tris base (BP152-500), Tween 20

(BP337-500), -mercaptoethanol (BP176-100) were purchased from Fisher Scientific (Fair

34 Lawn, NJ). Spray dried bovine hemoglobin powder (151235) was purchased from MP

Biomedicals (Solon, OH). β,β’-azino-bis (3-ethylbenzothiazoline-6-sulfonic acid) diammonium salt (ABTS, A1888), bovine hemoglobin lyophilized powder (H2500), goat anti-mouse IgG

(H+L) horseradish peroxidase conjugate (Fc specific, A2554, anti-IgG-HRP), human hemoglobin (H7379), hydrogen peroxide solution (30%, g/g, H1009), polyethylene glycol (MW

8000, P2139), porcine hemoglobin lyophilized powder (H4131), potassium phosphate dibasic anhydrous (P288), potassium phosphate monobasic anhydrous (p5379), p-Courmaric (C9008), and Ponceau S (P3504) were purchased from Sigma-Aldrich Co. (St. Louis, MO). Glycine

(0167), glycerol (BDH1172) and methanol (BDH1135) were purchased from VWR International

(West Chester, PA). All chemicals and reagents were of analytical grade. All solutions were prepared using pure deionized water from the NANOpure Diamond ultrapure water system

(Barnstead International, Inc., Dubuque, IA). All reagents formulae are summarized in Table 9, and research design and methodology is presented in Figure 5.

Beef shoulder steak, chicken fillet, pork loin and turkey breast were purchased from local markets in Tallahassee, FL. All meat samples were cut into small parts and ground twice using a meat grinder (Waring Consumer Products, East Windsor, N.J.) and stored at -80 ºC before usage.

Porcine blood and chicken blood were collected from Bradley’s Country Store (Tallahassee, FL) and Longview Farm (Havana, FL), respectively. Fifty milliliter of 3.8% (g/mL) sodium citrate solution was added to every 450-mL fresh blood as anti-coagulant. One liter of the whole blood was centrifuged at 3,000g using Avanti J30I high-performance centrifuge (Beckman Coulter,

Inc., Atlanta, GA) at 4 °C. The precipitates were collected as blood cells. The supernatant was centrifuged again at the same condition. The supernatant after second centrifugation was collected and aliquot as plasma. Part of porcine whole blood/plasma/blood cells were freeze-

35 dried at -49 °C, 144 µBar using Labconco Freeze Dry System (Laconic Corporation, Kansas

City, MO). Goat blood in Na-citrate (7202506) and sheep blood in Na-citrate (7209006) were purchased from Lampire Biological Laboratories, Inc. (Pipersville, PA). Bovine, horse, rabbit and turkey blood were purchased from Hemostat Laboratories (Dixon, CA). All materials were stored at -20 ºC before usage.

3.2 Methods

3.2.1 Production of mAbs

The animal immunization and antibody production were performed in the Hybridoma

Facility at Florida State University.

Animal immunization. All procedures using animal experiments were approved by the

Animal Care and Use Committee (ACUC) of Florida State University (ACUC approval number

1517). Three female BALB/c mice (6-8 weeks old) were immunized subcutaneously and intraperitoneally with a total of 50 μg of the immunogen mixed 1:1 (mL/mL) with Freund's complete adjuvant, followed by two or three booster injections at 4-week intervals with 25

μg/mouse of the mammalian blood immunogen mixed 1:1 (mL/mL) with Freund’s incomplete adjuvant. Test sera were collected by retro-orbital bleeding 10 days after each injection; the titers of the sera were determined by inELISA, as described in a previous study [199]. The mouse exhibiting the highest serum titer to the immunogen was received a final boost of β0 μg of the immunogen in saline before the fusion.

Antibody production. Using the hybridoma technique [200], spleen cells from the immunized mouse were fused with the myeloma cell line (NS-1, ATCC TIB-18) at a ratio of 4:1 in the presence of polyethylene glycol. The cells were diluted to an appropriate density and

36 cultured in hypoxanthine-aminopterin-thymidine (HAT) media (Sigma-Aldrich). After 10 days, the medium was screened against the immunogen using inELISA. Positive hybridomas were selected, cloned twice by limiting dilution, and expanded. For a secondary selection, the expanded positive hybridomas were screened for reactivity. Because IgM antibodies were generally more difficult to purify and store, only the IgG class of mAbs were chosen. This was achieved by using an IgG -chain specific secondary antibody as a probe in the ELISA screening procedures. mAbs were obtained in supernatants from the propagated cell cultures.

IgG purification. Antibodies were purified according to the method from Ey, Prowse, and

Jenkin (1978) with modifications. The Protein A MAPS II binding buffer salts (Bio-Rad) were added to antibody supernatant (31.4 g buffer salts in 100 g supernatant). After rotation, until fully dissolved, the supernatant was filtered through a 0.45 µm syringe filter (Millipore, Billerica,

MA) and centrifuged at 3000g at 4 ºC for 15 min using a 5810 R centrifuge (Eppendorf,

Hamburg, Germany). The supernatant was degassed for 15 min. The Econo-Pac Protein A cartridge (Bio-Rad) was connected to Bio-Rad Econo System. After equilibration with binding buffer, the supernatant was loaded at a flow rate of 0.5 mL/min. The successful binding of supernatant to the column was monitored by a Model EM-1 Econo UV Monitor (Bio-Rad) at 280 nm. The peaks started to show up once proteins reach the UV light. The effluent was collected and reloaded another three times to the column at the same speed. The column was washed with binding buffer at the same flow rate until the peak returned to the baseline. The antibody was eluted with elution buffer at 1 mL/min and collected by a fraction collector. 325 µ L of 1 M Tris-

HCl (pH 8.8) was added in each fraction in advance to neutralize the eluate. The absorbance of each fraction was read at 280 nm using a SmartSpec 3000 spectrometer (Bio-Rad). The fractions containing IgG were combined and concentrated at 4 ºC using a Centricon centrifugal filter

37 device (MWCO: 10 kDa, Millipore, Darmstadt, Germany) at 3000g using a 5810R Centrifuge

(Eppendorf). The IgG was dialyzed against 10 mM PBS at 4 °C with four changes of buffer and was filtered through a 0.22 µm syringe filter (Millipore). The IgG concentration was determined using a BioTek Take3 Micro-Volume Plate at 280 nm (BioTek, Winooski, VT). The extinction coefficient at 280 nm is 1.4 for a 1 mg/mL IgG solution [202]. The purity of IgG was checked using SDS-PAGE (15% separating gel and 4% stacking gel). All reagents in IgG purification were degassed for 15 min before usage.

3.2.2 Protein extracts preparation

Different blood and meat samples were prepared according to Rao and Hsieh (2008) with modifications. Briefly, for heated samples, around 4 g of whole blood from eight species

(bovine, chicken, goat, horse, porcine, rabbit, sheep and turkey), porcine blood cells, porcine blood plasma, and ground meat from four species (beef, chicken, pork and turkey) were heated at

100 ºC, 600 rpm for 15 min using a Thermomixer C (Eppendorf), respectively. All heated samples were cooled by immersing in ice-cold water immediately. Equal amount [g/g] of the extraction buffer was added to each sample. All samples were homogenized at 11,000 rpm for 1 min twice using a homogenizer (ULTRA-TURRAX T-25 basic homogenizer, IKA Works, Inc.,

Wilmington, NC, USA). Each homogenate was further sonicated at 50% amplitude for 10 s three times using a Q125 Sonicator (Qsonica, LLC. Newtown, CT, USA). After rotated end-over-end for at least 1 h at room temperature, the mixture was centrifuged twice at 20,000g for 15 min using a high-performance centrifuge (Beckman Coulter). The supernatant was collected and aliquoted. In addition, raw meat samples from four species were prepared using the same procedure as those heated meat samples.

38 For raw whole blood samples from eight species, porcine blood cells, and porcine blood plasma, after properly diluted in the extraction buffer, they were sonicated at 50% amplitude for

10 s three times using a Q125 Sonicator (Qsonica). The supernatant was collected and aliquoted after centrifuging at 20,000g using a 5424R centrifuge (Eppendorf). All meat and blood protein concentration was determined using BCA assay. All samples were stored at -20 °C before usage.

Laboratory spiked samples were prepared by adulterating 0.4 g freeze-dried porcine whole blood powder in 7.6 g meat (pork and chicken, wet basis). The mixture was heated, homogenized, sonicated and centrifuged as previously described.

For thermostability study, porcine hemoglobin powder (Sigma) and isolated porcine hemoglobin were dissolved in 50 mM Tris-HCl at three different pH levels (3, 7 and 10) or selected extraction buffer (12.5 mM Na2CO3 and 25 mM NaCl, pH 8.3) to make a final concentration as 1 mg/mL. At each pH, the protein extracts were divided into four portions and subjected to different heat treatment (unheated; 50 °C/600 rpm/15 min; 100 °C/600 rpm 15 min;

121 °C/2.3 Bar/15 min). Heat treatment at 50 °C and 100 °C were performed using a

Thermomixer C (Eppendorf), and autoclave at 121 °C was performed using a steam sterilizer

(Barnstead Co., Dubuque, IA). For storage stability study, unheated and heated (100 °C/15 min) commercial porcine hemoglobin powder dissolved in 50 mM Tris-HCl containing 0.05% (g/mL) sodium azide at different pHs (3, 7 and 10) were aliquoted into 200 µL per tube and stored at 4

°C up to one month. Every two days, one set was removed and stored at -80 °C.

3.2.3 Indirect non-competitive ELISA (inELISA)

inELISA was carried out (1) to screen the selectivity of mAbs antibody development; (2) to study affinity constant of the selected mAb; (3) to determine the optimal coating amount and

39 primary antibody concentration for icELISA and (4) to study the effect of temperature, pH and storage time on the immunoreactivity of target protein. Briefly, 100 μL of properly diluted animal blood/meat protein (1 ppm); or porcine hemoglobin (1 ppm, 2 ppm, 3 ppm and 4 ppm); or bovine hemoglobin (1 ppm) dissolved in carbonate-bicarbonate buffer were coated onto a 96- well polystyrene high binding microplate (Corning Inc., Corning, NY) and incubated at 37 °C for

1 h. After three washing steps with washing buffer using a Model 1575 ImmunoWash microplate washer (Bio-Rad), each well was blocked with blocking buffer and incubated at 37 °C for 1 h.

After being washed three times, 100 µL of antibody supernatant was loaded in the screening test.

For affinity study and icELISA optimization, mAb was serially diluted in antibody buffer from

12 ppm to 0.005 ppm by a two-fold dilution. 100 μL of each concentration was loaded to each well. For stability study, 0.75 ppm of mAb in antibody buffer was loaded to each well. The microplate was incubated with primary antibody at 37 ºC for 1 h and washed three times with washing buffer. Then 100 µL/well of anti-IgG-HRP conjugate (2.87 ppm) diluted in the antibody buffer was added. After incubation at 37 °C for 1 h and being washed for five times, the color was developed by adding 100 µL/well of 0.4 mM ABTS substrate solution followed by incubation at 37 °C for at least 10 min. The reaction was stopped by adding 100 µL of stopping buffer to each well. The absorbance was read using a PowerWave XS microplate reader

(BioTek) at 415 nm.

3.2.4 SDS-PAGE and Western blot.

SDS-PAGE was performed to (1) check the purity of IgG and isolated porcine hemoglobin; (2) study the effect of temperature and storage time on molecular integrity of target protein. Western blot was performed to (1) study the selectivity of the mAb; (2) investigate the

40 target analyte of mAb; (3) study the effect of pH, temperature and storage time on the antigenicity of target protein; (4) confirm the adulterated porcine hemoglobin detection.

First, raw and heated animal whole blood protein (bovine, chicken, goat, horse, porcine, sheep and turkey), porcine plasma/cells protein, animal meat protein (bovine, chicken, porcine and turkey), hemoglobin (porcine, bovine and human) and purified IgG were separated by SDS-

PAGE (4% stacking gel and 15% separating gel) using a Mini-PROTEAN Tetra Electrophoresis

Cell (Bio-Rad) [203]. The running was performed at 50 V for 30 min to reach the separating gel and then at 200 V for 1 h. One gel was stained with staining solution. After staining for 1 h, the gel was disdained by disdaining solution I for 2 h with two changes of solution. Then it was immersed in disdaining solution II overnight.

The separated protein bands from another gel were transferred to nitrocellulose membrane (Bio-Rad) using the Trans-Blot Turbo Blotting System (Bio-Rad) according to the instruction manual. The running condition was 25 V for 30 min. After transferring, the membrane was stained with Ponceau S staining solution and photographed. After removing visible bands using 10 mM phosphate buffered saline (PBS), the membrane was incubated in the blocking buffer for at least 1 h at room temperature. After washing the membrane with washing buffer for 5 min, the membrane was incubated with mAb (0.75 ppm) diluted in antibody buffer for at least 1 h at room temperature. After washing the membrane with PBST for five times, anti-

IgG-HRP (0.172 ppm) and Precision Protein StrepTactin-HRP conjugate (1:50,000, μL/μL) were dissolved in antibody buffer and added to the membrane. After six further washes with washing buffer, the chemiluminescent reagent was added. All images were captured by the ChemiDoc

XRS system (Bio-Rad) or Azure C600 Imaging System (Azure Biosystems Inc., Dublin, CA).

41 3.2.5 Aqueous two-phase system (ATPS)

Aqueous two-phase system (ATPS) was applied to isolate hemoglobin from porcine blood cells. ATPS was performed based on the method described by Selvakumar, Ling, Walker, et al. (2010) with modifications. Briefly, porcine blood cells stored at -20 °C were held at 4 °C overnight. After defrosting, two parts of deionized water were added to the blood cells (g/g). The mixture was sonicated for 10 s twice using a Q125 Sonicator (Qsonica). The mixture was then centrifuged at 3,000g for 10 min. PEG 8,000 (12.5%, g/g) was added to the supernatant, and the pH of the system was maintained at 10.0 by a small amount addition of NaOH. The separation was performed at 1000g for 15 min. After phase separation, the bottom phase was discarded.

PEG 8,000 (12.5%, g/g) was added to the top phase and the pH of the system was maintained at

7.5 by adding a mixture of 12.5% (g/g) phosphates [K2HPO4: KH2PO4 = 18:7 (g/g)]. The mixture was rotated end-over-end for 30 min and then centrifuged at 1,000g for 10 min to accelerate phase separation. The bottom phase which contains hemoglobin was dialyzed in dialysis tubing

(MWCO: 6,000-8,000, Fisher Scientific) against water for 30 h with four times of changing water. The dialyzed hemoglobin was collected and centrifuged at 3,000g for 15 min. The supernatant was collected and stored at -20 ºC before usage. All centrifuge steps were performed using a 5810R centrifuge (Eppendorf) at 4 ºC. The purity and immunoreactivity were confirmed using SDS-PAGE and Western blot, respectively. Part of the isolated porcine hemoglobin was freeze dried at -44 °C, 144 µBar using Labconco Freeze Dryer.

3.2.6 Gel filtration chromatography

Gel filtration chromatography was applied to separate porcine hemoglobin subunits based on the method described by Jandaruang, Siritapetawee, Thumanu, et al. (2012) with

42 modifications. Briefly, isolated porcine hemoglobin prepared in 3.2.5 was diluted in binding buffer (50 mM Tris and 200 mM containing 1 mM DTT, pH 8.1) before loading onto a Sephadex

G-75 resin column (1 cm × 30 cm, Bio-Rad). The column was equilibrated and eluted with binding buffer at a flow rate of 0.5 mL/min. The absorbance of each fraction was measured at

280 nm and 410 nm. SDS-PAGE and Western blot were performed to check the purity and immunoreactivity of each fraction, respectively.

3.2.7 Gradient gel electrophoresis

Gradient gel electrophoresis was performed to separate porcine hemoglobin subunits.

Porcine hemoglobin was electrophoresed on 15%-20% gradient separating gel and 4% stacking gel [67]. Briefly, 15%-20% linear gradient gel was prepared using a Gradient Gel Former (Jule

Inc., Milford, CT) according to the manufacturer’s instruction. Porcine hemoglobin was 1:1

(mL/mL) mixed with 2 × SDS-PAGE sample buffer and heated at 96 ºC, 600 rpm for 5 min using a Thermomixer C (Eppendorf). Then proteins were electrophoresed on the gel at 5 V for 1 h and then at 35 V for 10 h. After running the gel, staining and Western blot were performed as previously described.

3.2.8 Two-dimensional gel electrophoresis

Two-dimensional gel electrophoresis was carried out on heated porcine whole blood proteins (100 ºC, 15 min) with non-equilibrium polyacrylamide isoelectric gel (NEIEF) as the first dimension and SDS-PAGE as the second dimension. The IPG strip (pH 3 - 10) was rehydrated overnight in rehydration buffer containing 10 μg of heated porcine whole blood proteins. The strip was then focused at 10,000 Vh at 20 °C using a PROTEAN IEF Cell (Bio-

43 Rad). Before running SDS-PAGE, the strip was equilibrated in Equilibration buffer I and

Equilibration buffer II for 10 min each time. The 2nd dimension was performed on 15% separating gel according to the method by Laemmli using the Mini-PROTEAN Tetra

Electrophoresis Cell [203]. Ten micrograms of heated porcine whole blood proteins were 1:1 diluted using a 2 × sample buffer to be used as standards. Separated proteins on the gel were transferred electrophoretically onto a nitrocellulose membrane using the Mini Trans-Blot electrophoretic transfer cell (Bio-Rad) according to manufacturer’s instructions. Western blot was performed as previously described.

3.2.9 icELISA

The optimized icELISA was performed to (1) determine the affinity constant of the mAb;

(2) establish a standard inhibition curve for porcine hemoglobin quantification; (3) to detect porcine blood powder adulterated in pork/chicken. Briefly, 4 ppm of porcine hemoglobin in carbonate-bicarbonate buffer was coated onto each well of a 96-well polystyrene high binding microplate (Corning) and incubated at 37 °C for 1 h. The microplate was washed three times with washing buffer and then blocked by adding 200 µL/well of blocking buffer. The microplate was incubated at 37 °C for at least 1 h. The purified mAb IgG diluted in antibody buffer (1.5 ppm) was 1:1 (µL/µL) mixed with standards or adulterated samples. The standards were prepared by serially diluting porcine hemoglobin from 6000 ppm to 0.1 ppm using the antibody buffer. After antibody-antigen mixture being incubated at 37 °C for 1 h, the microplate was washed twice, and 100 µL of the equilibrated mixture was loaded to each well. The microplate was shaken for 5 min at room temperature at 500 rpm/min using a MS3 digital shaker (IKA

Works), and then incubated at 37 °C for 1 h. After incubation, the microplate was shaken for

44 another 5 min before washing three times. The incubation with secondary antibody and color development was performed the same in inELISA.

3.2.10 Data analysis

All images from SDS-PAGE and Western blot were analyzed using the Image Lab software (version 5.2, Bio-Rad) and the AzureSpot software (version 14.2, Azure Biosystems).

For inELISA, two-way analysis of variance (ANOVA) with Sidak and Tukey post-test was performed to compare the effect of pH and storage conditions on the immunoreactivity, respectively, using GraphPad Prism (version 7.02 for Windows, GraphPad Software, Inc., La

Jolla, CA): P < 0.05 was considered as statistically significant. For icELISA, one-way ANOVA with Dunnett post-test was performed to determine the limit of detection (LOD) of the assay using GraphPad Prism: P < 0.05 was considered as statistically significant.

3.3 Results and Discussion

3.3.1 Antibody screening

inELISA was performed to select the antibody that showed the greatest selectivity to porcine blood. Some supernatants were tested against animal meat/blood proteins extracts. From the results (Figure 6), mAb13F7 showed the greatest selectivity to both raw and heated porcine blood. Although positive interaction (A415 > 0.2) was also observed in other animal blood (deer, horse and bovine), this could be eliminated by purifying IgG from the supernatant and optimizing the assay parameters. It should be noted that no positive interaction was observed for both raw and heated meat samples. Therefore, mAb13F7 was chosen for further study.

45 3.3.2 Antibody purification

The IgG in the mAb13F7 supernatant was successfully purified by immunoaffinity. From

Figure 7, IgG has been purified because there are only two bands observed on the gel. Due to the disulfide bonds between the heavy chain and light chain were destroyed using -mercaptoethanol during heating (96 °C, 5 min), IgG was separated into the heavy chains (50 kDa) and light chains

(25 kDa). The IgG concentration was 0.8328 mg/mL.

3.3.3 Antibody characterization

Selectivity. The selectivity of mAb13F7 was studied using Western blot. From Figure 8, this mAb could detect the target protein from four animal blood (bovine, horse, porcine and sheep), but the band intensity of porcine blood was much higher compared to others. In addition, the positive band disappeared in bovine blood after heat treatment. The mAb has the strongest immunoreactivity toward porcine blood. It should also be noted that no cross reaction with animal meat was observed. This result paralleled with previous screening tests. The cross reaction is caused by the common epitope(s) they are sharing. The immunoaffinity will be studied and presented in the following sections.

Target analyte. To know the target protein of this mAb, the first step was to determine which part of the blood, plasma or cells, contains this protein. As shown in Figure 9, this mAb reacted strongly with porcine blood cells, a band at 14 kDa was observed in both raw and heated porcine blood cells. This pattern was also observed in the porcine whole blood protein extracts.

The major protein in blood cells is hemoglobin, which accounts up for more than 90% of total blood cells protein [3] and 50% of total blood protein concentration [47]. Porcine hemoglobin has a molecular weight of 64.5 kDa, composed of two identical α-chains and two identical -

46 chains, with molecular weights of 16,166 Da and 15,039 Da, respectively [206]. Each chain also contains a heme group (C34H32FeN4O4) which is composed of an organic component named protoporphyrin, and a central iron atom binds to the four pyrrole nitrogen atoms. The heme moiety in different animal species is the same with a molecular weight of 617 Da and can be easily released when hemoglobin is heated [46]. Because the value left when the molecular weight of subunit subtracts that of heme group was closed to 14 kDa, we hypothesized that the

14 kDa band was the subunit(s) of porcine hemoglobin.

Western blot was performed using commercial porcine hemoglobin powder (Sigma) to verify this hypothesis. As shown in Figure 9, a 14 kDa band was observed, indicating that porcine hemoglobin is immunoreactive with this mAb. Therefore, the target analyte of this mAb is porcine hemoglobin subunit(s). In addition, another faint band at 28 kDa was also observed, it is the dimer of porcine hemoglobin. Hemoglobin subunits are linked by non-covalent interactions primarily induced from positive and negative charges of side chains in the protein

[207]. As the changes in buffer pH and ion strength, the tetrameric structure of hemoglobin will dissociate into dimer or monomer. This was also observed in other studies [208; 209; 210]. It should also be noted that this mAb can cross-react with human and bovine hemoglobin, but the immunoreactivity toward bovine hemoglobin (Sigma) is much weaker compared to that of porcine and human (Sigma). The immunoaffinity difference between hemoglobin from different species will be discussed in the following section.

In addition, a 50 kDa band was observed in raw porcine plasma and porcine whole blood.

This false positive is caused by the interaction of porcine blood IgG with secondary antibody

(anti-IgG-HRP). Porcine plasma protein contains around 10% immunoglobulins while IgG takes up the most significant amount. Although the secondary antibody is mouse specific, the cross-

47 reactivity happens by the common epitopes mouse IgG and porcine IgG sharing. Due to the disulfide bonds were destroyed using -mercaptoethanol during heat treatment, the IgG is separated into two bands: the heavy chains (50 kDa) and light chains (25 kDa). The cross- reaction between porcine blood IgG with secondary antibody was proved by incubating the membrane directly with secondary antibody. A 50 kDa band can be observed in Figure 10. This false positive can only be observed in raw blood plasma or whole blood samples. This is because

IgG begins to denature at a temperature of 52 °C, and the degree of denaturation increased as a function of time and temperature [211]. After heat treatment (100 ºC, 15 min), porcine blood IgG denatured without showing positive in heated samples. In addition, IgG exists in plasma instead of blood cells.

Considering the porcine hemoglobin currently used was commercially purchased from

Sigma-Aldrich, the processing might cause some unknown structural or stability changes to hemoglobin. In addition, the purity is unknown. Therefore, porcine hemoglobin from fresh porcine blood was isolated using aqueous two-phase system (ATPS). ATPS is widely applied for protein separation and purification. The two phases are formed by the incompatibility between two solutions of polymers, or a polymer (usually polyethylene glycol, PEG) and salt at high ionic strength (e.g. phosphate, sulfate or citrate) [212]. Based on the difference in surface properties, target protein was partitioned into different phases. The partition coefficient (K) is defined as:

c K= T Equation 1 cB

Where CT and CB represent the equilibrium concentrations of partitioned protein in the top and bottom phases, respectively [213]. K is influenced by several factors. First, the electrochemical interactions, i.e. charges of proteins, play an important role [214]. Therefore, the pH of the mixture during separation is crucial. Second, hydrophobic interactions are one of the

48 most important effects involved in ATPS [212]. Other factors including PEG molecular weight and concentration, protein and salt concentration, the ionic strength of salt, or the forms of salt addition (solids, liquid, etc.) can also influence the separation. In order to isolate porcine hemoglobin to the greatest purity, an optimized two-stage extraction process was performed. The first stage was to create an alkaline pH (10), which is higher than the pI of hemoglobin, the negatively charged groups on the surface of hemoglobin tend to partition toward PEG phase. A second stage is in need because cell debris (e.g. cell membrane, DNA, and RNA) could partition strongly at the interface in the ATPS [215]. In order to remove the debris, the separation was created by adjusting the pH to neutral (7.5), and hemoglobin was separated in the bottom phase.

A brief operational flow chart was presented in Figure 12.

Porcine hemoglobin α- and -subunit shared more than 80% similarity in their amino acids sequence. In order to further investigate whether this mAb is immunoreactive to one of the subunits or both of them, porcine hemoglobin (Sigma) was applied to a 15%-20% gradient gel to separate subunits [67]. A Higher percentage of gel provides smaller pore size which would separate small molecular weight proteins with high resolution. As shown in Figure 11, the subunits were not successfully separated. This might be caused by (1) high loading mass lead to the overlap of two subunits although three different concentrations were studied; (2) running conditions and gradient formation.

Gel filtration chromatography was applied to the isolated porcine hemoglobin to separate subunits. Due to the molecular weight difference of these two subunits, their retention time in the

Sephadex column is also different, which enables the separation. α-subunit has a smaller molecular weight compared to -subunit which enables it eluted earlier. Sephadex is a gel filtration medium by crosslinking dextran with epichlorohydrin [216]. Sephadex G75 was chosen

49 because it could fraction proteins ranging from 3 kDa to 80 kDa. The absorbance of each fraction at 280 nm and 410 nm were recorded using a BioTek microplate reader (Figure 13A), those fractions (Fraction 2 to 14) contain hemoglobin were applied to SDS-PAGE. From Figure 13B, the subunits were not successfully separated. This is probably because the column we were using was 30 cm compared the 2-m (1.5 cm × 200 cm) column as Jandaruang, Siritapetawee,

Thumanu, et al. (2012) stated. The longer the column is, the greater resolution it can make. A 30 cm-column did not provide enough time space to separate two subunits.

Considering the column and chromatography limitations in the lab, two-dimensional gel electrophoresis was performed to separate subunits based on their different isoelectric points.

Porcine hemoglobin α- and -subunit have a pI of 8.76 and 7.10, respectively [50]. It was expected to see two spots on the membrane. As shown in Figure 14, several dots around 12 kDa were observed between pH 7 and 9. This phenomenon was also observed in hemoglobin from other species [209; 217; 218] and can be explained by variable phosphorylation, dimerization or impurities [217]. In general, two of the spots (at pH 8.0 and 8.6) were observed after Ponceau S staining and they reacted positively with mAb13F7 (Figure 14B). We hypothesize that this mAb could positive react with both α- and -subunit. This makes sense because there is over 80% similarity in amino acid sequence between these two subunits.

3.3.4 Immunoaffinity

From the results of Western blot (Figure 9), this mAb could cross-react with human and bovine hemoglobin. Considering that it is nearly impossible to use human hemoglobin in foods, the immunoaffinity of mAb13F7 to porcine and bovine hemoglobin was compared using indirect

ELISAs (inELISA and icELISA).

50 From the results of inELISA (Figure 15A), with the same hemoglobin coating concentration (4 ppm), when the antibody concentration increased, the absorbance of porcine hemoglobin was always higher than that of bovine hemoglobin at the same antibody concentration. When antibody concentration was 0.75 ppm, this mAb bound 8 times more porcine hemoglobin than bovine hemoglobin. This suggested that porcine hemoglobin has a higher immunoaffinity toward this mAb. From the results of icELISA, the mobilized antigen was

4 ppm of porcine hemoglobin. According to Figure 15B, at the same mobilized hemoglobin concentration (1, 10 and 25 ppm), the absorbance of bovine hemoglobin was significantly higher than that of porcine. This is because bovine hemoglobin has a lower immunoaffinity than porcine hemoglobin. Therefore, its ability of binding to this mAb is weaker, making more mAb available to bind to the immobilized porcine hemoglobin on the plate, leading to a higher absorbance. As for hemoglobin concentration decreased to 0.1 ppm, there was no significant difference in absorbance between two species. At such a low concentration, very less amount of porcine hemoglobin has already been saturated with the antibody, nearly equal amounts of left antibody were bound to the mobilized hemoglobin compared with bovine hemoglobin at 0.1 ppm. In summary, two formats of ELISA gave the same conclusion that porcine hemoglobin has a stronger immunoaffinity than bovine hemoglobin. The difference is caused by the shared epitope(s). Bovine hemoglobin shares more than 80% of similarity in amino acids compared to porcine hemoglobin (Table 5). Any portion contains the epitope(s) may lead to the immunoreactivity. However, due to the structure or the number of the epitopes are different, this mAb is more likely to bind with porcine hemoglobin. The epitope information can be further confirmed by epitope mapping.

51 Antibody affinity is defined as the strength of binding an epitope of an antigen with an antibody. All the interactions between antibody and antigen including Van der Waals, hydrogen, hydrophobic and ionic strength contribute to the affinity [219]. It has a significant influence on the performance of the developed immunoassay [220]. High-affinity antibody is proved to be superior in a variety of biological reactions [39]. In addition, the sensitivity and working range of the assay are also related to antibody affinity [40]. Several methods including ELISA, surface plasma resonance (SPR) and isothermal titration calorimetry have been applied to determine affinity constant. Overall, in an antibody-antigen interaction, the affinity constant can be described as followed:

Ag+Ab →AgAb Equation 2

[AgAb] K= Equation 3 Ab [Ag]

[ ] Based on the Law of Mass Action and using serial dilutions of both antigen and antibody, antibody affinity constants can be obtained via inELISA by the Equation 4 [36].

n-1 K = Equation 4 aff β×(n[Ab']-[Ab])

' Where n=[Ag]/[Ag ]. [Ag] is the antigen concentration (M) and [Ab] is the antibody concentration (M) at half of the maximum reading when the antigen concentration is [Ag].

Similarly, Ag' is the antigen concentration (M, less than [Ag]) and [Ab'] is the antibody concentration[ (M)] when its absorbance is half of the maximum absorbance when coating

' -1 concentration is Ag . Kaff is the affinity constant (M ).

The curve[ was] plotted using a five-parameter non-linear regression model (5-PL, Figure

16). The antibody affinity constant was calculated using Equation 4 and summarized in Table 11.

The operation of inELISA is simple, easy and rapid. However, one concern in using inELISA to measure antibody affinity constant is that it does not permit the measurement of the 52 true equilibrium dissociation constant since antigen and antibody are in separate phases. The antigen is immobilized on the solid phase while the antibody is mobilized in the solution. In addition, proteins might be induced by conformational change when they were coated on the polystyrene microplate. The changes might lead to the hindrance in binding with the antibody

[221]. Therefore, an icELISA was also performed to calculate and compare the affinity constant.

The antigen at various concentrations was first incubated with the antibody at constant concentration until the equilibrium state was reached. The dissociation constant was measured as the free antibody remaining in the solution. icELISA has the advantage to measure true antibody affinity constant since antigen and antibody are both in the immobilize state. In addition, it can be applied to measure the affinity constant of impure antibodies. The revised Scatchard equation was established according to Friguet, Chaffotte, Djavadiohaniance, et al. (1985):

Amax-A Amax Amax-A Amax-A =Kaff×(1- ) Equation 5 a0-i0× Amax Amax

Where Amax is the maximum absorbance when there is no immobilized target protein; i0 is the antibody concentration; is the mobilized antigen concentration; is the absorbance when immobilized antigen concentration�0 is at . A linear regression model� was established

0 Amax-A � Amax Amax-A using Amax-A as y-axis and (1- ) as x-axis, and the affinity constant was obtained as the a0-i0× Amax Amax slope. The affinity constant on average is 5.80 × 10-8 M-1 (Table 11). The affinity constant obtained from icELISA have high reproducibility and goodness of fit (R2 ≥ 0.9996) compared with inELISA. However, this method was also evaluated and reported to have several drawbacks. First, it is only applicable to very simple circumstances when there is only one monovalent antibody [222]. For divalent antibodies and polyclonal antibodies, other equations need to be used. Second, the derivation is complex [223].

53 In summary, two different ELISA methods were applied to determine the affinity constant of the same mAb. Both methods gave an affinity constant at nanomolar level (10 -7 to

10-9). The ELISA method is simple to operate; however, one common limitation is that what type of antibody has to be known before measuring affinity constant. Different formulations have to be applied based on antibody purity, valency, and number. Further methods including SRP and

ITC can be performed to validate the affinity constants.

3.3.5 Hemoglobin stability

Effect of pH and temperature on the immunoreactivity of porcine hemoglobin. The effects of pH and temperature on the immunoreactivity of porcine hemoglobin were studied using inELISA. According to Figure 17A, raw porcine hemoglobin (Sigma) reacted strongly with mAb at three different pH levels, but the immunoreactivity at pH 10.0 is significantly higher compared to acidic and neutral pH. It was known that alkaline pH had little effect on hemoglobin [224]. Hemoglobin has the minimal denaturation occurred at high pHs because their solubility was largely retained [225; 226].

After heat treatment at 100 °C for 15 min, the immunoreactivity of porcine hemoglobin at pH 10.0 significantly decreased compared with raw. However, this decrease did not cause a significant difference in immunoreactivity when compared to porcine hemoglobin at pH 3. A negative reaction was observed at pH 7.0 (A415 < 0.2). This was also observed by Rieder (1970) that up to 50% of porcine hemoglobin were precipitated when they were dissolved in potassium phosphate (pH 7.4). The neutral pH is close to the pI of porcine hemoglobin [50], the decrease amount of soluble protein caused the decrease in immunoreactivity. At pH levels away from pI

54 of porcine hemoglobin (pH 3.0 and pH 10.0), they still showed positive interaction (A415 > 0.2).

Overall, after heat treatment, the weakest immunoreactivity was obtained at neutral pH.

Effect of pH and temperature on the molecular integrity and antigenicity of porcine hemoglobin. SDS-PAGE and Western blot were performed to study the effect of pH and temperature on the molecular integrity and antigenicity of porcine hemoglobin, respectively.

From the results of SDS-PAGE (Figure 18A), no degradation was observed in raw porcine hemoglobin at three different pHs. After heat treatment, some bands at the lower molecular weight were observed at pH 3.0, indicating that the degradation happens after heat treatment at acidic pH. This parallels with Ivanov, Karelin, and Yatskin (2005) that degradation process can be activated under acidic conditions regardless of the acid nature. At pH 7.0, only faint bands can be observed, suggesting that porcine hemoglobin were fragmented and precipitated after heat treatment at neutral pH. At pH 10.0, no degradation was observed. In addition, a higher molecular weight band (28 kDa) was observed. Based on preliminary data, this higher molecular weight band may be the dimer of hemoglobin.

Western blot results paralleled the previous observations in SDS-PAGE (Figure 18B) and inELISA. In raw samples, hemoglobin subunits reacted with mAb at three different pHs. The band intensity at pH 10.0 was higher compared to that of other two pHs. After heat treatment, although hemoglobin subunits degraded at pH 3.0, the degraded polypeptides were still immunodetectable using the mAb. This indicates that the epitopes were kept in those degraded peptides. In addition, only very faint bands can be observed at pH 7.0, confirmed that majority of hemoglobin powder was precipitated at neutral pH. Therefore the immunoreactivity decreased significantly after heat treatment. At pH 3.0 and pH 10.0, the observed higher molecular weight bands were the polymers hemoglobin because they were immunoreactive toward mAb13F7.

55 Hemoglobin dimers and tetramers are likely formed during heat treatment [229]. The dimers were also observed in raw samples; this formation might be caused by PAGE sample preparation

(96 °C for 5 min). Moreover, the immunoreactivity for heated hemoglobin at pH 10.0 was much stronger than the one at pH 3.0. The difference in the immunoreactivity of pH 10.0 and pH 3.0 using inELISA and Western blot is caused by the principles of these two methods. inELISA is a liquid-phase analysis performed by coating certain amount of antigen on the microplate.

Hemoglobin at different pHs might suffer conformational changes which lead to the differences in binding capacity. Western blot is a solid-phase analysis with transferred protein on a membrane. Therefore, previously hidden epitopes were exposed which can lead to a higher immunoreactivity.

In order to further study the effect of pH on thermostability of porcine hemoglobin, isolated porcine hemoglobin was dissolved in Tris-HCl buffer at the same pH conditions (pH 3, pH 7 and pH 10) and heated at 50 °C, 100 °C, and 121 °C, respectively. After heating at 50 ºC, no precipitates were observed in all samples. However, after heating at 100 ºC for 15 min, precipitates were observed at pH 7. After heating at 121 ºC, precipitates were observed at pH 3 and pH 7. The precipitates were removed by centrifugation at 20,000g. At each pH level, no significant decrease in band intensity was observed after heating at 50 ºC. However, once the temperature went above 100 ºC, the band intensity of isolated porcine hemoglobin monomer decreased as a function of temperature increase (Figure 19A). This result matched previous thermostability using commercial porcine hemoglobin. Especially after 121 ºC heating for 15 min, the band intensity at three pH levels significantly decreased. The most intense band at this condition was observed at alkaline pH, and the weakest was observed at neutral pH. On the contrary, the band intensity of degraded peptides increased accordingly with the increase of

56 heating temperature. This proved that heating would degrade and fragment porcine hemoglobin.

The dimers were also observed in three sets of samples. In conclusion, at three pH levels, the degradation of porcine hemoglobin increased as a function of heating temperature.

Based on Western blot (Figure 19B), at pH 10, the isolated porcine hemoglobin showed the greatest immunoreactivity after heating at 100 ºC and 121 ºC. At pH 3, the degraded polypeptides were immunoreactive to the mAb after heating at 100 ºC for 15 min, and a very faint band can be observed when heated at 121 ºC. As for the degraded peptides showed up in

SDS-PAGE but did not show immunoreactivity on Western blot, there are several explanations.

First, those degraded peptides did not contain any epitope(s). Second, the small amount did not cause the positive interaction. Third, they have been lost during transfer due to such a small molecular weight. At pH 7, a single and faint band was observed when samples were heated at

100 ºC, and it completely disappeared after 121 ºC heat treatment. In summary, only porcine hemoglobin can be observed on the SDS-PAGE, suggesting that it has been successfully isolated using ATPS. In addition, this result paralleled with previous commercial porcine hemoglobin that after heat treatment, alkaline pH produced the highest immunoreactivity and best molecular integrity. The worst solubility and immunoreactivity after heat treatment were observed at neutral pH. Heat treatment at 100 ºC degraded porcine hemoglobin at pH 3 but facilitate dimer formation at pH 10.

In addition, isolated porcine hemoglobin and commercial porcine hemoglobin were also extracted using extraction buffer (12.5 mM NaHCO3 with 25 mM NaCl, pH 8.3). The same heat treatment conditions (50 ºC, 100 ºC, and 121 ºC) were applied. According to Figure 19, they share the similar pattern as previously mentioned. In general, as the heating temperature increased to 50 ºC, it did not decrease the immunoaffinity significantly. However, when the

57 temperature increased to 100 ºC, the monomer decreased, in the meanwhile, the degraded polypeptides showed up. After 121 ºC for 15 min, porcine hemoglobin was fragmented, and only very faint bands can be observed. Porcine hemoglobin dissolved at pH 8.3 had better thermostability compared to the neutral pH (pH 7) but was weaker than pH 10.

Regardless of the buffer type and hemoglobin origin, the results showed that alkaline pH provided the best immunoreactivity, solubility for heated sample. Acid pH worked better compared to the neutral pH. This was also observed by Jansson and Swenson (2008). Heating at

50 ºC for 15 min did not cause any significant decrease in immunoreactivity compared with raw samples at three pH level. This is because hemoglobin can remain undissociated at 50 ºC but begin dissociating above 60 ºC [231]. Once the temperature increased to 100 ºC, the hemoglobin begins to fragment into small polypeptides. In addition, the dimers can be observed because aggregation is accompanied with the unfolding process. The unfolding process happened between 42 ºC to 72 ºC while the aggregation starts around 57 ºC [230]. In summary, our data are consistent with previous results and conclude that alkaline pH can provide the best thermostability of porcine hemoglobin.

Effect of storage time on immunoreactivity of porcine hemoglobin. Because of the desirable thermostability at acidic and alkaline pH for hemoglobin, pH 3.0 and pH 10.0 were chosen to study the immunoreactivity during storage time. inELISA, SDS-PAGE and Western blot were performed to study the immunoreactivity, molecular integrity and antigenicity of commercial porcine hemoglobin during storage time, respectively. From inELISA (Figure 17B), for both raw and heated samples, no significant differences in immunoreactivity were observed during storage at two pH levels. From SDS-PAGE (Figure 18A), there are two bands observed in both raw and heated samples which should be the α- and -subunit of porcine hemoglobin. There

58 were no additional bands at smaller molecular weights were observed during storage. It indicates that hemoglobin retained its molecular integrity during storage. From Western blot (Figure 18B), the results parallel with the previous inELISA in that there was no significant change in immunoreactivity during storage time. In conclusion, the immunoreactivity of porcine hemoglobin was constant over the storage at acidic and alkaline pH.

3.3.6 Assay development

icELISA involves the competition towards a limited amount of antibody between mobilized and immobilized antigen, which is the target analyte presented in the unknown sample. According to the law of mass action, with the increase of mobilized antigen, less antibody will be able to bind to the immobilized antigen. Therefore, the measured response which is optical density (OD) will decrease. The extent of decrease is proportional to the target analyte concentration in the unknown sample. The maximum OD will be obtained when there is no immobilized antigen incubated with the antibody.

icELISA is widely applied in quantitatively determining a target in the unknown samples.

The target analyte amount in the unknown samples is determined using a standard inhibition curve established by a set of standards. In this study, first, a standard curve was established using porcine hemoglobin (Sigma); second, the standard curve was validated according to FDA

Guidance for Industry; Third, an optimized extraction buffer was developed, and this standard curve was applied to porcine blood powder adulteration samples to determine hemoglobin concentration.

Construction of standard curve. During assay development, coating amount and primary antibody concentration are two important parameters to be optimized. inELISA was performed

59 by a two-dimensional titration with coating concentration as the first dimension and IgG concentration as the second. In this study, four coating concentrations, 100 ng/µL, 200 ng/µL,

300 ng/µL and 400 ng/µL were compared. The IgG concentration of mAb13F7 was a two-fold dilution from 1,200 ng/mL to 0.52 ng/µL. From the inELISA results (Figure 16), at same antibody concentration, the absorbance increased as the increase of coating antigen concentration. The absorbance of 400-ng porcine hemoglobin was always higher than the other three coating concentrations. The absorbance was around 1.6 when it reacted with 75 ng/100 μL of mAb. In a competitive assay, the antibody amount in the mixture should be limited so that competition between mobilized and immobilized antigen can form. In addition, the limited antibody can achieve a desirable assay sensitivity [232]. Higher absorbance can be achieved by increasing antibody concentration, but this might lead to the saturation and decrease in sensitivity. In addition, substrate incubation time is another factor that will influence the sensitivity and efficiency of the assay. Longer incubation might increase the sensitivity but lead to false positive. Therefore, taking all these factors into considerations, 400 ng/100 µL of porcine hemoglobin and 75 ng/100 μL of mAb1γF7 were chosen as coating and primary antibody concentration, respectively, in the icELISA.

In the icELISA, the amount of target analyte in the unknown sample can be predicted with a standard curve generated from a set of standards. The standard curve was plotted in Figure

20 using Equation 6 [32]:

A -A % inhibition= 415 min ×100 Equation 6 Amax-Amin

Where Amax is the maximum absorbance, which is obtained when no mobilized target analyte is presented; Amin is the minimum absorbance, which is obtained when there is the maximum amount of antigen incubated with antibody.

60 Currently, a nonlinear relationship is generally chosen compared to a linear relationship in immunoassays. An accepted model including 4-parameter logistic (4-PL) model (Equation 7) and 5-parameter logistic (5-PL) model (Equation 8), which have the most accuracy and precision

[233]. Compared to 4-PL, 5-PL tend to be more asymmetric and show a higher accuracy [234].

a-d y=d+ x Equation 7 [1+( )b] c

Where y is the response, a is the maximum response obtained when there is no target analyte presented in the standards; b is the slope which defines the steepness of the curve; c is the inflection point of IC50, which can be calculated by b × (d – a)/4c; d is the minimum response when it is at infinite concentration.

a-d y=d+ x Equation 8 [1+( )b]g c

For 5-PL model, another factor, g, is added, which is an asymmetry factor. This factor allows the function to approach the asymptotes at different rates [233].

After pig slaughtering, pork skeletal muscle contains around 0.29 mg hemoglobin per 100 g of muscle [235]. Because the amount of hemoglobin needs to be added in order to function as a color enhancer, a ppm-level standard curve was established.

In the construction of ppm standard curve, porcine hemoglobin was first dissolved in extraction buffer. Then it was diluted to 400 ng/100 µL in the carbonate-bicarbonate buffer to be coated. The standards were prepared by serially diluting the porcine hemoglobin stock from

6,000 ppm to 0 using antibody buffer. The extra sum-of-squares F test was performed to compare the models. A standard curve was established using a 5-PL model and plotted in Figure

20.

Assay validation. The assay was validated on its precision, sensitivity, working range and recovery according to FDA Guidance for Industry [236]. The sensitivity of the assay is 61 characterized as the limit of detection (LOD), which is defined as the smallest amount of analyte that could be significantly differentiated from the background. In this study, LOD was obtained as the smallest quantity of porcine hemoglobin that could be differentiated from 0 ppm using one-way ANOVA, Dunnett posttest, P < 0.05. The LOD of the established standard curve is 0.5 ppm. In addition, the IC50 of this assay is 18.9 ppm. IC50 is defined as the immobilized antigen concentration needed to inhibit 50% of maximum antibody binding in competitive ELISA. In this case, it means that 18.9 ppm of porcine hemoglobin can inhibit the maximum response by

50%. The smaller this number is, the higher sensitivity of this assay.

Precision is defined as the closeness of individual measures of an analyte when the procedure is repeatedly applied to multiple aliquots of a single homogeneous volume of the biological matrix [12]. Precision can be described as the reproducibility which can be represented by the coefficient of variation (CV). The CV is calculated by dividing the standard deviation with the mean and classified into intra-assay and inter-assay CV. Intra-assay CV describes the variability between different replications within the same microplate while inter- assay CV illustrates the variation in results obtained from different days. Therefore, each experiment would provide an intra-CV. Eventually, the intra-CV should be a range; while inter-

CV is a certain value. The CVs of this assay were calculated using the Equations 9 [237] and summarized in Table 12.

CV % =(Sp/Xp)×100 Equation 9

Where Sp is the pooled standard deviation which is calculated using the Equation 10; Xp is the pooled mean which is calculated using the Equation 11.

n β n Sp= [ i=1 (dfi × si )] / i=1 dfi Equation 10

√ ∑ n ∑n Xp= [ i=1 (dfi × xi )] / i= 1 dfi Equation 11

√ ∑ ̅ ∑ 62 Where s is individual standard deviation (SD); xi is individual mean; df is degrees of freedom, which is equal to (n-1), n is the sample size of each experiment. A beneficial advantage of using pooled SD and mean to calculate CV is that it will take sample size into consideration.

According to Table 12, inter- and intra-assay CVs were all less than 20%, indicating the good reproducibility of this assay.

According to FDA, the working range of the assay is defined as the range between a lower limit of quantification (LLOQ) to the upper limit of quantification (ULOQ). The inter- and intra-CVs of LLOQ and ULOQ should be less than 20%, while at least another five points should be within the range and all should have CVs less than 15%. The working range of this standard curve is 0.5 ppm to 1000 ppm (Table 12).

The recovery of an analyte in an assay is defined as the response obtained from an amount of the analyte added to and extracted from the biological matrix, compared to the response obtained from the true concentration of the standard curve. It described the extraction efficiency of an analytical process. The closer it to 100%, there is less difference between true value and interpolated value. It is required that recovery experiments should be performed at three concentrations (low, medium, and high). In this case, 0.5 ppm, 20 ppm and 1000 ppm of porcine hemoglobin in extraction buffer were chosen to represent the LOD, approximate IC50 and

ULOQ. The recovery is calculated using Equation 12, and the results are summarized in Table

13. According to FDA, the recovery of the analyte need not be 100%.

% recovery=(calculated concentration/true concentration) × 100 Equation 12

In conclusion, the constructed standard curve is sensitive, reliable and reproducible. It can be applied to detect porcine hemoglobin in adulteration samples.

63 Selection of hemoglobin extraction methods. To effectively extract porcine hemoglobin from sample matrix, different extraction buffers and their evaluation were summarized in Table

14. The evaluation is based on its ability to extract more amount of target analyte in the samples.

Generally, Tris-HCl buffer was chosen because it is a commonly used lysis buffer. It can break the cell membrane and release components inside. This buffer has been used to extract hemoglobin from chicken muscle [192]. SDS was added because it is an ionic detergent which can function as an efficient solubilizer [238]. Its ability in solubilization of membrane proteins was well studied [239; 240; 241]. Lie-Injo, Solai, and Ganesan (1978) also illustrated that SDS could prevent globin chains in hemoglobin from precipitating. High level of SDS addition will interfere with enzymatic protein digestion [242] and lead to denaturation [240]. Therefore, 0.1% of SDS was chosen to help solubilize protein. Tween 20 is another detergent which belongs to non-ionic surfactant. It could also enhance protein extraction ability and solubilize protein but is less disruptive compared to SDS [243]. It is effective in membrane protein, periplasmic transport proteins, and inner membrane metalloprotease extraction [243]. However, one major disadvantage of Tween 20 is its non-selectivity in nature and may extract proteins along with lipids [244]. TEK buffer contains 50 mM Tris-HCl, 80 mM KCl and 1 mM EDTA. In the TEK buffer, Tris is a large organic cation, which is not only able to permeabilize outer membrane but also is able of helping the action of EDTA [245]. EDTA has a strong divalent cation-chelating function; it can increase outer membrane permeability. When dealing with animal tissues, KCl is commonly used in homogenization buffer and procedures of protein extraction [246; 247; 248].

It has the function to facilitate the release of membrane-associated molecules. However, the higher concentration could breach the electrostatic force between lipids and proteins [248]. The criterion of buffer selection is whether there will be a significant difference in absorbance

64 between different adulteration levels. Theoretically, a significant decrease in absorbance should be noticed when hemoglobin adulteration increased. Sidak post-test was performed, P < 0.05 was considered to have a significant difference. As summarized in Table 14, the extraction buffers at neutral pH were eliminated at first place because of their poor solubility and extraction ability.

Another criterion of buffer selection is that such an extraction buffer would not have any side effects on the assay itself. This was conducted by measuring the maximum absorbance by mixing 1 part of extraction buffer with 1 part of mAb in the antibody buffer. There is not any target protein in the mixture. Therefore, a maximum absorbance should be observed. According to Figure 21, the antibody-antigen interaction is significantly influenced by pH and buffer components. Firstly, at acidic pH, after mixing buffer (pH 3) with antibody buffer (pH 7.2), the pH of the whole system reaches the point that is close to the pI of bovine serum albumin [249].

During incubation at 37 ºC for 1 h, precipitation was observed. Although 0.02% fish gelatin was also applied to replace bovine serum albumin in the antibody buffer, precipitates were still observed. Secondly, alkaline pH in this study can cause a false negative. It is surprising to notice the CB buffer (pH 10) gave a negative absorbance (A415 < 0.2) even if there is not any porcine hemoglobin in the mixture. Similarly, Tris-HCl (pH 10) also gave a smaller absorbance compared with antibody buffer (pH 7.2) and Tris-HCl (pH 3). This is probably because alkaline pH inhibits the free antibody from binding to the antigen coated on the plate. Alkaline pH is effectively used to dissociate antigen from the immunoaffinity column [250]. Thirdly, SDS has a negative effect on the antibody-antigen interaction. With 0.1% of SDS added to PBS, the absorbance decreased significantly. This was confirmed by Qualtiere, Anderson, and Meyers

(1977) that when SDS concentration is greater than 0.01%, almost all immunochemical reactivity is destroyed. Although SDS has strong solubilizing properties, it inhibits the immunoreaction

65 with the antibody significantly [252]. Taking into all those factors into consideration and based on literature review, an optimized extraction buffer (12. 5 mM NaHCO3 and 25 mM NaCl, pH

8.3) was applied in extraction [253]. This buffer has been evaluated by previous requirements; it turns out it had good extraction ability and did not cause any side effects on the assay.

Detection of porcine blood in the animal meat. Compare with liquid whole blood and porcine hemoglobin powder, porcine whole blood powder is more frequently used in the meat industry. Table 10 summarized the results from the freeze-drying process. The water content in whole blood, plasma and cells is 83.7%, 85.6%, and 65.8%, respectively. These values were comparable with previous literature [254; 255]. Porcine whole blood powder was adulterated in chicken meat and pork. As shown in Figure 22, the color is significantly enhanced. The standard curve was applied to determine hemoglobin concentration. The calculated porcine hemoglobin concentration (average ± SEM) in pork is 133 ± 39 ppm (n = 14) and in chicken is 136 ± 26 ppm

(n = 14). Western blot was performed to evaluate these two adulteration levels (Figure 24). By using the volume intensity ratio, the calculated adulteration level in pork is 133 ppm and for chicken is 146 ppm. These results suggested the reliability of using the assay to detect porcine hemoglobin concentration in meat.

3.4 Conclusions

In summary, a mAb which is selective to porcine blood has been developed using hybridoma technique. IgG was successfully purified using immunoaffinity column. After characterization, the target analyte of this mAb is porcine hemoglobin. By applying different methods including gradient gel electrophoresis, two-dimensional gel electrophoresis, and column chromatography, we hypothesize that this mAb could react with both α and subunit. The

66 selectivity of this mAb was studied using Western blot. Although cross-reactivity can be observed, the immunoaffinity was compared using ELISAs (inELISA and icELISA). Porcine hemoglobin showed a higher immunoaffinity compared with others. The affinity constant of this mAb was also studied using ELISAs (inELISA and icELISA). This is the first study, to the best of our knowledge, to study the affinity constant of the same mAb using two different ELISA methods. The calculated affinity constant was at nanomolar level (10-8), indicating that this is a high-affinity antibody.

Porcine hemoglobin has been successfully isolated from porcine blood using the aqueous two-phase system. The purity and immunoreactivity were confirmed using SDS-PAGE and

Western blot, respectively. The effect of pH on thermostability and storage stability of porcine hemoglobin were studied. In the thermostability study, regardless of commercial porcine hemoglobin powder or the powder we isolated, they both showed that porcine hemoglobin has the greatest solubility, molecular integrity, and immunoreactivity at alkaline pH. After heat treatment, protein denaturation and degradation were observed as the increase of temperature. However, heat treatment at 100 ºC facilitates the dimerization of subunits. When the temperature increased to 121

ºC, the immunoreactivity significantly decreased at three pH levels. Overall, the strongest immunoreactivity was achieved at pH 10 while the weakest one was observed at pH 7. In the storage study, for both raw and heated (100 ºC, 15 min) porcine hemoglobin, there is no significant changes in immunoreactivity and molecular integrity during storage conditions using inELISA and

SDS-PAGE. In conclusion, porcine hemoglobin remains its stability at acidic and alkaline pH during storage.

An improved hemoglobin extraction method was developed to enhance the extraction efficiency of porcine hemoglobin in animal meat. It was found that SDS, acid and alkaline pH can cause negative effects on antibody-antigen interaction or assay. The finalized and optimized

67 extraction buffer was 15 mM NaHCO3 and 25 mM NaCl (pH 8.3). This buffer was applied in preparing porcine hemoglobin standards and extraction of adulteration samples.

Finally, an anti-porcine hemoglobin antibody (mAb13F7) was applied to develop an indirect competitive ELISA. The five-parameter logistic model was used to establish the standard curve. After validation based on FDA Guidance for Industry, this assay has good performance including wide working range (0.5 ppm to 1000 ppm); sensitivity (LOD: 0.5 ppm); reproducibility (Intra- and inter-assay CVs < 15%). This assay was also applied to lab- adulterated samples. Porcine hemoglobin adulteration in animal meat (pork and chicken) has been successfully determined. Therefore, this assay has the potential to detect porcine blood adulteration in animal meat. In addition, it also has the potential to identify diseased pork through quantitatively determining residual porcine hemoglobin level in pork.

3.5 Limitations

1. The false positive caused by blood IgG interaction with secondary antibody (goat anti-mouse

IgG-HRP conjugate). Currently, only heated adulteration samples were studied. This

limitation can be eliminated by labeling mAb with HRP and develop into a direct competitive

ELISA assay. In addition, a sandwich ELISA can be established using another mAb whose

target analyte is also hemoglobin.

2. A correlation between hemoglobin concentration and whole blood powder addition could be

established in the further study.

3. Its application in identifying diseased pork is not well studied due to the unavailability of

sample currently in the US. In addition, this is an indirect way to identify the diseased pork. A

68 more specific method such as PCR should be applied in the meantime to identify the pathogen.

69 CHAPTER 4

CHARACTERIZATION OF MONOCLONAL ANTIBODIES SPECIFIC

FOR PORCINE PLASMA

4.1 Materials

Bromophenol blue (343) was purchased from Allied Chemical Corporation (Morristown,

NJ). Coomassie Brilliant Blue R-250 (BP101-50), 0.5 M Tris-HCl buffer (pH 6.8, 161-0799), 1.5

M Tris-HCl buffer (pH 8.8, 161-0798), 7 cm pH 3 to 10 immobilized pH gradient (IPG) strips

(163-2007), Precision Protein StrepTactin-HRP conjugate (161-0380), Precision Plus Protein All

Blue Prestained Protein Standards (161-0373), Precision Plus Protein Unstained Standards (161-

0363), Precision Plus Protein WesternC Standards (161-0376), Protein A MAPSII binding (151-

6161) and elution buffer solids (153-6162), rehydration/sample buffer (161-2306), ReadyPrep 2-

D Starter Kit Equilibration Buffer I (163-2107) and II (163-2108), SDS-PAGE Molecular

Weight Standards, Broad Range (161-0317), supported nitrocellulose membrane (162-0115),

10× Tris/Glycine buffer (161-0734), 10× Tris/Glycine/SDS electrophoresis buffer (161-0732), were purchased from Bio-Rad Laboratories, Inc. (Hercules, CA). Luminol (21864) was purchased from Chem-Impex International (Wood Dale, IL). Acetic acid (AX0073-9), ammonium persulfate (APS, BP179-100), bovine serum albumin (BP1605-100), citric acid

(A110-3), dimethyl sulfoxide (DMSO, BP231-1), dithiothreitol (DTT, BP172), hydrochloric acid

(A508), Pierce BCA Protein Assay Kit (23225), sodium bicarbonate (NaHCO3, BP233-500), sodium carbonate (Na2CO3, S495-500), sodium chloride (NaCl, S271-3), sodium citrate dehydrate (S279), sodium dodecyl sulfate (SDS, BP166-500), sodium phosphate dibasic anhydrous (Na2HPO4, S374-500), sodium phosphate monobasic anhydrous (NaH2PO4, S397-

500), Tris base (BP152-500), Tween 20 (BP337-500) , -mercaptoethanol (BP176-100) were

70 purchased from Fisher Scientific (Fair Lawn, NJ). Chicken anti-pig plasminogen (APP) was purchased from Gallus Immunotech Inc. (Cary. NC). Rabbit anti-pig haptoglobin was purchased from Life Diagnostic, Inc. (West Chester, PA). Murine anti-human C7 antibody (A211) was purchased from Quidel Cooperation (San Diego, CA). β,β’-azino-bis (3-ethylbenzothiazoline-6- sulfonic acid) diammonium salt (ABTS, A1888), ethanolamine (411000), goat anti-mouse IgG

(H+L) horseradish peroxidase conjugate (Fc specific, A2554, anti-IgG-HRP), goat anti-rabbit

IgG (whole molecule)-HRP (A0545), hydrogen peroxide solution (30%, g/g, H1009),

N,N,N’,N’-tetra-methylethylenediamine (TEMED, T9281), porcine hemoglobin lyophilized powder (H4131), potassium phosphate dibasic anhydrous (P288), potassium phosphate monobasic anhydrous (p5379), p-Courmaric (C9008), and Ponceau S (P3504) were purchased from Sigma-Aldrich Co. (St. Louis, MO). Transferrin antibody (MAI-19013) was purchased from Thermo Scientific (Waltham, MA). Glycine (0167), glycerol (BDH1172) and methanol

(BDH1135) were purchased from VWR International (West Chester, PA). All chemicals and reagents were of analytical grade. All solutions were prepared using distilled, deionized pure water from the NANOpure Diamond ultrapure water system (Barnstead International, Inc.,

Dubuque, IA).

Porcine blood was collected from Bradley’s Country Store (Tallahassee, FL) and

Longview Farm (Havana, FL), respectively. One hundred milliliters of 3.8% (g/mL) sodium- citrate solution was added to every 900-mL fresh blood as anti-coagulant. One liter of the whole blood was centrifuged at 3,000g using Avanti J30I high-performance centrifuge (Beckman

Coulter, Inc., Atlanta, GA) at 4 °C. The supernatant was centrifuged again at the same condition.

The supernatant after second centrifugation was collected and aliquoted as plasma. The protein

71 concentration of porcine whole blood and plasma was determined using BCA assay. All protein extracts were stored at -20 ºC before usage.

4.2 Methods

4.2.1 Production of mAbs

The animal immunization and antibody production were performed in the Hybridoma

Facility at Florida State University.

Animal immunization. All procedures using animal experiments were approved by the

Animal Care and Use Committee (ACUC) of Florida State University (ACUC approval number

1517). Three female BALB/c mice (6-8 weeks old) were immunized subcutaneously and intraperitoneally with a total of 50 μg of the immunogen mixed 1:1 (mL/mL) with Freund's complete adjuvant, followed by two or three booster injections at 4-week intervals with 25

μg/mouse of the mammalian blood immunogen mixed 1:1 (mL/mL) with Freund’s incomplete adjuvant. Test sera were collected by retro-orbital bleeding 10 days after each injection; the titers of the sera were determined by inELISA, as described in a previous study [199]. The mouse exhibiting the highest serum titer to the immunogen was received a final boost of β0 μg of the immunogen in saline before the fusion.

Antibody production. Using the hybridoma technique [200], spleen cells from the immunized mouse were fused with the myeloma cell line (NS-1, ATCC TIB-18) at a ratio of 4:1 in the presence of polyethylene glycol. The cells were diluted to an appropriate density and cultured in hypoxanthine-aminopterin-thymidine (HAT) media (Sigma-Aldrich). After 10 days, the medium was screened against the immunogen using inELISA. Positive hybridomas were selected, cloned twice by limiting dilution, and expanded. For a secondary selection, the

72 expanded positive hybridomas were screened for reactivity. Because IgM antibodies were generally more difficult to purify and store, only the IgG class of mAbs were chosen. This was achieved by using an IgG -chain specific secondary antibody as a probe in the ELISA screening procedures. mAbs were obtained in supernatants from the propagated cell cultures.

IgG purification. mAb16F9 was purified according to the method from Ey, Prowse, and

Jenkin (1978) with modifications. The Protein A MAPS II binding buffer salts (Bio-Rad) were added to antibody supernatant (31.4 g buffer salts in 100 g supernatant). After rotation, until fully dissolved, the supernatant was filtered through a 0.45 µm syringe filter (Millipore, Billerica,

MA) and centrifuged at 3000g at 4 ºC for 15 min using a 5810 R centrifuge (Eppendorf,

Hamburg, Germany). The Econo-Pac Protein A cartridge (Bio-Rad) was connected to Bio-Rad

Econo System. After equilibration with binding buffer, the supernatant was loaded at a flow rate of 0.5 mL/min. The successful binding of supernatant to the column was monitored by a Model

EM-1 Econo UV Monitor (Bio-Rad) at 280 nm. The peaks started to show up once proteins reach the UV light. The effluent was collected and reloaded another three times at the same speed. The column was washed with binding buffer at the same flow rate until the peak returned to the baseline. The antibody was eluted with elution buffer at 1 mL/min and collected by a fraction collector. Three hundred microliter of 1 M Tris-HCl (pH 8.8) were added to each fraction in advance to neutralize the eluate. The absorbance of each fraction was read at 280 nm using a SmartSpec 3000 spectrometer (Bio-Rad). The fractions containing IgG were combined and be concentrated at 4 ºC using a Centricon centrifugal filter device (MWCO: 10 kDa,

Millipore, Darmstadt, Germany) at 3000g using a 5810R Centrifuge (Eppendorf). The IgG was dialyzed against PBS at 4 °C with four changes of buffer and be filtered through a 0.22 µm syringe filter (Millipore). The IgG concentration was determined using a BioTek Take3 Micro-

73 Volume Plate at 280 nm (BioTek). The purity of IgG was checked using SDS-PAGE (15% separating gel and 4% stacking gel).

4.2.2 Protein extracts preparation

Proteins were extracted from porcine whole blood and plasma as described Ofori and

Hsieh (2016) with modifications. Briefly, 10 g of porcine whole blood or plasma were weighed into 50 mL centrifuge tubes, respectively. The samples were heated at 100 ºC, 600 rpm for 15 min using a Thermomixer C (Eppendorf). After heat treatment, the solids were mashed into fine particles, and equal amounts [g/g] of PBS were added. Then samples were homogenized at

11,000 rpm for 1 min twice using a homogenizer (ULTRA-TURRAX T-25 basic homogenizer).

The mixture was sonicated at 50% amplitude for 10 s three times using a Q125 Sonicator

(Qsonica) and then rotated end-over-end for 1 h at room temperature. Then they were centrifuged at 20,000g for 15 min using Avanti J-30I high-performance centrifuge (Beckman).

This centrifugation step was repeated twice. The supernatant was collected and aliquoted. For raw porcine plasma, it directly extracted by PBS, then homogenization, sonication, and centrifugation were performed. The whole process was performed on ice, and all centrifugation was performed at 4 °C. Protein extracts concentration was determined using BCA assay. All the samples were stored at -20 °C before usage.

4.2.3 SDS-PAGE and Western blot

SDS-PAGE and Western blot was performed to confirm the target analyte of mAb19C5 and mAb16F9 using four commercial antibodies (anti-C7 IgG, Transferrin antibody, anti- plasminogen IgY and anti-haptoglobin IgG).

74 First, raw and heated porcine plasma were separated by SDS-PAGE (4% stacking gel and

15% separating gel) using a Mini-Protean Tetra Cell (Bio-Rad) according to the method of

Laemmli (1970). The running was performed at 50 V for 30 min and then at 150 V for 90 min.

One gel was stained with staining solution. After staining for 1 h, the gel was stained by disdain solution I for 2 h with two changes of solution. Then it was immersed in disdain solution II overnight.

The separated protein bands from another gel were transferred to nitrocellulose membrane (Bio-Rad) using the Trans-Blot Turbo Blotting System (Bio-Rad) according to the instruction manual. The running condition was 25 V for 30 min. After transferring, the membrane was stained with Ponceau S staining solution, photographed and cut into individual lanes. Each lane was incubated with different antibodies. Briefly, the membrane using five antibodies (mAb19C5, mAb16F9, anti-C7 IgG, transferrin antibody and anti-haptoglobin) was developed by chemiluminescence, while the anti-plasminogen IgY incubated membrane was developed using colorimetric substrate. Table 15 summarized the detailed information applied in

Western blot.

For chemiluminescent-developed membrane, after removing visible bands using PBS, the membrane was incubated in the blocking buffer for 1 h at room temperature. After washing the membrane with PBST for 5 min, the membrane was incubated with different primary antibodies.

After washing the membrane with PBST for five times, the secondary antibody was dissolved in antibody buffer and added to the membrane. After six further washes with washing buffer, the chemiluminescent reagent was added. For colorimetric development, the blocking and washing buffer changed into TBST, and the antibody buffer was 1% BSA in TBST. In addition, a different enzyme labeled secondary antibody (goat anti-mouse IgG-AP conjugate) was diluted by

75 1:3000 in antibody buffer and 1 ml was added to each lane. The colorimetric development was carried out by AP conjugate substrate kit (Bio-Rad). All images were analyzed with the

ChemiDoc XRS system (Bio-Rad) and the Image Lab version 4.1 (Bio-Rad).

Non-reducing SDS-PAGE was also performed. Protein samples were mixed with sample buffer without adding -mercaptoethanol. In addition, to study the disulfide information of target protein, another four -mercaptoethanol concentrations (2.5%, 5%, 10% and 20%) were applied in Laemmli buffer and 1:1 (mL/mL) mix with protein extracts. The running conditions and transfer procedure were the same as mentioned before.

4.2.4 Two-dimensional gel electrophoresis

Two-dimensional gel electrophoresis was carried out on heated porcine whole blood protein with isoelectric focusing (IEF) as the first dimension and SDS-PAGE as the second dimension. The IPG strip (pH 3 - 10) was rehydrated overnight in rehydration buffer containing

10 μg of porcine whole blood protein. The strip was then focused at 10,000 Vh at β0 °C using a

PROTEAN IEF Cell (Bio-Rad). Before running SDS-PAGE, the strip was equilibrated in

Equilibration buffer I and Equilibration buffer II for 10 min each. The 2nd dimension was performed on 15% separating gel according to the method by Laemmli using the Mini-Protean

Tetra Cell[203]. Ten microgram of heated porcine whole blood protein was 1:1 diluted with 2 × sample buffer to be used as standards. Separated proteins on the gel were transferred electrophoretically onto a nitrocellulose membrane using the Mini Trans-Blot electrophoretic transfer cell (Bio-Rad) according to manufacturer’s instruction. After transfer, staining, incubation and color development were performed as described earlier.

76 4.2.5 Target protein isolation

The target analyte of mAb19C5 and mAb16F9 was isolated using NHS-activated agarose dry resin according to the manufacturer’s instructions. Briefly, 400 µL of purified 19C5 IgG (1 mg/mL) was loaded with 33 mg of dry NHS-activated agarose resin in an empty spin column

(Fisher Scientific). After rotating end-over-end for 1 h at room temperature, it was continued rotating at 4 °C overnight to reach 80% coupling efficacy. The column was brought out and rotated at room temperature for 1 h the next day. Then the unbound antibody was removed using an Eppendorf Centrifuge 5424R centrifuge. 400 µL of washing buffer was added to the column, and the washing buffer was removed via centrifugation. After washing the column twice with washing buffer, 400 µL of quenching buffer was added to the column and mixed end-over-end at room temperature for 20 min. The flow-through was collected as previously described. The column was equilibrated with 800 µL binding buffer before loading the sample. Raw/heated porcine plasma protein extracts were brought from -20 ºC and centrifuged at 20,000g for 15 min at 4 °C. Four hundred microliter of the supernatant was loaded onto the column. After incubation at room temperature for at least 1 h, 3 mL of washing buffer were loaded to remove unbound protein. Then 100 µL of elution buffer was loaded, and flow-through was collected after centrifugation and neutralized with 5 µL of neutralization buffer immediately. The column was equilibrated with binding buffer for 6 times before loading new protein extracts. All the centrifugation was performed at 1,000g for 2 min at room temperature using a 5424R centrifuge

(Eppendorf). The isolation from raw and heated porcine plasma was repeated 12 times individually, and the collected fractions were concentrated using a Microcon Centrifugal Filter

Device (Merck Millipore Ltd., Cork, Ireland) at 14,000g, 4 °C. The concentrated sample protein concentration was determined using Bradford assay. The isolated proteins were applied to SDS-

77 PAGE and Western blot to confirm the purity and immunoreactivity, respectively. After confirmation, the proteins were sent to College of Medicine, Florida State University to be sequenced.

4.2.6 Target protein sequence

The following procedure was performed in Translational Laboratory, College of

Medicine, Florida State University. A solution of an isolated target analyte in 96.9% water / 3% acetonitrile / 0.1% formic acid was infused via syringe pump at 500 nL/min into a Q Exactive

HF mass spectrometer equipped with a nanospray source (Thermo Scientific). MS1 data were acquired at 60,000 resolving power over the m/z range 350-1,500. The MS1 data were deconvoluted to find the masses of the species present using BioFinder Pharma 1.0.92.0 (Thermo

Scientific). In solution, partial tryptic digestion was performed using ProteoExtract All-in-One

Trypsin Digestion Kit (650212, Calbiochem) according to manufacturer's instructions. Briefly, about 20 µg of dried protein sample were resuspended in 25 µL of extraction buffer by vigorous vortexing. The sample was centrifuged at 21000g for 10 minutes, and the supernatant was carefully transferred to a fresh tube. 25 µL of digestion buffer was added along with 1 µL of reducing agent and incubated at 37 °C for 10 minutes. Samples were cooled to room temperature and then blocked using blocking reagent for 10 minutes at room temperature. Trypsin at a final concentration of 2 ng/µL (one fourth of recommended) was added and incubated for 1 hour at 37

°C with shaking. Peptides were eluted in 40 μL 0.1% FA and run on LC/MS as described below.

An externally calibrated Thermo Q Exactive HF (high-resolution electrospray tandem mass spectrometer) was used in conjunction with Dionex UltiMate3000 RSLCnano System

(Thermo Scientific). A 2 μL (~1ug) sample was aspirated into a 50 μL loop and loaded onto the

78 trap column (Thermo µ-Precolumn 5 mm). The flow rate was set to 300 nL/min for separation on the analytical column (Acclaim pepmap RSLC 75 μm× 15 cm). Mobile phase A was composed of 99.9% water (EMD Omni Solvent), and 0.1% formic acid and mobile phase B was composed of 99.9% acetonitrile and 0.1% formic acid. A 60-minute linear gradient from 3% to 45% B was performed. The LC eluent was directly nanosprayed into a Q Exactive HF mass spectrometer.

During the chromatographic separation, the mass spectrometer was operated in a data-dependent mode and under the direct control of the Xcalibur 3.1.66 instrument software (Thermo

Scientific). The MS data were acquired using the following parameters: 20 data-dependent

MS/MS scans per full scan (350 to 1,700 m/z) at 60,000 resolution. MS2 were acquired in centroid mode at 15,000 resolution. Ions with single charge or charges more than 7 as well as unassigned charge were excluded. A 15-second dynamic exclusion window was used. All measurements were performed at room temperature. Resultant raw files were searched with

PEAKS Studio 7.5 (Bioinformatics Solutions, Inc).

4.2.7 Data analysis

All images from SDS-PAGE and Western blot were analyzed using the Image Lab software (version 5.2, Bio-Rad) and the AzureSpot software (version 14.1, Azure Biosystems).

4.3 Results and Discussion

4.3.1 Target analyte characterization

According to Table 3, this unknown target protein is in porcine plasma with a molecular weight of 90 kDa, and it is thermal stable. The first step is to use three commercial antibodies

(chicken anti-pig plasminogen, transferrin antibody, and murine anti-human C7) to see if they

79 can share the same pattern as previously observed. Porcine plasminogen has a molecular weight of around 90 kDa [256]. It is a zymogen of plasmin which can degrade blood proteins especially fibrin [257]. Porcine transferrin has an estimated molecular weight of 79 kDa, and it is reported to be stable against extreme pHs and thermal denaturation [258]. Complement component C7 is a single glycoprotein with an approximate molecular weight of 90 kDa and 100 kDa under reducing and non-reducing conditions, respectively. It plays a significant role in the formation of biologically active terminal complement complexes [259]. These three proteins have similar molecular weight as the one reported. Western blot results were represented in Figure 23. In both raw and heated samples, a 90 kDa band was observed using mAb16F9 and mAb19C5. Using anti-C7, no bands could be observed in both raw and heated samples; this is probably due to the low C7 concentration in the plasma. For anti-transferrin and anti-plasminogen, although bands were observed in raw samples, their molecular weight is smaller compared to the target protein showed up in mAb16F9 and mAb19C5. In addition, the bands were not observed in heated samples, indicating that those two proteins are not thermal stable. In conclusion, the target analyte of mAb16F9 and mAb19C5 is not transferrin, complement component 7 nor plasminogen.

In order to further investigate if the target analyte is the subunit of protein or a single protein, non-reducing SDS-PAGE was performed. In the non-reducing conditions, reducing reagents such as dithiothreitol (DTT) and beta-mercaptoethanol were not included so that disulfide information can be obtained. The results were shown in Figure 24.

Two-dimensional gel electrophoresis. In order to further know the pI of the target protein, two-dimensional gel electrophoresis was performed. According to Figure 25, a dot at the same level of target analyte was observed on the membrane. It has a pI around 4.9. Heat shock protein

80 HSP 90 has an alpha and a beta subunit, whose molecular weight is 83.2 kDa and 84.7 kDa, respectively. In addition, their pI is 4.96 and 4.93, respectively [50]. We hypothesize that the target protein could be the HSP.

Target analyte isolation. Target analyte was isolated from porcine plasma using NHS- activated dry resin agarose. This isolation was performed based on immunoaffinity. NHS- activated agarose is crosslinked that contains N-hydroxysuccinimide (NHS) functional group. It reacted with primary amines to form stable linkages so that the mAb has been covalently immobilized. After loading protein samples, target analyte was bound to the antibody through immunoaffinity. Unbound proteins were washed away by centrifugation. During elution step, the acidic pH would break the bond between antibody and the target protein. Target protein was isolated from both raw and heated porcine plasma. The purity and immunoreactivity were confirmed using SDS-PAGE and Western blot, respectively. According to Figure 26A, a single band was observed in the protein isolated from heated porcine plasma. However, a band at the lower molecular weight was observed in the samples isolated from raw porcine plasma. We hypothesize that this is the heavy chain of IgG, it might be produced during the washing or elution step. On the Western blot (Figure 26B), the bands at the same molecular weight were observed compared with plasma protein extracts. It was confirmed that the target analyte had been successfully isolated. The smearing effect was observed and might be caused by high loading concentration or protein degradation during heat treatment. Any degraded peptides contain the epitopes can show the immunoreactivity. The isolated proteins were sent to the

Translational Laboratory at College of Medicine, Florida State University.

Sequencing results suggest that pig haptoglobin has maximum confidence and amount.

Haptoglobin is a protein produced by the liver and has a 0.5 mg/mL – 0.7 mg/mL in healthy pigs

81 based on different growing stages [260]. It is an acute phase protein that exists in plasma [261] and works as an indicator of infection, inflammation or trauma. It can also form a haptoglobin- hemoglobin complex to prevent iron loss and renal damage [262]. Porcine haptoglobin is composed of α and chain connected by disulfide bonds [263]. However, the polymer structure of haptoglobin can be formed in individuals with different haptoglobin genotypes [264]. In addition, it was reported that mammalian haptoglobin has a pI of 4.0 – 4.30 [265]. Therefore, we hypothesize that porcine haptoglobin is the target analyte. To further test this hypothesis, commercial anti-haptoglobin-IgG was applied in Western blot. According to Figure 26, we observe bands in both raw and heated plasma when the membrane was incubated with anti- haptoglobin-IgG. Two major bands were observed at around 14 kDa and 44 kDa, which could be the α and subunit of porcine haptoglobin [266]. The other bands could be the polymers of subunit interaction. The major two bands at the similar position were also observed in the isolated sample, which confirms that the isolated protein contains porcine haptoglobin. However, this pattern was completely different as what were observed when the membrane was incubated with mAb16F9. We hypothesize that this 90 kDa band could be the polymer of subunits.

However, they were strongly linked together. In order to completely break the disulfide bonds between the subunits, increased -mercaptoethanol concentrations (2.5%, 5%, 10% and 20%) were applied to the sample buffer. According to Figure 27, although reducing reagents concentration has increased to 20%, the target analyte still has a molecular weight of 90 kDa.

Our previous hypothesis that the 90 kDa was a polymer of haptoglobin is wrong.

82 4.4 Conclusions

The target analyte of mAb16F9 and mAb19C5 was studied using reducing/non-reducing

SDS-PAGE, Western blot, and two-dimensional gel electrophoresis. The target analyte was successfully isolated via immunoaffinity. Its purity and immunoreactivity have been confirmed.

Isolated protein has been sequenced. In total four commercial antibodies were tested. However, we have ruled out the possibility that target protein is haptoglobin, transferrin, complement component 7 and plasminogen. In general, this target analyte has a molecular weight of 90 kDa with a pI around 4.9, and it is still under investigation.

83 APPENDIX A

TABLES

Table 1. List of worldwide incidents worldwide related to pork/porcine blood. Time Location Incidence Consequences Jan 2009 Tianjin, China Water was injected into Decrease the nutritional value pigs. of meat. Increase the possibility of microbial infection [267]. Mar 2010 - Traces of porcine blood The unaware touch with was detected in cigarette porcine blood for religious filters. people [268]. Feb 2013 Vietnam Consumption of raw More than 10 people were porcine blood to celebrate infected with the pork-based Lunar New Year Festival. pathogen. Eventually, cause 4 people death. Mar 2013 Shanghai, China Around 16,000 heads of Rotting water supply. Carry pigs were found in a river porcine circovirus [6]. near Shanghai. May 2014 Canada, Mexico, Porcine epidemic diarrhea Although the virus is not Japan virus attacked the piglets. harmful to humans or food, it caused economic loss and the usage of dried blood as feed supplement got influenced [269]. Dec. 2014 Jiangxi, China Around 70,000 heads of It is a food safety issue that dead pigs were made into can pose harmful effect on sausages and cured meat. human’s health June 2015 - Blood can be added to wine It can cause unaware as a fining agent. consumption of blood for Muslims, Jewish and vegetarians [270]. April. 2016 Texas, US Muslims try to cleanse the Local Muslims asked pig area they live. farmer to move out [271]. - US Meat industry glued scraps Decrease meat nutritional of meat together instead of value. Increase the chance of serving a genuine piece of E.coli infection [272]. meat.

84 Table 2. Characteristics of mAbs that are specific to heat-treated porcine blood. MW of target Location of target mAb(IgG1) 121 °C/15 min 100 °C/15 min Raw protein (kDa) protein B4A1 ++ +++ ++ 60 B4B1 +++ +++ +++ B4D1 +++ +++ +++ B4E1 +++ ++ ++ B7F1 ++ +++ + 85-120 B7F3 ++ + + B5A4 +++ ++ + B7E1 +++ +++ + Plasma B7A2 +++ +++ + B7C2 +++ +++ + B7A3 +++ +++ + B7G1 ++ + - B7H1 ++ + - B7B1 +++ +++ + 250 B7D1 ++ +++ + “+” denotes weak reaction; “++” denotes moderate reaction; “+++” denotes strong reaction; “-” denotes no reaction [10].

85 Table 3. Characteristics of mAb19C5 and mAb16F9. mAb19C5 mAb16F9 Selectivity Raw porcine blood + + Heated porcine blood ++ ++ Raw/heated other animala blood - -

Reactivity with PPP +++ commercial products PFP ++ APP ++ APS +

MW of target protein 90 90

Location of target protein Porcine plasma Porcine plasma

Subclass IgG1 IgG1

Stability Thermal stability Yes Yes Hydrolysis stability No No

sELISA Capture antibody Detection antibody

Detection limit Lab adulteration samples 0.3% (v/v) porcine blood in heated chicken meat Commercial products 0.01% (v/v) PPP in heated beef/pork/lamb/cod 0.03% (v/v) PPP in heated chicken 0.03% (v/v) PFP in heated beef/pork/lamb/cod/chicken 0.01% (v/v) APP in heated beef/pork/lamb/cod/chicken a denotes “cattle, donkey, horse, goat sheep, rabbit, turkey, chicken”; “-” denotes no reaction (A415 < 0.β); “+” denotes weak reaction (0.2 < A415 < 0.5); “++” denotes moderate reaction (0.5 ≤ A415 < 1); “+++” denotes strong reaction (1 ≤ A415 < 2). PPP denotes porcine plasma powder purchased from SONAC BV (Netherland); PFP denotes porcine Fibrimex powder purchased from SONAC BV (Netherland); APP denotes ApoPORK purchased from Proliant (Spain); APS denotes AProSan purchased from Proliant (Spain) [9].

86 Table 4. Major porcine blood proteins. % of whole Accession Fraction Protein Gene MW (kDa) pI porcine blood numbera Blood cells Hemoglobin 14.2 64.5 6.76 α HBA 15.0 8.76 P01965 HBB 16.1 7.1 P02067 Plasma Albumin ALB 3.8 69.7 6.08 P08835 Apolipoprotein A-I APOA1 30.3 5.48 P18648 C-II APOC3 107.0 4.76 P27917 E APOE 36.6 5.62 P18650 M APOM 21.2 6.42 Q2LE37 Fibrinogen 0.65 P14477 α FGA 1.8 4.41 P14460 FGB 2.2 4.32 Haptoglobin HP 38.5 6.51 Q8SPS7 Plasminogen PLG 90.6 7.04 P06867 Prothrombin F2 70.0 5.68 F1SIB1 Serotransferrin TF 77.0 6.93 P09571 a denotes the accession number obtained from UniProt Protein Database [50].

87 Table 5. Comparison of hemoglobin between different species. Molecular Similarity (%) Accession Species Subunit Genes weight (Da) Intra-a Inter-b numberc Porcine α HBA 15,039 38.9 P01965 (Sus scrofa) HBB 16,166 P02067 Bison α 15,139 86.6 40.5 P09423 (Bison bonasus) HBB 15,976 79.6 P09422 Bovine α HBA 15,184 85.9 39.2 P01966 (Bos Taurus) HBB 15,954 81.0 P02070 Buffalo α-1 15,128 85.2 39.2 Q9TSN7 (Bubalus bubalis) α-2 15,185 84.5 38.5 Q9TSN8 α-3 15,189 85.2 39.2 Q9TSN9 α-4 15,171 85.9 39.2 Q9XSK1 HBB 15,986 81.0 P67820 Chicken α-A HBAA 15,429 66.2 34.9 P01944 (Gallus gallus) α-D HBAD 15,695 56.7 43.6 P02001 HBB 16,466 66.7 P02112 Deer α 15,127 78.0 37.4 P01972 (Virginia white-tailed 15,824 79.6 P02074 deer) HBB Goat α-1 HBA1 15,164 87.3 41.2 P0CH25 (Capra hircus) α-2 HBA2 15,191 87.3 38.5 P0CH26 -A 16,021 79.6 P02077 -C HBBC 15,751 73.5 P02078 Duck α-A HBAA 15,420 66.9 33.6 P01987 (Cairnia moschata) α-D HBAD 15,721 57.4 42.3 P02003 HBB 16,436 66.7 P14260 Horse α HBA 15,245 85.9 41.2 P01958 (Equus caballus) HBB 16,008 81.6 P02062 α HBA1, HBA2 15,258 83.8 40.3 P69905 HBB 15,998 85.0 P68871

88 Table 5 – continued

Molecular Similarity Accession Species Subunit Genes weight (Da) Intra-a Inter-b numberc Mouse α HBA 15,085 81.7 P01942 (Mus musculus) -H0 HBB-BH0 16,384 64.0 32.9 P04443 -H1 HBB-BH1 16,494 66.0 32.9 P04444 -2 HBB-B2 15,878 78.9 38.3 P02089 Rabbit α-1/2 15,589 81.0 36.9 P01948 (Oryctolagus cuniculus) -1/2 HBB1, HBB2 16,133 83.7 P02057 Sheep α-1/2 15,164 87.3 41.2 P68240 (Ovis aries) HBB 16,073 82.3 P02075 Turkey α-A HBAA 15,441 66.2 35.8 P81023 (Meleagris gallapavo) α-D HBAD 15,665 58.2 43.9 P81024 16,307 66.0 P84479 Zebra fish α HBAA1 15,522 52.5 Q90487 (Danio rerio) -1 BA1 16,389 52.0 36.0 Q90486 -2 BA2 16,389 50.7 36.0 Q90485 a denotes the similarity of α and subunits between 14 species and pig; b denotes the similarity of α- or - subunit within the species; c denotes the accession number obtained from UniProt Protein Database.

89 Table 6. Porcine blood application. Food products Food name Country Ingredients and recipe Biroldo [273] Italy Pork blood, raisins, pine nuts and a touch of cinnamon.

Black pudding [274] United Kingdom, Ireland, Spanish Pork blood and fat, a high proportion of oatmeal. Bloodpalt [275] Sweden, Finland, and Sweden A potato dumpling with porcine blood added to the dough. Blutwurst [275] Germany Blood sausages contain pork, beef, porcine blood, spices, and herbs. Dinuguan [276] Philippines Stew of pig’s stomach, intestines, ears, heart and snout simmered in rich, spicy dark gravy of porcine blood, garlic, chilli, and vinegar Doi Huyet [275] Vietnam Porcine blood sausages made with special herbs, Vietnamese leaves, shrimp paste, etc. Moronga [275] Cuba, Puerto Rico, Central A sausage made from porcine blood, spices, herbs, onions and chiles. They will be America, Mexico boiled in pig’s large intestine for several hours as casing. Mustamakkara [277] Finland Mixing pork, porcine blood, crushed rye, and flour. Often eaten with lingonberry jam. Sangerete [275] Romania A kind of sausage which is composed of pork shoulder, porcine blood, and filler. Soondae [275] Korea A blood sausage which is stuffed with noodles, rice, kimchi or other ingredients. Ti-Hoeh-Koe [275] China Porcine blood and rice are fried or steamed. Tiet Canh [275] Vietnam Raw porcine blood is refrigerated to allow coagulation. Usually consumed with fresh herbs. Zungenwurst German Known as blood tongue. The cheese is made with porcine blood, suet, bread crumbs, oatmeal, and chunks of pickled ox’s tongue—bears some resemblance to blood sausage with large cubes of fat and tongue throughout.

Animal feed Brand name Company Components Functions Proglobulin [278] SONAC BV, Netherland Plasma Protein source in diets of piglets during the weaning period. Hemoglobin powder [91] SONAC BV, Netherland Hemoglobin It provides high protein content and good digestibility for shrimp/fish. Pet food enhancer [279] Patent Whole blood The incorporation of porcine blood in pet food enhances nutritional value and palatability. Porcine hemoglobin Shanghai Genon Hemoglobin It has the advent ages of good taste, high protein content. It is a good powder [280] Bioengineering Co., Ltd, source of protein. It can be applied as feed for poultry and aquatic China creatures.

90 Table 6 – continued

Food additives Brand name Company Components Description AproPORK PLUS 80F Proliant Meat Ingredients, Plasma It is a natural product with a high protein content and an excellent Spain emulsifying and gelling property. It can be sued in processed foods, especially sausages, meat cans, reconstructed meat. AproRED Proliant Meat Ingredients, Red blood cells It can increase the natural color of final products. It is chromatically Spain stable and ideal for heated sausages, salami and other meat products. AProSan BHJ Protein Foods, Denmark Whole blood It is a dark red, microgranulated powder. It is soluble in water which exhibits properties perfectly suited to replace the blood once the product has been hydrated. AProThem BHJ Protein Foods, Denmark Red blood cells It functions as colorant additives. It can be applied to processed meat products. Porcine plasma protein Shenzhen Taier Plasma High emulsifying and gelling capacity can stabilize water and lipids. It powder (food grade) Biotechnology Co., Ltd., improves the nutritional value. It can be applied in meat products such China as sausages and hams. Prietin Lican Functional Protein Whole blood It can be used in sausage production. Source, Chile Fibrimex [281] SONAC BV, Netherland Fibrinogen It works as a binder in meat products. It allows for better texture and flavor of the meat. It can also increase nutritional value. Harimix SONAC BV, Netherland Hemoglobin It provides the products with a stable, uniform and attractive red color.

Bioactive compounds Activity Source Functional components References ACE inhibitory Hemoglobin LGFPTTK, VVYPWT [136] Hemoglobin PGLVVA, GLLVLG [137] Hemoglobin WVPSV, YTVF [138] Plasma LVL [282] Plasma GVHGV [283] Anti-microbial activity Blood cells Protegrin [284] [285] Anti-oxidant activity Plasma Plasma albumin and globulin [286] Calcium-binding ability Plasma VSGVEDVN [287] Iron-binding ability Plasma DLGEQFKG [288] Opioid activity Hemoglobin TPTT, VVTPTTGAP, LVVTPTTGAP [289]

91

Table 7. Porcine blood detection methods. Adulteration in foods Diseased pork Chromatography Principle The different retention time of Through detection of hemoglobin via size specific protein presented in exclusion chromatography to measure the foods, such as fibrinogen, residual blood in the meat. hemoglobin, etc. [290]. Cation-exchange Size exclusion chromatography is applied to Applications chromatography is applied to determine hemoglobin concentration in meat detect adulteration via [20]. hemoglobin determination [24]. RP-HPLC is applied to detect fibrinopeptides A and B in different animal meat [190]. Limitations Difficult sample preparations, Difficult sample preparations, rigorous rigorous procedures and high procedures and inconsistent results from different costs of running the experiments. researches.

PCR Principle Through amplifying a specific Through amplifying a specific region of the region of the target DNA in target DNA. porcine meat. Mainly applied to meat It can be used in detecting diseases/virus Applications adulteration detection [291; 292]. presented in porcine blood [28]. Limitations Its application in blood It is not required to know which disease exist in adulteration is not reported. pork. What in need is a rapid detection method to tell the pork’s safety.

Spectrometry Principle Special hemoglobin absorbance Special hemoglobin absorbance at specific at specific wavelengths (422 nm wavelengths (540 nm, 580 nm) [195]. and 500 nm) [195]. It can be applied to detect blood Hemoglobin concentration can be converted to Applications adulteration using a constructed blood residual in pork, indicating the health of standard curve. pork indirectly. Limitations Specific coefficients for porcine Tedious process and involve poisonous hemoglobin is needed. Poisonous chemicals. Inconsistent results obtained from reagents are involved. different researchers.

Others Principle The color of fresh pork is pinkish while diseased pork is dark with blood observed when pressing the meat. Peroxidase-based detection kit. It is convenient and instinct for customers to Applications differentiate raw pork on the market. It belongs to rapid detection. Limitations Processed pork can interfere with the judgment.

92

Table 7 – continued

Adulteration in foods Diseased pork Immunoassays Principle The interaction of a specific antibody with the antigen, the immunoassay can detect trace amount of target analyte in mixtures quantitatively and qualitatively. Currently, no immunodetection methods are Applications ELISAs have been applied to detect available to identify diseased pork. bovine/porcine blood in foods and feed [7; 8; 9]. Limitations The effective method for blood cells detection is missing.

93

Table 8. Commercial ELISA kits for the porcine blood detection. Company Product name Target in blood Catalog number Assay type Sensitivity Range Recovery (%) CV Assay fraction (ng/mL) (ng/mL) duration Abcam Pig albumin ELISA kit Plasma ab108794 Sandwich 1.1 1.563 - 100 98 4 h Abcam Pig Haptoglobin ELISA Plasma ab205091 Sandwich 3.107 6.25 - 400 > 85 Inter-assay: <10% 40 min kit Intra-assay: <10% Abcam Pig IFN gamma ELISA Plasma ab113353 Sandwich < 0.5 0.62 - 150 97 kit Abcam Pig plasminogen ELISA Plasma ab190538 Sandwich 2.414 85 kit Abcam Lactoferrin Pig ELISA Plasma ab156516 Sandwich 0.001 0.1 - 2.5 98.5 Kit Abcam IL-1 beta Pig ELISA kit Plasma ab100754 Sandwich < 0.006 0.008 - 6 90 Abcam IL-6 (Interleukin-6) Plasma ab100755 Sandwich < 0.045 0.04 - 10 89 ELISA kit Abnova Albumin (pig) ELISA kit Plasma KA0497 Sandwich 1.1 1.563 - 100 94 - 105 Inter-assay: 10.6% Intra-assay: 3.6% Abnova CRP (Pig) ELISA kit Plasma KA1920 Sandwich 6.25 - 200 Bethyl Pig Albumin ELISA kit Plasma E101-110 Sandwich 1.23 - 900 Laboratories

Bethyl Pig IgA ELISA kit Serum E101-102 Sandwich 1.37 - 1000 Laboratories, Inc., Bethyl Pig IgG ELISA kit Plasma E101-104 Sandwich 1.37 - 1000 Laboratories, Inc., Cusabio Pig coagulation factor V Plasma CSB-EL007929PI Sandwich 27 27 - 20,000 Inter-assay: <8% 1-5 h Biotech Co., (F5) ELISA kit Intra-assay: <10% Ltd Cusabio Pig coagulation factor X Plasma CSB-EL007915PI Sandwich 78 312 - 20,000 Inter-assay: <8% 1-5 h Biotech Co., (F10) ELISA kit Intra-assay: <10% Ltd Innovative Porcine Fibrinogen Plasma IPFBGNKT Sandwich 0.6 1.56 - 800 Research ELISA kit Innovative Pig Plasminogen ELISA Plasma IRKTAH1167 Sandwich Research kit LifeSpan Pig Transferrin ELISA kit Plasma LS-F8759 Sandwich 0.313 - 20 Inter-assay: <10% BioSciences, Intra-assay: <12% Inc. LifeSpan Pig serum albumin Plasma LS-F10159 Competitive 30 30 - 700,000 Inter-assay: <8%; BioSciences, ELISA kit Intra-assay: <10% Inc. LifeSpan Pig serum albumin Plasma LS-F15679 Sandwich BioSciences ELISA kit ， Inc.

94

Table 8 – continued

Company Product name Target in blood Catalog number Assay type Sensitivity Range Recovery (%) CV Assay fraction (ng/mL) (ng/mL) duration My Glycated hemoglobin Blood cells MBS280244 Sandwich 310 780 - 50,000 Inter-assay: <8%; 3 h BioSource A1c (GHbA1c), ELISA Intra-assay: <12% kit

My Pig albumin ELISA kit Plasma MBS705964 Competitive 30 30 - 7,000,000 Inter-assay: <8%; 3 h BioSource Intra-assay: <10% R&D Porcine IL-6 Quantikine Plasma P6000B Sandwich 0.0043 0.0188 - 1.2 85 -109 Inter-assay: <5.1%; 4.5 h Systems ELISA kit Intra-assay: <7.5% Thermal IFN gamma ELISA kit Serum KSC4021 Sandwich < 0.002 0.0078 - 0.5 84 3 h Scientific Thermal IL-8 ELISA kit (Porcine) Serum KSC0082 Sandwich < 0.01 0.00312 - 2 93 Inter-assay: <8%; 4 h Scientific Intra-assay: <10% Tridelta Porcine CRP ELISA kit Plasma TA-901 Development Ltd

95

Table 9. Reagents formula. Reagent Formula IgG purification Binding buffer, pH 9.0 ± 0.2 31.4% [g/mL] MAPS II binding buffer solids Elution buffer, pH 3.0 ± 0.2 2.2% [g/mL] MAPS II elution buffer solids 1 M Tris-HCl, pH 8.8 1 M Tris base, pH adjusted to 8.8 using 12.1 M HCl

ELISA 50 mM Carbonate-bicarbonate buffer, pH 9.6 15 mM Na2CO3 and 35 mM NaHCO3 10 mM Phosphate buffered saline, pH 7.2 (PBS) 76.8 mM Na2HPO4, 26.7 mM NAH2PO4 and 1.49 M NaCl PBST, pH 7.2 0.05% [mL/mL] Tween 20 in PBS Blocking buffer, pH 7.2 1% BSA [g/mL] in PBS Antibody buffer, pH 7.2 1% BSA [g/mL] in PBST 50 mM Phosphate-citrate buffer, pH 5.0 (PCB) 50 mM Na2HPO4 and 25 mM citric acid 0.4 mM ABTS substrate 0.4 mM ABTS and 1.5% [mL/mL] H2O2 in PCB Stopping buffer 1% [g/mL] SDS

SDS-PAGE and Western blot 2 × Sample buffer 62.5 mM of 0.5 M Tris-HCl (pH 6.8), 25% (mL/mL) glycerol, 2% [g/mL] SDS, 0.01% [g/mL] Bromophenol Blue and 1/19 [mL/mL] -mercaptoethanol Running buffer 25 mM Tris base, 192 mM glycine and 0.1% [w/v] SDS Staining solution 0.025% [g/mL] Coomassie Brilliant Blue R250, 40% [mL/mL] Methanol and 7% [mL/mL] acetic acid Disdaining solution 1 40% [mL/mL] Methanol and 7% [mL/mL] acetic acid Disdaining solution 2 5% [mL/mL] Methanol and 7% [mL/mL] acetic acid Transfer buffer 25 mM Tris and 192 mM glycine Ponceau S staining solution 0.1% [g/mL] Ponceau S and 5% [mg/mL] acetic acid Washing buffer PBST Blocking buffer 1% BSA [g/mL] in PBST Antibody buffer 1% BSA [g/mL] in PBST Chemiluminescent color development solution A 25 mM Luminol, 9 mM p-Coumaric acid and 100 mM of 1 M Tris-HCl (pH 8.8) Chemiluminescent color development solution B 0.06% [mL/mL] 30% H2O2 and 100 mM of 1 M Tris-HCl (pH 8.8)

96

Table 9 – continued

Reagent Formula Two-dimensional gel electrophoresis Rehydration buffer 8 M Urea, 2% [g/mL] CHAPS, 50 mM DTT, 0.2% [mL/mL] Bio-Lyte 3/10 Ampholyte, 0.001% [mL/mL] of 1% Bromophenol Blue Equilibration buffer I 6 M Urea, 2% [g/mL] SDS, 0.375 mM of 1.5 M Tris-HCl (pH 8.8), 20% [mL/mL] glycerol and 2% [g/mL] DTT Equilibration buffer II 6 M Urea, 2% [g/mL] SDS, 0.375 mM of 1.5 M Tris-HCl (pH 8.8), 20% [mL/mL] glycerol and 2.5% [g/mL] iodoacetamide

Target protein isolation Washing buffer PBS Binding buffer PBS Quenching buffer 1 M ethanolamine in PBS Elution buffer 0.1 M Glycine-HCl, pH 2.5-3.0 Neutralization buffer 1 M Tris-HCl, pH 8.8

97

Table 10. The water content of porcine blood. Portion Before freeze dryinga (g) Weight loss (g) Water contentb (%) Whole blood 1 11.13 9.32 83.7 Whole blood 2 10.13 8.48 83.7 Average 83.7 Plasma 1 5.03 4.32 85.9 Plasma 2 4.13 3.52 85.2 Average 85.6 Cells 1 7.06 4.65 65.9 Cells 2 5.59 3.67 65.7 Average 65.8 a denotes freeze-drying was performed at –44 ºC, 144 µBar. b water content is calculated as the ratio of weight loss to the weight before freeze drying.

Table 11. The affinity constant (M-1) of mAb13F7/C8. Indirect non-competitive ELISA to calculate Kaff using Equation 4 -8 -1 Ab (nM) Ab’ (nM) n Kaff (× 10 M ) 3.23 13.29 2 4.67 1.32 13.29 3 3.86 0.87 13.29 4 3.48 0.87 3.23 2 1.12 Average: 3.28

-8 -1 Indirect competitive ELISA to calculate Kaff using Equation 5 (× 10 M ) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 5.83 5.82 5.81 5.81 5.81 5.81 5.82 5.81 5.80 5.78 5.77 5.77 5.82 5.78 Average: 5.80

98

Table 12. Coefficient of variations (CVs) of standard curve established by icELISA using mAb13F7/C8. Inter-CV (%) Intra-CV ppm 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Range (%) 0.5 6.95 4.18 0.85 0.79 1.47 7.04 3.55 1.28 0.41 1.57 0 4.85 10.19 4.03 0.52 0 - 10.19 5.35 1 0.52 1.21 1.56 0.46 2.32 5.91 8.42 0.36 1.13 1.71 0.63 6.08 0.85 2.77 3.17 0.46 - 8.42 7.53 10 0.52 0.77 3.16 0.16 0.14 0.26 2.52 0.95 3.56 11.07 2.03 4.79 2.38 0.92 0.38 0.14 - 11.07 14.32 25 0.96 2.98 3.28 2.41 0.52 1.79 3.92 2.21 2.42 3.18 1.06 7.18 0.75 0.82 0.54 0.52 - 3.92 10.52 50 2.29 0.54 5.78 1.26 5.14 3.92 1.84 9.43 4.04 3.73 1.11 2.53 1.53 2.71 1.98 0.54 - 9.43 13.01 100 3.96 1.66 12.46 2.06 3.95 0.76 0 3.58 0.18 7.22 1.87 0.71 7.50 1.09 0.53 0-12.46 13.32 250 1.47 9.01 3.90 3.23 4.67 1.27 1.83 1.88 0.58 0 4.49 4.89 6.52 1.99 1.01 0 - 9.01 14.98 500 0.14 3.50 3.35 0.21 0.55 3.70 3.05 1.22 7.19 1.04 4.64 2.34 6.22 1.11 1.05 0.14 - 7.19 12.58 1000 4.46 6.13 2.09 1.37 0.85 7.25 8.38 1.37 5.89 0.71 0.65 0.14 0.68 5.57 1.07 0.14 - 8.38 17.14

Table 13. Recovery of porcine hemoglobin extracted by extraction buffer. True value (ppm) Calculating value (ppm) Recovery (%) 0.5 0.49 ± 0.03 98.5 ± 6.8 20 19.0 ± 2.2 94.8 ± 10.8 1000 1106 ± 317 124.8 ± 30.8 Results were expressed as the mean ± SEM.

99

Table 14. Summary of extraction buffers.

Component 1 mM 0.05% Heat Solubilityb Buffer type pH 0.1% SDS Abbreviation Group Ac Group Bd EDTA and Tween 20 treatmenta (1 mg/mL) (g/mL) 80 mM KCL (mL/mL) 50 mM Tris- 3 - - - Tris-HCL H X HCL 7 - - - H √ X 8 + - - TEK R Y H X 10 - - - Tris-HCL R Y H X - + - Tris-SDS H X - - + Tris-TWEEN H X + + - TEK-SDS R Y H X

10 mM PBS 7.2 - + - PBS-SDS H √ X - - + PBST H √ X 10 - - + H X

50 mM 10 - - - CB R Y Carbonate- H X bicarbonate 10 - + - CB-SDS R Y buffer H Y a: R: raw; H: heated (100 ºC for 15 min); b: the solubility of 1 mg/mL of porcine hemoglobin in extraction buffer. √ denotes that porcine hemoglobin solution is crowded; c: whether there is a significant difference in absorbance between 50,000 ppb and 500 ppb of adulteration extracted by the buffer, Sidak post-test was performed, P < 0.05 was considered significant different; d: whether there is a significant difference in absorbance between 100 ppm of adulteration and pork meat base extracted by the buffer, Sidak post-test was performed, P < 0.05 was considered significant different; Y denotes as yes, there is a significant difference; X: denotes as no, there is no significant difference.

100

Table 15. Summary of commercial antibodies used in Western blot. Chemiluminescent development Colorimetric development 1st antibody mAb19C5 mAb16F9 anti-C7 Transferrin Anti-haptoglobin Anti-plasminogen IgG antibody 1st antibody 1,000 1,000 concentration (ng/mL)

2nd antibody Goat anti-mouse IgG-HRP conjugate Goat anti-rabbit Anti-chicken IgY-AP conjugate IgG-HRP conjugate 2nd antibody 170 333,333 concentration (ng/mL)

Blocking buffer 1% BSA in PBST 1% BSA in TBST

Washing buffer PBST TBST

Color development Chemiluminescent substrate A and B AP conjugate substrate kit (Bio- Rad)

101

APPENDIX B

FIGURES

Figure 1. Illustration of pig farm and pork. (a) pig farm [293]; (b) ill pigs [294]; (c) safe pork [295]; (d) diseased pork [18]

102

Figure 2. The process of pig slaughtering.

103

Figure 3. Three-dimensional structure of porcine hemoglobin.

104

Figure 4. Two-dimensional structure of heme moiety.

105

Figure 5. Illustration of research objective and methodology. 106

Figure 6. Screening test for mAb13F7/C8 using indirect non-competitive ELISA. The results were denoted as A415 ± SEM (n = 2). A415 > 0.2 was considered as positive. (A) animal meat; (B) animal whole blood (B: bovine; Bi: bison; Bu: buffalo; C: chicken; D: deer; Du: duck; E: elk; G: goose; H: horse; L: lamb; P: porcine; R: rabbit; S: sheep; T: turkey). 107

Figure 7. SDS-PAGE of purified mAb13F7/C8 IgG. The loading mass was 0.5 μg. The Precision Plus Protein Kaleidoscope standards were used to indicate the molecular weight.

108

Figure 8. SDS-PAGE and Western blot to study the selectivity of the mAb. (A) SDS-PAGE of raw animal blood and meat protein extracts; (B) SDS-PAGE of proteins from animal blood and meat after heat treatment (100 ºC, 15 min); (C) Western blot of (A); (D) Western blot of (B). T: turkey; S: sheep, R: rabbit; P: porcine, H: horse; G: goat; C: chicken; B: bovine. The loading mass of each lane was 1.25 µg. mAb13F7 concentration was 0.75 ppm. The Precision Plus Protein WesternC standards were loaded on the left indicating molecular weights. 109

Figure 9. SDS-PAGE and Western blot to study the target analyte of mAb13F7. (A) SDS-PAGE of proteins from porcine whole blood (W); porcine plasma (P) and blood cells (C); (B) Western blot (PHb: porcine hemoglobin; BHb: bovine hemoglobin; HHb: Human hemoglobin; r: raw; h: heated at 100 ºC for 15 min). The loading mass for blood proteins was 1.25 µg and for pure hemoglobin from different species was 0.5 µg. mAb13F7 concentration was 0.75 ppm. The Precision Plus Protein WesternC standards were loaded on the left indicating molecular weights. 110

Figure 10. Western blot to study the cross-reactivity of porcine blood IgG with anti-IgG-HRP. Proteins were from porcine whole blood (W) and porcine plasma (P); the subscript r denotes as raw while h denotes as heated samples (100 ºC for 15 min). The loading mass for each lane was 1.25 µg. mAb13F7 concentration was 0.75 ppm. The Precision Plus Protein WesternC standards were loaded on the left indicating molecular weights.

111

Figure 11. Gradient gel electrophoresis and Western blot. The loading mass for porcine hemoglobin was 5 µg, 2.5 µg, and 1.25 µg, respectively. mAb13F7 concentration was 0.75 ppm.

112

Figure 12. Isolation of porcine hemoglobin using aqueous two-phase system (APTS).

113

Figure 13. Column chromatography purification and SDS-PAGE. (A) A280 and A410 of different fractions during column chromatography. (B) SDS-PAGE of collected fractions (F2 – F14). Each fraction was 50 times diluted with binding buffer (50 mM Tris-HCl, 200 mM NaCl and 1 mM DTT).

114

Figure 14. Two-dimensional gel electrophoresis of heated whole porcine blood. (A) Ponceau S staining; (B) Western blot. Heated porcine whole blood (100 ºC, 15 min) was used as a standard, the loading mass for the first dimension was 10 μg. Five microgram of heated porcine whole blood protein extracts was used as standards. The mAb concentration was 0.75 ppm.

115

Figure 15. Comparison of immunoaffinity of mAb13F7 to porcine and bovine hemoglobin using inELISA and icELISA. (A) The five-parameter logistic curves using inELISA. The coating concentration of bovine and porcine hemoglobin was 4 ppm. mAb13F7 concentration was from 12 ppm to 0.005 ppm by a two-fold dilution. Results were expressed as A415 ± SEM, n = 2. (B) icELISA. 4 ppm of porcine hemoglobin was coated. Results were expressed as A415 ± SEM, n = 2. Different letters denote the significant difference between same mobilized hemoglobin concentration using Sidak post- test, P < 0.05 was considered as significantly different.

116

Figure 16. Two-dimensional titration using indirect non-competitive ELISA. The five-parameter logistic model was used to establish the curves. The coating concentration of porcine hemoglobin was 1 ppm, 2 ppm, 3 ppm and 4 ppm, respectively. mAb13F7 concentration was from 12 ppm to 0.005 ppm by a two-fold dilution. Results were expressed as A415 ± SEM, n = 2. The purple circled point was chosen in the icELISA which porcine hemoglobin concentration was 4 ppm, and antibody concentration was 0.75 ppm.

117

Figure 17. Effect of pH on thermostabililty and storage stability of porcine hemoglobin. (A) Effect of pH and temperature on the immunoreactivity of porcine hemoglobin using inELISA. The mAb concentration was 0.75 ppm. Results were denoted as A415 ± SEM, n = 4. Different letters indicate the significant difference using Sidak post-test, P < 0.05 was considered as significantly different. (B) Effect of storage time on the immunoreactivity of porcine hemoglobin using inELISA. The mAb concentration was 0.75 ppm. Results are denoted as % of absorbance compared to Day 0 ± SEM, for raw samples, n = 4 and for heated samples, n=6. Sidak post-test was performed, P < 0.05 was considered as significantly different. There is no significant difference during storage time for all four groups. 118

Figure 18. The effect of pH on molecular integrity and storage stability of porcine hemoglobin. (A) SDS-PAGE; (B) Western blot. The loading mass for each lane was 1 µg. mAb13F7 concentration was 0.75 ppm.

119

Figure 19. SDS-PAGE and Western blot to study the effect of pH on thermostability and storage stability of isolated porcine hemoglobin. (A) SDS-PAGE; (B) Western blot. The loading mass for each lane was 1 µg. Primary antibody concentration was 0.75 ppm. All heat a treatment had a duration of 15 min. : extraction buffer was 12.5 mM NaHCO3 and 25 mM NaCl. 120

Figure 20. A standard curve of indirect competitive ELISA using mAb13F7. The five-parameter logistic model was used to establish the standard curve. The corresponding values of IC50, limit of detection 2 (LOD) and an upper limit of quantification (ULOQ) were indicated by the red and green dotted lines, respectively. R quantifies the goodness of fit. Data were reported as % inhibition ± SEM, n =14. 121

Figure 21. A415 of 0% inhibition of different extraction buffers. AbB: antibody buffer. icELISA was performed. Results were denoted as mean ± SEM, n = 2.

122

Figure 22. Image of laboratory adulterated samples. (A) Pork protein extracts; (B) porcine whole blood adulterated in pork; (C) chicken meat protein extracts; (D) porcine whole blood adulterated in chicken meat.

123

Figure 23. Western blot to determine porcine hemoglobin concentration in laboratory spiked samples. (A) Western blot. The standard (STD) was 1 µg of porcine hemoglobin. Pa: porcine whole blood powder spiked in pork. Ca: porcine whole blood powder spiked in chicken. mAb concentration was 0.75 ppm. (B) Three dimension view of (A).

124

Figure 24. Western blot using different commercial antibodies. The samples were raw and heated (100 ºC, 15 min) porcine plasma protein extracts. Anti-P: anti- plasminogen IgY; anti-T: anti-transferrin IgG; anti-C7: anti-C7 IgG. The loading mass in each lane was 5 µg. Primary antibody concentration was 1 ppm. This image was acquired by Bio- Rad’s ChemiDoc XRS system. Anti-P was captured using EPI while the rest were captured using chemiluminescence.

125

Figure 25. Reducing and non-reducing SDS-PAGE. (A) Non-reducing SDS-PAGE of raw porcine plasma proteins (Ppr). The loading mass was 5 µg. The SDS-PAGE Molecular Weight Standards, Broad Range were loaded on the left indicating molecular weights in the non-reducing gel. (B) Reducing SDS-PAGE of raw porcine plasma proteins (Ppr). The loading mass was 5 µg. The Precision Plus Protein Unstained Standards were loaded on the right indicating the molecular weights in the reducing gel. 126

t Figure 26. Two-dimensional gel electrophoresis of heated porcine plasma proteins. (A) Ponceau S staining; (B) Western blot. Heated porcine plasma protein extracts (100 ºC, 15 min) were used as a standard with a loading mass of 5 µg. The loading mass for the first dimension was 10 μg. The mAb19C5 concentration was 1 ppm.

127

Figure 27. SDS-PAGE and Western blot to verify isolated target analyte purity and immunoreactivity. (A) SDS-PAGE; (B) Western blot. (Ppr: raw porcine plasma protein; Ppc: heated porcine plasma protein; TPr: target protein isolated from Ppr; TPc: target protein isolated from Ppc). The loading mass for Ppr and Ppc was 5 µg. For the isolated protein, they were diluted using 6 × sample buffer before loading to the gel. The mAb19C5 concentration was 1 ppm. 128

Figure 28. Western blot using commercial anti-haptoglobin mAb. Ppc: heated porcine plasma protein; TPr: target protein isolated from Ppr; TPc: target protein isolated from Ppc; Hp: haptoglobin. The loading mass for Ppc was 5 µg. For the isolated protein, they were diluted using 6 × sample buffer before loading to the gel. Primary antibody concentration (anti-haptoglobin and mAb19C5) was 1 ppm.

129

Figure 29. Western blot with protein treated with -mercaptoethanol at different concentrations. The protein was isolated from raw porcine plasma; it was diluted using 6 × sample buffer before loading to the gel. Primary antibody concentration (anti-haptoglobin and mAb19C5) was 1 ppm.

130

REFERENCES

1. Gorbatov, V. M. (1988). Collection and utilization of blood and blood proteins for edible purposes in the USSR. Advances in Meat Research, 5, 167-195.

2. Campderros, M., Furln, L. T. R., Padilla, A. P., & Campderr¨s, M. E. (2010). Functional and physical properties of bovine plasma proteins as a function of processing and pH, application in a food formulation. Advance Journal of Food Science and Technology, 2(5), 256-267.

3. Blanc, L., Papoin, J., Debnath, G., Vidal, M., Amson, R., Telerman, A., An, X. L., & Mohandas, N. (2015). Abnormal erythroid maturation leads to microcytic anemia in the TSAP6/Steap3 null mouse model. American Journal of Hematology, 90(3), 235-241.

4. Jack, A. O., & Hsieh, Y. H. (2012). The use of blood and derived products as food additives. In Y. El-Samragy (Ed.), Food Additives, (pp. 229-256): InTech, Croatia.

5. Huong, V. T. L., Hoa, N. T., Horby, P., Bryant, J. E., Kinh, N. V., Toan, T. K., & Wertheim, H. F. L. (2014). Raw pig blood consumption and potential risk for streptococcus suis infection, Vietnam. Emerging Infectious Diseases, 20(11), 1895-1898.

6. Michael, G. R. 16,000 dead pigs found in Chinese river, threatening Shanghai's water supply (2013). http://www.treehugger.com/clean-water/16000-dead-pigs-found-chinese- river-threatening-shanghais-water-supply.html. Accessed: March 12, 2017.

7. Ofori, J. A., & Hsieh, Y. H. (2007). Sandwich enzyme-linked immunosorbent assay for the detection of bovine blood in animal feed. Journal of Agricultural and Food Chemistry, 55(15), 5919-5924.

8. Rao, Q. C., & Hsieh, Y. H. (2008). Competitive enzyme-linked immunosorbent assay for quantitative detection of bovine blood in heat-processed meat and feed. Journal of Food Protection, 71(5), 1000-1006.

9. Ofori, J. A., & Hsieh, Y. H. (2016). Monoclonal antibodies as probes for the detection of porcine blood derived food ingredients. Journal of Agricultural and Food Chemistry, 64(18), 3705-3711.

10. Raja Nhari, R. M., Hamid, M., Rasli, N. M., Omar, A. R., El Sheikha, A. F., & Mustafa, S. (2016). Monoclonal antibodies specific to heat-treated porcine blood. Journal of the Science of Food and Agriculture, 96(7), 2524-2531.

11. Regenstein, J. M., Chaudry, M. M., & Regenstein, C. E. (2013). An introduction to kosher and halal food laws. In P. A. Curtis (Ed.), Guide to US Food Laws and Regulations Second ed., (pp. 171-212): John Wiley & Sons, Ltd, Chichester, UK.

12. US Food and Drug Administration. Guidance for industry: A food labeling guide (2013). http://www.fda.gov/downloads/Food/GuidanceRegulation/UCM265446.pdf. Accessed: March 13, 2017. 131

13. US Department of Agriculture. Food standards and labeling policy book (2005). http://www.fsis.usda.gov/OPPDE/larc/Policies/Labeling_Policy_Book_082005.pdf. Accessed: March 13, 2017.

14. Addeen, A., Benjakul, S., Wattanachant, S., & Maqsood, S. (2014). Effect of Islamic slaughtering on chemical compositions and post-mortem quality changes of broiler chicken meat. International Food Research Journal, 21(3), 897-907.

15. Limquizon, M. C., Benabaye, R. M., White, F. M., Dayrit, M. M., & White, M. E. (1994). Cholera in metropolitan Manila - foodborne transmission via street vendors. Bulletin of the World Health Organization, 72(5), 745-749.

16. Wang, J., Xu, C. H., & Guo, Q. H. (2010). Consumer cognition, willingness to pay and purchasing behavior of safety pork: A case of Jilin province. Journal of Jilin Agricultural University, 32(5), 586-590.

17. Edwards, S. A. (2002). Perinatal mortality in the pig: Environmental or physiological solutions? Livestock Production Science, 78(1), 3-12.

18. The comparison between safe pork and diseased pork. http://cif.mofcom.gov.cn/site/html/shizuishan/html/572341/2010/11/30/1291103519082.h tml. Accessed: March 12, 2017.

19. Warriss, P. D., & Rhodes, D. N. (1977). Hemoglobin concentrations in beef. Journal of the Science of Food and Agriculture, 28(10), 931-934.

20. Warriss, P. D. (1977). The residual blood content of meat - a review. Journal of the Science of Food and Agriculture, 28(5), 457-462.

21. Reporter, D. M. Revealed: How chefs use 'meat glue' made from pig blood to stick steaks together (2012). http://www.dailymail.co.uk/news/article-2144981/How-chefs-use-meat- glue-pig-blood-stick-steaks-together.html. Accessed: March 12, 2017.

22. Portnoy, H. Buyer beware: restaurants use ‘meat glue’ to create ‘steaks’ from scraps (2012). http://hotair.com/greenroom/archives/2012/05/20/buyer-beware-restaurants-use- meat-glue-to-create-steaks-from-scraps/. Accessed: March 12, 2017.

23. Noyes, D. "Meat glue" poses health risks for consumers (2012). http://abc7.com/archive/8642900/. Accessed: March 12, 2017.

24. Wissiack, R., de la Calle, B., Bordin, G., & Rodriguez, A. R. (2003). Screening test to detect meat adulteration through the determination of hemoglobin by cation exchange chromatography with diode array detection. Meat Science, 64(4), 427-432.

25. Alaraidh, I. A. (2008). Improved DNA extraction method for porcine contaminants, detection in imported meat to the Saudi market. Saudi Journal of Biological Sciences, 15(2), 225-229.

132

26. Lysandra L. Slocombe, & Colditz, l. G. (2011). A rapid colorimetric assay for measuring low concentrations of haemoglobin in large numbers of bovine plasma samples. Food and Agricultural Immunology, 22(2), 135-143.

27. Aida, A. A., Man, Y. B. C., Wong, C. M. V. L., Raha, A. R., & Son, R. (2005). Analysis of raw meats and fats of pigs using polymerase chain reaction for Halal authentication. Meat Science, 69(1), 47-52.

28. Shibata, I., Okuda, Y., Yazawa, S., Ono, M., Sasaki, T., Itagaki, M., Nakajima, N., Okabe, Y., & Hidejima, I. (2003). PCR detection of porcine circovirus type 2 DNA in whole blood, serum, oropharyngeal swab, nasal swab, and feces from experimentally infected pigs and field cases. Journal of Veterinary Medical Science, 65(3), 405-408.

29. Hoelzle, L. E., Adelt, D., Hoelzle, K., Heinritzi, K., & Wittenbrink, M. M. (2003). Development of a diagnostic PCR assay based on novel DNA sequences for the detection of Mycoplasma suis (Eperythrozoon suis) in porcine blood. Veterinary Microbiology, 93(3), 185-196.

30. Ramahefarisoa, R. M., Rakotondrazaka, M., Jambou, R., & Carod, J. F. (2010). Comparison of ELISA and PCR assays for the diagnosis of porcine cysticercosis. Veterinary Parasitology, 173(3-4), 336-339.

31. Wang, Q., Zhuang, M., Yu, B., Zhang, L., & Chen, X. (2007). Study on the activity of peroxidase in pork using TMB method. Chinese Journal of Analysis Laboratory, 26(7), 14-16.

32. Crowther, J. R. (2000). The ELISA guidebook (Second ed. Vol. 149): Springer Science and Business Media, Berlin, Germany.

33. Harlow, E., & Lane, D. (1988). Antibodies: A laboratory manual (Second ed. Vol. 139): Cold Spring Harbor Laboratory Press, New York.

34. Lipman, N. S., Jackson, L. R., Trudel, L. J., & Weis-Garcia, F. (2005). Monoclonal versus polyclonal antibodies: Distinguishing characteristics, applications, and information resources. Ilar Journal, 46(3), 258-268.

35. Friguet, B., Chaffotte, A. F., Djavadiohaniance, L., & Goldberg, M. E. (1985). Measurements of the true affinity constant in solution of antigen-antibody complexes by enzyme-linked immunosorbent-assay. Journal of Immunological Methods, 77(2), 305- 319.

36. Beatty, J. D., Beatty, B. G., & Vlahos, W. G. (1987). Measurement of monoclonal- antibody affinity by noncompetitive enzyme-immunoassay. Journal of Immunological Methods, 100(1-2), 173-179.

37. Dam, T. K., Torres, M., Brewer, C. F., & Casadevall, A. (2008). Isothermal titration calorimetry reveals differential binding thermodynamics of variable region-identical

133

antibodies differing in constant region for a univalent ligand. The Journal of Biological Chemistry, 283(46), 31366-31370.

38. Malmqvist, M. (1993). Surface plasmon resonance for detection and measurement of antibody-antigen affinity and kinetics. Current Opinion in Immunology, 5(2), 282-286.

39. Steward, M. W. (1981). The biological significance of antibody affinity. Immunology Today, 2(7), 134-140.

40. Self, C. H., Dessi, J. L., & Winger, L. A. (1994). High-performance assays of small molecules: Enhanced sensitivity, rapidity, and convenience demonstrated with a noncompetitive immunometric anti-immune complex assay system for digoxin. Clinical Chemistry, 40(11), 2035-2041.

41. Chang, B. S., Kendrick, B. S., & Carpenter, J. F. (1996). Surface-induced denaturation of proteins during freezing and its inhibition by surfactants. Journal of Pharmaceutical Sciences, 85(12), 1325-1330.

42. Sloan, A. E. The top ten food trends (2015). http://www.ift.org/food-technology/past- issues/2015/april/features/the-top-ten-food-trends.aspx?page=viewall. Accessed: March 12, 2017.

43. Cal, K., & Sollohub, K. (2010). Spray drying technique. I: Hardware and process parameters. Journal of Pharmaceutical Sciences, 99(2), 575-586.

44. Mehansho, H. (2006). Iron fortification technology development: New approaches. Journal of Nutrition, 136(4), 1059-1063.

45. US Department of Agriculture. Livestock slaughter 2015 summary (2016). http://usda.mannlib.cornell.edu/usda/current/LiveSlauSu/LiveSlauSu-04-20-2016.pdf. Accessed: March 13, 2017.

46. J. Wismer-Pedersen. (1988). Use of haemoglobin in foods - a review. Meat Science, 24(1), 31-45.

47. Clara S.F. Bah, Alaa El-Din A. Bekhit, Alan Carne, & McConnell, M. A. (2013). Slaughterhouse blood: An emerging source of bioactive compounds. Comprehensive Reviews in Food Science and Food Safety, 12(3), 314-331.

48. Nowak, B., & von Mueffling, T. (2006). Porcine blood cell concentrates for food products: Hygiene, composition, and preservation. Journal of Food Protection, 69(9), 2183-2192.

49. Christine V. Sapan, & Lundblad, R. L. (2006). Considerations for the use of blood plasma and serum for proteomic analysis. Cancer Genomics and Proteomics, 3(1), 227- 230.

134

50. Zhang, K., Liu, Y., Shang, Y., Zheng, H., Guo, J., Tian, H., Jin, Y., He, J., & Liu, X. (2012). Analysis of pig serum proteins based on shotgun liquid chromatography-tandem mass spectrometry. African Journal of Biotechnology, 11(57), 12118-12127.

51. Perutz, M. F., Miurhead, H., Cox, J. M., Goaman, L. C., Mathews, F. S., McGandy, E. L., & Webb, L. E. (1968). Three-dimensional Fourier synthesis of horse oxyhaemoglobin at 2.8 A resolution: (1) x-ray analysis. Nature, 219(5149), 29-32.

52. Wyman, J., Jr. (1964). Linked functions and reciprocal effects in hemoglobin: a second look. Advances in Protein Chemistry, 19, 223-286.

53. Katz, D. S., White, S. P., Huang, W., Kumar, R., & Christianson, D. W. (1994). Structure determination of aquomet porcine hemoglobin at 2.8 angstrom resolution. Journal of Molecular Biology, 244(5), 541-553.

54. Antonini, E. (1965). Interrelationship between structure and function in hemoglobin and myoglobin. Physiological Reviews, 45(1), 123-170.

55. Schroeder, W. A., Huisman, T. H., Shelton, J. R., Shelton, J. B., Kleihauer, E. F., Dozy, A. M., & Robberson, B. (1968). Evidence for multiple structural genes for the gamma chain of human fetal hemoglobin. Proceedings of the National Academy of Sciences, 60(2), 537-544.

56. Giardina, B., Messana, I., Scatena, R., & Castagnola, M. (1995). The multiple functions of hemoglobin. Critical Reviews in Biochemistry and Molecular Biology, 30(3), 165-196.

57. Hall, F. G., & Gray, I. (1929). The hemoglobin concentration of the blood of marine fishes. Journal of Biological Chemistry, 81(3), 589-594.

58. Larimer, J. L. (1959). Hemoglobin concentration and oxygen capacity of mammalian blood. Journal of the Mitchell Society, 174-188.

59. Huisman, T., Reynolds, C., Dozy, A., & Wilson, J. (1965). The structure of sheep hemoglobins, the amino acid compositions of the α and chains of the hemoglobins, A, B, and C. Journal of Biological Chemistry, 240(6), 2455-2460.

60. Hashimoto, K., & Wilt, F. H. (1966). The heterogeneity of chicken hemoglobin. Proceedings of the National Academy of Sciences, 56(5), 1477-1483.

61. Vesterberg, O., & Svensson, H. (1966). Isoelectric fractionation analysis and characterization of ampholytes in natural pH gradients: 4. Further studies on resolving power in connection with separation of myoglobins. Acta Chemica Scandinavica, 20(3), 820-834.

62. James, W. D., Piergiorgio, R., & H. Franklin, B. (1971). The separation of human and animal hemoglobins by isoelectric focusing in polycarylamide gel. Biochimica et Biophysica Acta, 229(1), 42-50.

135

63. Basset, P., Beuzard, Y., Garel, M. C., & Rosa, J. (1978). Isoelectric focusing of human hemoglobin: its application to screening, to the characterization of 70 variants, and to the study of modified fractions of normal hemoglobins. Blood, 51(5), 971-982.

64. Altland, K., Kaempfer, M., & Granda, H. (1979). Improved screening test for abnormal hemoglobins from dried blood samples. Human Genetics, 53(1), 97-100.

65. Giovanni, R., Claudette, M., & Thaddeus, W. B. (1978). Resolution of hemoglobin subunits by electrophoresis in acid urea polyacrylamide gels containing Triton X-100. Analytical Biochemistry, 85(2), 506-518.

66. Kanko, S., & Autio, K. (1985). Electrophoretic studies of blood globin preparations. Journal of Chromatography, 324(2), 395-406.

67. Lie-Injo, L. E., Solai, A., & Ganesan, J. (1978). Globin chain separation by SDS polyacrylamide gel electrophoresis: simple screening method for elongated hemoglobin chains. Journal of Chromatography, 153(1), 161-165.

68. Wada, Y. (2002). Advanced analytical methods for hemoglobin variants. Journal of Chromatography B-Analytical Technologies in the Biomedical and Life Sciences, 781(1- 2), 291-301.

69. Castagnola, M., Messana, I., Cassiano, L., Rabino, R., Rossetti, D. V., & Giardina, B. (1995). The use of capillary electrophoresis for the determination of hemoglobin-variants. Electrophoresis, 16(8), 1492-1498.

70. Sahin, A., Laleli, Y. R., & Ortancil, R. (1995). Hemoglobin analysis by capillary zone electrophoresis. Journal of Chromatography A, 709(1), 121-125.

71. Hempe, J. M., & Craver, R. D. (1994). Quantification of hemoglobin variants by capillary isoelectric focusing. Clinical Chemistry, 40(12), 2288-2295.

72. Harper, S. B., Hurst, W. J., & Lang, C. M. (1994). Use of capillary electrophoresis isoelectric-focusing for the determination of bovine hemoglobin-variants. Journal of Chromatography B-Biomedical Applications, 657(2), 339-344.

73. Sittivilai, R., Sribhen, C., Isariyodom, S., Songserm, T., & Choothesa, A. (2004). A chromatographic and electrophoretic study of hemoglobin of domestic fowl. Kasetsart Journal, 38(6), 132-136.

74. Jandaruang, J., Siritapetawee, J., Songsiriritthigul, C., Preecharram, S., Azuma, T., Dhiravisit, A., Fukumori, Y., & Thammasirirak, S. (2014). Purification, characterization, and crystallization of crocodylus siamensis hemoglobin. Protein Journal, 33(4), 377-385.

75. Size exclusion chromatography: Principles and methods. https://www.gelifesciences.com/gehcls_images/GELS/Related%20Content/Files/146615 8841539/litdoc18102218_20161012165335.pdf. Accessed: March 12, 2017.

136

76. Wan, J. H., Tian, P. L., Luo, W. H., Wu, B. Y., Xiong, F., Zhou, W. J., Wei, X. C., & Xu, X. M. (2012). Rapid determination of human globin chains using reversed-phase high- performance liquid chromatography. Journal of Chromatography B-Analytical Technologies in the Biomedical and Life Sciences, 90(1), 53-58.

77. Nemati, H., Bahrami, G., & Rahimi, Z. (2011). Rapid separation of human globin chains in normal and thalassemia patients by RP-HPLC. Molecular Biology Reports, 38(5), 3213-3218.

78. Masala, B., & Manca, L. (1994). Detection of globin chains by reversed-phase high- performance liquid-chromatography. Hemoglobins, Part B: Biochemical and Analytical Methods, 231, 21-44.

79. Wilson, J. B., Headlee, M. E., & Huisman, T. H. J. (1983). A new high-performance liquid-chromatographic procedure for the separation and quantitation of various hemoglobin-variants in adults and newborn babies. Journal of Laboratory and Clinical Medicine, 102(2), 174-186.

80. Frederuck, E. R., & Shung-Ho, C. (1978). Patent: Method for rapid analysis of hemoglobin variants by high speed liquid chromatography. Patent #: 4108603.

81. Kutlar, A., Kutlar, F., Wilson, J. B., Headlee, M. G., & Huisman, T. H. J. (1984). Quantitation of hemoglobin components by high-performance cation-exchange liquid- chromatography - its use in diagnosis and in the assessment of cellular-distribution of hemoglobin-variants. American Journal of Hematology, 17(1), 39-53.

82. Mandal, P. K., Rao, V. K., Kowale, B. N., & Pal, U. K. (1999). Utilization of slaughter house blood in human food. Journal of Food Science and Technology-Mysore, 36(2), 91- 105.

83. Jack, A. O., & Hsieh, Y. H. (2011). Blood-derived products for human consumption Revelation and Science, 1(1), 14-21.

84. Wang, F., & Lin, C. (1994). The effects of heating and chemical treatment on the haem and non-haem iron content of heat-induced porcine blood curd. Journal of the Science of Food and Agriculture, 65(2), 209-213.

85. Wang, F.-S., & Lin, C.-W. (1994). Molecular forces involved in heat-induced porcine blood curd. Journal of Agricultural and Food Chemistry, 42(5), 1085-1088.

86. Churillo, T. O. (2014). Filipino "Kulinarya" cuisine and chronic kidney disease. Journal of Renal Nutrition, 24(4), 31-35.

87. Pierce, J. L., Cromwell, G. L., Lindemann, M. D., Russell, L. E., & Weaver, E. M. (2005). Effects of spray-dried animal plasma and immunoglobulins on performance of early weaned pigs. Journal of Animal Science, 83(12), 2876-2885.

137

88. Jamroz, D., Wiliczkiewicz, A., Orda, J., Skorupinska, J., Slupczynska, M., & Kuryszko, J. (2011). Chemical composition and biological value of spray dried porcine blood by- products and bone protein hydrolysate for young chickens. British Poultry Science, 52(5), 589-605.

89. Ermer, P. M., Miller, P. S., & Lewis, A. J. (1994). Diet preference and meal patterns of weanling pigs offered diets containing either spray-dried porcine plasma or dried skim milk. Journal of Animal Science, 72(6), 1548-1554.

90. Hansen, J. A., Nelssen, J. L., Goodband, R. D., & Weeden, T. L. (1993). Evaluation of animal protein supplements in diets of early-weaned pigs. Journal of Animal Science, 71(7), 1853-1862.

91. Hemoglobin powder in aqua feed. https://d118ospkkl5uqf.cloudfront.net/Sonac/Sonac_leaflets/Hemoglobin%20Low%20Re s_01.pdf. Accessed: March 12, 2017.

92. Duangrat, C., Chutima, T., Amornrat, P., & Manee, S. (2010). Effect of hemoglobin powder substituted for fishmeal on growth performance, protein digestibility, and trypsin gene expression in Litopenaeus vannamei. Songklanakarin Journal of Science and Technology, 32(2), 119-127.

93. Martínez-Llorens, S., Vidal, A. T., Moñino, A. V., Gómez Ader, J., Torres, M. P., & Cerdá, M. J. (2008). Blood and haemoglobin meal as protein sources in diets for gilthead sea bream (Sparus aurata): Effects on growth, nutritive efficiency and fillet sensory differences. Aquaculture Research, 39(10), 1028-1037.

94. Lee, K.-J., & Bai, S. C. (1997). Hemoglobin powder as a dietary animal protein source for juvenile nile tilapia. The Progressive Fish-Culturist, 59(4), 266-271.

95. Ramos-Clamont, G., Fernandez-Michel, S., Carrillo-Vargas, L., Martinez-Calderon, E., & Vazquez-Moreno, L. (2003). Functional properties of protein fractions isolated from porcine blood. Journal of Food Science, 68(4), 1196-1200.

96. Visessanguan, W., Benjakul, S., & An, H. (2000). Porcine plasma proteins as a surimi protease inhibitor: Effects on actomyosin gelation. Journal of Food Science, 65(4), 607- 611.

97. Howell, N. K., & Lawrie, R. A. (1984). Functional aspects of blood-plasma proteins: 2. Gelling properties. Journal of Food Technology, 19(3), 289-295.

98. Hurtado, S., Daga, I., Espigule, E., Pares, D., Saguer, E., Toldra, M., & Carreteroa, C. (2011). Use of porcine blood plasma in "phosphate-free frankfurters". 11th International Congress on Engineering and Food, 1, 477-482.

99. Pares, D., Saguer, E., Saurina, J., Sunol, J. J., & Carretero, C. (1998). Functional properties of heat induced gels from liquid and spray-dried porcine blood plasma as influenced by pH. Journal of Food Science, 63(6), 958-961. 138

100. Pares, D., & Ledward, D. A. (2001). Emulsifying and gelling properties of porcine blood plasma as influenced by high-pressure processing. Food Chemistry, 74(2), 139-145.

101. Davila, E., Pares, D., Cuvelier, G., & Relkin, P. (2007). Heat-induced gelation of porcine blood plasma proteins as affected by pH. Meat Science, 76(2), 216-225.

102. Tsai, C. M., Tseng, T. F., Yang, J. H., & Chen, M. T. (2006). Study on a binder by using porcine blood plasma transglutaminase, thrombin and fibrinogen. Asian-Australasian Journal of Animal Sciences, 19(1), 137-143.

103. Tseng, T. F., Liu, D. C., & Chen, M. T. (2000). Characteristics of transglutaminase derived from pig plasma. Journal of the Agricultural Association of China, 1(4), 452-458.

104. Tseng, T. F., Tsai, C. M., Yang, J. H., & Chen, M. T. (2006). Porcine blood plasma transgluataminase combined with thrombin and fibrinogen as a binder in restructured meat. Journal of Animal Science, 19(7), 1054-1058.

105. Nakamura, R., Hayakawa, S., Yasuda, K., & Sato, Y. (1984). Emulsifying properties of bovine blood globin: A comparison with some proteins and their improvement. Journal of Food Science, 49(1), 102-104.

106. Tybor, P. T., Dill, C. W., & Landmann, W. A. (1973). Effect of decolorization and lactose incorporation on emulsification capacity of spray-dried blood protein concentrates. Journal of Food Science, 38(1), 4-6.

107. Toldrà, M., Dàvila, E., Saguer, E., Fort, N., Salvador, P., Parés, D., & Carretero, C. (2008). Functional and quality characteristics of the red blood cell fraction from biopreserved porcine blood as influenced by high pressure processing. Meat Science, 80(2), 380-388.

108. Salvador, P., Saguer, E., Pares, D., Carretero, C., & Toldra, M. (2010). Foaming and emulsifying properties of porcine red cell protein concentrate. Food Science and Technology International, 16(4), 289-296.

109. Toldra, M., Elias, A., Pares, D., Saguer, E., & Carretero, C. (2004). Functional properties of a spray-dried porcine red blood cell fraction treated by high hydrostatic pressure. Food Chemistry, 88(3), 461-468.

110. Cozzone, A. J. (2001). Proteins: Fundamental chemical properties. In A. Finazzi-Agrò (Ed.), eLS, (pp. 1-12): Wiley, U.S.

111. Tybor, P. T., Dill, C. W., & Landmann, W. A. (1973). Effect of decolorization and lactose incorporation on the emulsification capacity of spraydried blood protein concentrates. Journal of Food Science, 38(1), 4-6.

112. Sato, Y., Hayakawa, S., & Hayakawa, M. (1981). Preparation of blood globin through carboxymethyl cellulose chromatography. Journal of Food Technology, 16(1), 81-91.

139

113. Yang, J. H., & Lin, C. W. (1996). Effects of various viscosity enhancers and pH on separating haem from porcine red blood cells. Journal of the Science of Food and Agriculture, 70(3), 364-368.

114. Hayakawa, S., Matsuura, Y., Nakamura, R., & Sato, Y. (1986). Effect of heat-treatment on preparation of colorless globin from bovine hemoglobin using soluble carboxymethyl cellulose. Journal of Food Science, 51(3), 786-790.

115. Yang, J. T., & Lin, C. W. (2000). Decolorization of porcine red blood cell globin with ion exchanger method and modification of its protein functionalities. Asian-Australasian Journal of Animal Sciences, 13(12), 1770-1774.

116. Yang, J. H., & Lin, C. W. (1998). Functional properties of porcine blood globin decolourized by different methods. International Journal of Food Science and Technology, 33(4), 419-427.

117. Miller, J. L. (2013). Iron deficiency anemia: A common and curable disease. Cold Spring Harbor Perspectives in Medicine, 3(7), 1-13.

118. Kongkachuichai, R., Napatthalung, P., & Charoensiri, R. (2002). Heme and nonheme iron content of animal products commonly consumed in Thailand. Journal of Food Composition and Analysis, 15(4), 389-398.

119. Conrad, M. E., Weintraub, L. R., Sears, D. A., & Crosby, W. H. (1966). Absorption of hemoglobin iron. American Journal of Physiology, 211(5), 1123-1130.

120. Reizenstein, P. (1979). Hemoglobin fortification of food and prevention of iron deficiency with heme iron. Acta Medica Scandinavica, 629, 1-47.

121. South, P. K., Lei, X. G., & Miller, D. D. (2000). Meat enhances nonheme iron absorption in pigs. Nutrition Research, 20(12), 1749-1759.

122. Fly, A. D., & Czarnecki-Maulden, G. L. (2000). Iron bioavailability from hemoglobin and hemin in chick, rat, cat, and dog: A comparative study. Nutrition Research, 20(2), 237-248.

123. Gera, T., Sachdev, H. S., & Boy, E. (2012). Effect of iron-fortified foods on hematologic and biological outcomes: Systematic review of randomized controlled trials. The American Journal of Clinical Nutrition, 96(2), 309-324.

124. Vaghefi, N., Guillochon, D., Bureau, F., Jauzac, P., Neuville, D., Jacob, B., Arhan, P., & Bougle, D. (1998). Effect of stabilising amino acids on the digestive absorption of heme and non-heme iron. Reproduction Nutrition Development, 38(5), 559-566.

125. Uzel, C., & Conrad, M. E. (1998). Absorption of heme iron. Seminars in Hematology, 35(1), 27-34.

140

126. Olivares, M., Hertrampf, E., Pizzarro, F., Walter, T., Cayazzo, M., Llaguno, S., Chadud, P., Cartagena, N., Vega, V., Amar, M., & et al. (1990). Hemoglobin-fortified biscuits: bioavailability and its effect on iron nutriture in school children. Archivos Latinoamericanos de Nutrición, 40(2), 209-220.

127. Eichler, K., Wieser, S., Ruthemann, I., & Brugger, U. (2012). Effects of micronutrient fortified milk and cereal food for infants and children: a systematic review. BMC Public Health, 12, 506.

128. Piedad, M. M. C., Yesid, A. M. L., & Lesbia, C. J. G. (2015). Protein quality of rice drinks fortified with bovine and porcine blood plasma. Revista Facultad Nacional de Agronomía, 68(1), 7487-7496.

129. Delaney, R. A. (1975). The nutritive value of porcine blood plasma concentrates prepared by ultrafiltration and spray drying. Journal of Science of Food and Agriculture, 26(3), 303-310.

130. Bates, R. P., Wu, L. C., & Murphy, B. (1974). Use of animal blood and cheese whey in bread: nutritive-value and acceptance. Journal of Food Science, 39(3), 585-587.

131. Kuppevelt, A. V., Levin, G., & Reizenstein, P. (1976). Small-scale production and net protein-utilization of globin-globin fortification of cereals. Nutrition Reports International, 13(5), 429-442.

132. Slinde, E. (1987). Color of black salami sausage - dissociation of heme from myoglobin and hemoglobin. Journal of Food Science, 52(5), 1152-1154.

133. Salvador, P., Toldra, M., Pares, D., Carretero, C., & Saguer, E. (2009). Color stabilization of porcine hemoglobin during spray-drying and powder storage by combining chelating and reducing agents. Meat Science, 83(2), 328-333.

134. Fontes, P. R., Gomide, L. A. M., Ramos, E. M., Stringheta, P. C., & Parreiras, J. F. M. (2004). Color evaluation of carbon monoxide treated porcine blood. Meat Science, 68(4), 507-513.

135. Sorheim, O., Aune, T., & Nesbakken, T. (1997). Technological, hygienic and toxicological aspects of carbon monoxide used in modified-atmosphere packaging of meat. Trends in Food Science & Technology, 8(9), 307-312.

136. Yu, Y., Hu, J. N., Miyaguchi, Y. J., Bai, X. F., Du, Y. G., & Lin, B. C. (2006). Isolation and characterization of angiotensin I-converting enzyme inhibitory peptides derived from porcine hemoglobin. Peptides, 27(11), 2950-2956.

137. Mito, K., Fujii, M., Kuwahara, M., Matsumura, N., Shimizu, T., Sugano, S., & Karaki, H. (1996). Antihypertensive effect of angiotensin I-converting enzyme inhibitory peptides derived from hemoglobin. European Journal of Pharmacology, 304(1-3), 93-98.

141

138. Yan, R., Wan, A., Lu, X.-m., Chen, X.-m., Zhang, T.-e., & Guo, J.-L. (2011). Isolation and characterization of angiotensin Iconverting enzyme inhibitor peptides derived from porcine hemoglobin. Scientific Research and Essay, 6(30 ), 6262-6269.

139. Shi, J. H., Ross, C. R., Chengappa, M. M., Sylte, M. J., McVey, D. S., & Blecha, F. (1996). Antibacterial activity of a synthetic peptide (PR-26) derived from PR-39, a proline-arginine-rich neutrophil antimicrobial peptide. Antimicrobial Agents and Chemotherapy, 40(1), 115-121.

140. Harwig, S. S. L., Kokryakov, V. N., Swiderek, K. M., Aleshina, G. M., Zhao, C. Q., & Lehrer, R. I. (1995). Prophenin-1, an exceptionally proline-rich antimicrobial peptide from porcine leukocytes. FEBS Letters, 362(1), 65-69.

141. Kokryakov, V. N., Harwig, S. S. L., Panyutich, E. A., Shevchenko, A. A., Aleshina, G. M., Shamova, O. V., Korneva, H. A., & Lehrer, R. I. (1993). Protegrins-leukocyte antimicrobial peptides that combine features of corticostatic defensins and tachyplesins. FEBS Letters, 327(2), 231-236.

142. Huang, X. X., Gao, C. Y., Zhao, Q. J., & Li, C. L. (2015). Antimicrobial characterization of site-directed mutagenesis of porcine beta defensin 2. Plos One, 10(2), e0118170.

143. Veldhuizen, E. J. A., Rijnders, M., Claassen, E. A., van Dijk, A., & Haagsman, H. P. (2008). Porcine beta-defensin 2 displays broad antimicrobial activity against pathogenic intestinal bacteria. Molecular Immunology, 45(2), 386-394.

144. Xu, X., Cao, R., He, L., & Yang, N. (2009). Antioxidant activity of hydrolysates derived from porcine plasma. Journal of the Science of Food and Agriculture, 89(11), 1897-1903.

145. Liu, Q., Kong, B. H., Xiong, Y. L. L., & Xia, X. F. (2010). Antioxidant activity and functional properties of porcine plasma protein hydrolysate as influenced by the degree of hydrolysis. Food Chemistry, 118(2), 403-410.

146. Lertittikul, W., Benjakul, S., & Tanaka, M. (2007). Characteristics and antioxidative activity of Maillard reaction products from a porcine plasma protein-glucose model system as influenced by pH. Food Chemistry, 100(2), 669-677.

147. Benjakul, S., Lertittikul, W., & Bauer, F. (2005). Antioxidant activity of Maillard reaction products from a porcine plasma protein-sugar model system. Food Chemistry, 93(2), 189-196.

148. Liu, Q., Li, J., Kong, B. H., Jia, N., & Li, P. J. (2014). Antioxidant capacity of Maillard reaction products formed by a porcine plasma protein hydrolysate-sugar model system as related to chemical characteristics. Food Science and Biotechnology, 23(1), 33-41.

149. The future of world religions: population growth projections, 2010-2050. (2015). http://www.pewforum.org/2015/04/02/religious-projections-2010-2050/. Accessed: March 12, 2017.

142

150. Nakyinsige, K., Man, Y. B., & Sazili, A. Q. (2012). Halal authenticity issues in meat and meat products. Meat Science, 91(3), 207-214.

151. Proferrin Forte: Heme iron polypeptide + folic acid. http://www.proferrin.com/proferrinforte-home/. Accessed: March 12, 2017.

152. Immunolin: Superior immune system support. http://buyersguide.supplysideshow.com/media/23/library/NPI_08_08_wilke_immunolin. pdf. Accessed: March 12, 2017.

153. Rohman, A., & Man, Y. B. C. (2012). Analysis of pig derivatives for halal authentication studies. Food Reviews International, 28(1), 97-112.

154. Nakyinsige, K., Fatimah, A. B., Aghwan, Z. A., Zulkifli, I., Goh, Y. M., & Sazili, A. Q. (2014). Bleeding efficiency and meat oxidative stability and microbiological quality of new zealand white rabbits subjected to halal slaughter without stunning and gas stun- killing. Asian-Australasian Journal of Animal Sciences, 27(3), 406-413.

155. Kotula, A. W., & Helbacka, N. V. (1966). Blood retained by chicken carcasses and cut- up parts as influenced by slaughter method. Poultry Science, 45(2), 404-410.

156. Regenstein, J. M., Chaudry, M. M., & Regenstein, C. E. (2003). The kosher and halal food laws. Comprehensive Reviews in Food Science and Food Safety, 2(3), 111-127.

157. Warriss, P. D. (1984). Exsanguination of animals at slaughter and the residual blood content of meat. Veterinary Record, 115(12), 292-295.

158. Calvo, J. H., Osta, R., & Zaragoza, P. (2002). Quantitative PCR detection of pork in raw and heated ground beef and pate. Journal of Agricultural and Food Chemistry, 50(19), 5265-5267.

159. Rohman, A., Sismindari, Erwanto, Y., & Man, Y. B. C. (2011). Analysis of pork adulteration in beef meatball using Fourier transform infrared (FTIR) spectroscopy. Meat Science, 88(1), 91-95.

160. Ayaz, Y., Ayaz, N. D., & Erol, I. (2006). Detection of species in meat and meat products using enzyme-linked immunosorbent assay. Journal of Muscle Foods, 17(2), 214-220.

161. Hsieh, Y. H., Woodward, B. B., & Ho, S. H. (1995). Detection of species substitution in raw and cooked meats using immunoassays. Journal of Food Protection, 58(5), 555-559.

162. von Bargen, C., Brockmeyer, J., & Humpf, H. U. (2014). Meat authentication: A new HPLC-MS/MS based method for the fast and sensitive detection of horse and pork in highly processed food. Journal of Agricultural and Food Chemistry, 62(39), 9428-9435.

163. Al-Bachir, M., & Mehio, A. (2001). Irradiated luncheon meat: microbiological, chemical and sensory characteristics during storage. Food Chemistry, 75(2), 169-175.

143

164. Ballin, N. Z. (2010). Authentication of meat and meat products. Meat Science, 86(3), 577-587.

165. Wertheim, H. F. L., Nghia, H. D. T., Taylor, W., & Schultsz, C. (2009). Streptococcus suis: An emerging human pathogen. Clinical Infectious Diseases, 48(5), 617-625.

166. Yu, H. J., Jing, H. Q., Chen, Z. H., Zheng, H., Zhu, X. P., Wang, H., Wang, S. W., Liu, L. G., Zu, R. Q., Luo, L. Z., Xiang, N. J., Liu, H. L., Liu, X. C., Shu, Y. L., Lee, S. S., Chuang, S. K., Wang, Y., Xu, J. G., & Yang, W. Z. (2006). Human Streptococcus suis outbreak, Sichuan, China. Emerging Infectious Diseases, 12(6), 914-920.

167. Nguyen, T. H. M., Ngo, T. H., Tran, V. T. N., Le, D. L., Tran, T. H. C., Dinh, X. S., Nguyen, H. P., Ly, V. C., To, S. D., James, C., Ho, D. T. N., Tran, N. M., Nguyen, V. V. C., Menno, D. D., Nguyen, T. C., Tran, T. H., Jeremy, F., & Constance, S. (2008). Streptococcus suis meningitis in adults in Vietnam. Clinical Infectious Diseases, 46(5), 659-667.

168. Huong, V. T. L., Ha, N., Huy, N. T., Horby, P., Nghia, H. D. T., Thiem, V. D., Zhu, X. T., Hoa, N. T., Hien, T. T., Zamora, J., Schultsz, C., Wertheim, H. F. L., & Hirayama, K. (2014). Epidemiology, clinical manifestations, and outcomes of streptococcus suis infection in humans. Emerging Infectious Diseases, 20(7), 1105-1114.

169. Stevenson, G. W., Hoang, H., Schwartz, K. J., Burrough, E. R., Sun, D., Madson, D., Cooper, V. L., Pillatzki, A., Gauger, P., Schmitt, B. J., Koster, L. G., Killian, M. L., & Yoon, K. J. (2013). Emergence of porcine epidemic diarrhea virus in the United States: clinical signs, lesions, and viral genomic sequences. Journal of Veterinary Diagnostic Investigation, 25(5), 649-654.

170. Brassard, J., Gagne, M. J., Lamoureux, L., Inglis, G. D., Leblanc, D., & Houde, A. (2008). Molecular detection of bovine and porcine Torque teno virus in plasma and feces. Veterinary Microbiology, 126(1-3), 271-276.

171. Allan, G. M., & Ellis, J. A. (2000). Porcine circoviruses: A review. Journal of Veterinary Diagnostic Investigation, 12(1), 3-14.

172. Pig producers warned to be wary when using spray-dried porcine plasma. (2014). Veterinary Record, 174(12), 291-291.

173. NASDBPP. Spray dried porcine blood products are safe porcine epidemic diarrhea virus (2014). https://www.aasv.org/pedv/NASDBPP_PR.pdf. Accessed: March 12, 2017.

174. Polo, J., Opriessnig, T., O'Neill, K. C., Rodriguez, C., Russell, L. E., Campbell, J. M., Crenshaw, J., Segales, J., & Pujols, J. (2013). Neutralizing antibodies against porcine circovirus type 2 in liquid pooled plasma contribute to the biosafety of commercially manufactured spray-dried porcine plasma. Journal of Animal Science, 91(5), 2192-2198.

144

175. Opriessnig, T., Xiao, C. T., Gerber, P. F., Zhang, J. Q., & Halbur, P. G. (2014). Porcine epidemic diarrhea virus RNA present in commercial spray-dried porcine plasma is not infectious to naive pigs. Plos One, 9(8), 1-10.

176. Fitzgerald, R. F., Stalder, K. J., Matthews, J. O., Kaster, C. M. S., & Johnson, A. K. (2009). Factors associated with fatigued, injured, and dead pig frequency during transport and lairage at a commercial abattoir. Journal of Animal Science, 87(3), 1156-1166.

177. Wang, X. Y. 60 million heads of pigs die from illness each year in China: Small ratio of them are of safety-disposal (2016). http://news.163.com/16/0928/20/C22V3KJ900014SEH.html. Accessed: March 12, 2017.

178. Guo, K., Zhang, Y., Ren, X., Wang, X., & Wang, J. (2010). Comparison of detection methods of diseased pork in the way of peroxide enzyme and bacterial toxin enzymatic colorimetry. Progress in Veterinary Medicine, 31(S), 48-50.

179. Han, D., Mcmillin, K. W., & Godber, J. S. (1994). Hemoglobin, myoglobin, and total pigments in beef and chicken muscles - chromatographic determination. Journal of Food Science, 59(6), 1279-1282.

180. Warriss, P. D. (1978). Factors affecting the residual blood content of meat. Meat Science, 2(2), 155-159.

181. Smith, C. J., & Bonwick, G. A. (2007). Considerations on labour- and cost-efficient immunoassay protocols and formats exemplified by the detection of food adulteration and fraud In A. Van, D. Barug & M. Lauwaars (Eds.), Rapid Methods for Food and Feed Quality Determination., (pp. 125-137): Wageningen Academic Publishers, Wageningen.

182. Hargin, K. D. (1996). Authenticity issues in meat and meat products. Meat Science, 43, S277-S289.

183. Kitty, S. EFSA backs use of controversial binding agent in Europe (2015). http://www.globalmeatnews.com/Safety-Legislation/EFSA-backs-use-of-controversial- binding-agent-in-Europe. Accessed: March 12, 2017.

184. Kim, K. B. W. R., Lee, S. Y., Song, E. J., Park, J. G., Lee, J. W., Byun, M. W., Kim, K. E., & Ahn, D. H. (2010). Changes in allergenicity of porcine serum albumin by gamma irradiation. Korean Journal for Food Science of Animal Resources, 30(3), 397-402.

185. Ayuso, R., Lehrer, S. B., Lopez, M., Reese, G., Ibanez, M. D., Esteban, M. M., Ownby, D. R., & Schwartz, H. (2000). Identification of bovine IgG as a major cross-reactive vertebrate meat allergen. Allergy, 55(4), 348-354.

186. Del Hoyo, P., Rendueles, M., & Diaz, M. (2008). Effect of processing on functional properties of animal blood plasma. Meat Science, 78(4), 522-528.

187. Del Hoyo, P., Moure, F., Rendueles, M., & Diaz, M. (2007). Demineralization of animal blood plasma by ion exchange and ultrafiltration. Meat Science, 76(3), 402-410.

145

188. Grundy, H. H., Reece, P., Sykes, M. D., Clough, J. A., Audsley, N., & Stones, R. (2008). Method to screen for the addition of porcine blood-based binding products to foods using liquid chromatography/triple quadrupole mass spectrometry. Rapid Communications in Mass Spectrometry, 22(12), 2006-2008.

189. Irger, B., Birgit, H., Desmond, H., & Lisbeth, T. (1978). A two-step fibrinogen-fibrin transition in blood coagulation. Nature, 275, 501-505.

190. Sellers, J. P., & Clark, H. G. (1981). High pressure liquid chromatography of fibrinopeptides derived from eight mammalian fibrinogens. Thrombosis Research, 23(1- 2), 91-95.

191. Espinoza, E. O., Kirms, M. A., & Filipek, M. S. (1996). Identification and quantitation of source from hemoglobin of blood and blood mixtures by high performance liquid chromatography. Journal of Forensic Sciences, 41(5), 804-811.

192. Lyon, B. G., Davis, C. E., & Lyon, C. E. (1986). Detection of blood (hemoglobin) contamination in poultry muscle tissue. Poultry Science, 65(10), 1919-1924.

193. Tanabe, S., Miyauchi, E., Muneshige, A., Mio, K., Sato, C., & Sato, M. (2007). PCR method of detecting pork in foods for verifying allergen labeling and for identifying hidden pork ingredients in processed foods. Biosci Biotechnol Biochem, 71(7), 1663- 1667.

194. Tanabe, S., Hase, M., Yano, T., Sato, M., Fujimura, T., & Akiyama, H. (2007). A real- time quantitative PCR detection method for pork, chicken, beef, mutton, and horseflesh in foods. Biosci Biotechnol Biochem, 71(12), 3131-3135.

195. Maxstadt J. J, & Pollman M. R. (1980). Spectrophotometric determination of sulfites, benzoates, sorbates, ascorbates, and added blood in ground beef: collaborative study. Journal - Association of Official Analytical Chemists, 63(3), 667-674.

196. Gantner Jr, G., Sturner, W., & Caffrey, P. (1962). Ascorbic acid levels in the post- mortem vitreous humor: their use in the estimation of time of death. Journal of forensic medicine, 9, 156-159.

197. Lawrie, R. (1950). Some observations on factors affecting myoglobin concentrations in muscle. The Journal of Agricultural Science, 40(04), 356-366.

198. Dahlberg, E. (1983). Estimation of the blood contamination of tissue extracts. Analytical Biochemistry, 130(1), 108-113.

199. Rao, Q. C., & Hsieh, Y. H. (2014). Enhanced immunodetection of bovine central nervous tissue using an improved extraction method. Food Control, 46, 282-290.

200. Kohler, G., & Milstein, C. (1975). Continuous cultures of fused cells secreting antibody of predefined specificity. Nature, 256(5517), 495-497.

146

201. Ey, P. L., Prowse, S. J., & Jenkin, C. R. (1978). Isolation of pure IgG1, IgG2a and IgG2b immunoglobulins from mouse serum using protein A-Sepharose. Immunochemistry, 15(7), 429-436.

202. Scientific, T. Extinction coefficients https://tools.thermofisher.com/content/sfs/brochures/TR0006-Extinction-coefficients.pdf. Accessed: March 28, 2017.

203. Laemmli, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature, 227(5259), 680-685.

204. Selvakumar, P., Ling, T. C., Walker, S., & Lyddiatt, A. (2010). Redefinition of working aqueous two-phase systems: A generic description for prediction of the effective phase chemical composition for process control and biorecovery. Journal of Chromatography B-Analytical Technologies in the Biomedical and Life Sciences, 878(21), 1784-1790.

205. Jandaruang, J., Siritapetawee, J., Thumanu, K., Songsiriritthigul, C., Krittanai, C., Daduang, S., Dhiravisit, A., & Thammasirirak, S. (2012). The effects of temperature and pH on secondary structure and antioxidant activity of Crocodylus Siamensis hemoglobin. Protein Journal, 31(1), 43-50.

206. Crystal structure determination of porcine hemoglobin at 1.8A resolution. http://www.rcsb.org/pdb/ngl/ngl.do?pdbid=1QPW&preset=validationReport. Accessed: March 12, 2017.

207. Frieden, E. (1975). Non-covalent interactions: Key to biological flexibility and specificity. Journal of Chemical Education, 52(12), 754-761.

208. Kranen, R. W., Van Kuppevelt, T. H., Goedhart, H. A., Veerkamp, C. H., Lambooy, E., & Veerkamp, J. H. (1999). Hemoglobin and myoglobin content in muscles of broiler chickens. Poultry Science, 78(3), 467-476.

209. Ofori, J. A., & Hsieh, Y. H. (2011). Characterization of a 12 kDa thermal-stable antigenic protein in bovine blood. Journal of Food Science, 76(9), 1250-1256.

210. Yun, Q., He, M. L., Xing, W. C., Bi, J. X., Ma, G. H., & Su, Z. G. (2004). A novel tetrafunctional reagent for simultaneous intramolecular and intermolecular cross-linking of bovine hemoglobin. Biotechnology Letters, 26(17), 1359-1363.

211. Gonzalez, M., Murature, D. A., & Fidelio, G. D. (1995). Thermal stability of human immunoglobulins with sorbitol: Critical evaluation. Vox Sang, 68(1), 1-4.

212. Asenjo, J. A., & Andrews, B. A. (2011). Aqueous two-phase systems for protein separation: A perspective. Journal of Chromatography A, 1218(49), 8826-8835.

213. Asenjo, J. A., & Andrews, B. A. (2012). Aqueous two-phase systems for protein separation: Phase separation and applications. Journal of Chromatography A, 1238, 1-10.

147

214. Olivera-Nappa, A., Lagomarsino, G., Andrews, B. A., & Asenjo, J. A. (2004). Effect of electrostatic energy on partitioning of proteins in aqueous two-phase systems. Journal of Chromatography B-Analytical Technologies in the Biomedical and Life Sciences, 807(1), 81-86.

215. Kan, P., & Lee, C. J. (1994). Application of aqueous two-phase system in separation/purification of stroma free hemoglobin from animal blood. Artificial Cells Blood Substitutes and Immobilization Biotechnology, 22(3), 641-649.

216. Porath, J., & Flodin, P. (1959). Gel filtration: A method for desalting and group separation. Nature, 183(4676), 1657-1659.

217. Xu, L., & Glatz, C. E. (2009). Predicting protein retention time in ion-exchange chromatography based on three-dimensional protein characterization. Journal of Chromatography A, 1216(2), 274-280.

218. Zilberstein, G. V., Baskin, E. M., & Bukshpan, S. (2003). Parallel processing in the isoelectric focusing chip. Electrophoresis, 24(21), 3735-3744.

219. Reverberi, R., & Reverberi, L. (2007). Factors affecting the antigen-antibody reaction. Blood Transfus, 5(4), 227-240.

220. Fremy, J. M., & Usleber, E. (2003). Policy on characterization of antibodies used in immunochemical methods of analysis for mycotoxins and phycotoxins. Journal of AOAC International, 86(4), 868-871.

221. Friguet, B., Djavadiohaniance, L., & Goldberg, M. E. (1984). Some monoclonal- antibodies raised with a native protein bind preferentially to the denatured antigen. Molecular Immunology, 21(7), 673-677.

222. Bobrovnik, S. A. (2003). Determination of antibody affinity by ELISA. Theory. Journal of Biochemical and Biophysical Methods, 57(3), 213-236.

223. Bobrovnik, S. (1998). Determination of antibody affinity using ELISA. Ukrains' kyi biokhimichnyi zhurnal (1999), 71(6), 90-102.

224. Kristinsson, H. G. (2002). Conformational and functional changes of hemoglobin and myosin induced by pH: Functional role in fish quality. Dissertation, University of Massachusetts - Amherst.

225. Kristinsson, H. G., Theodore, A. E., Demir, N., & Ingadottir, B. (2005). A comparative study between acid‐and alkali‐aided processing and surimi processing for the recovery of proteins from channel catfish muscle. Journal of Food Science, 70(4), 298-306.

226. Kristinsson, H. G., & Hultin, H. O. (2004). Changes in trout hemoglobin conformations and solubility after exposure to acid and alkali pH. Journal of Agricultural and Food Chemistry, 52(11), 3633-3643.

148

227. Rieder, R. F. (1970). Hemoglobin stability: Observations on denaturation of normal and abnormal hemoglobins by oxidant dyes, heat, and alkali. Journal of Clinical Investigation, 49(12), 2369-2376.

228. Ivanov, V. T., Karelin, A. A., & Yatskin, O. N. (2005). Generation of peptides by human erythrocytes: Facts and artifacts. Biopolymers, 80(2-3), 332-346.

229. Elmer, J., Harris, D., & Palmer, A. F. (2011). Purification of hemoglobin from red blood cells using tangential flow filtration and immobilized metal ion affinity chromatography. Journal of Chromatography B-Analytical Technologies in the Biomedical and Life Sciences, 879(2), 131-138.

230. Jansson, H., & Swenson, J. (2008). Dynamical changes of hemoglobin and its surrounding water during thermal denaturation as studied by quasielastic neutron scattering and temperature modulated differential scanning calorimetry. Journal of Chemical Physics, 128(24).

231. Carvalho, J. W. P., Carvalho, F. A. O., Santiago, P. S., & Tabak, M. (2013). Thermal denaturation and aggregation of hemoglobin of Glossoscolex paulistus in acid and neutral media. International Journal of Biological Macromolecules, 54, 109-118.

232. Diamandis, E. P., & Christopoulos, T. K. (1996). Immunoassay configurations. In E. P. Diamandis & T. K. Christopoulos (Eds.), Immunoassay, (pp. 227-236): Academic Press, San Diego, CA.

233. Findlay, J. W. A., & Dillard, R. F. (2007). Appropriate calibration curve fitting in ligand binding assays. Aaps Journal, 9(2), E260-E267.

234. Gottschalk, P. G., & Dunn, J. R. (2005). The five-parameter logistic: A characterization and comparison with the four-parameter logistic. Analytical Biochemistry, 343(1), 54-65.

235. Obrien, P. J., Shen, H., Mccutcheon, L. J., Ogrady, M., Byrne, P. J., Ferguson, H. W., Mirsalimi, M. S., Julian, R. J., Sargeant, J. M., Tremblay, R. R. M., & Blackwell, T. E. (1992). Rapid, simple and sensitive microassay for skeletal and cardiac-muscle myoglobin and hemoglobin - use in various animals indicates functional-role of myohemoproteins. Molecular and Cellular Biochemistry, 112(1), 45-52.

236. US Food and Drug Administration. Guidance for industry bioanalytical method validation (2001). http://www.fda.gov/downloads/drugs/guidancecomplianceregulatoryinformation/guidanc es/ucm175984.pdf. Accessed: March 13, 2017.

237. H. Edward Grotjan, & Keel, B. A. (1996). Data interpretation and quality control. In E. P. Diamandis & H. Christensen (Eds.), Immunoassay). E.P.: Academic Press.

238. le Maire, M., Champeil, P., & Moller, J. V. (2000). Interaction of membrane proteins and lipids with solubilizing detergents. Biochimica Et Biophysica Acta-Biomembranes, 1508(1-2), 86-111. 149

239. Zhou, J. Y., Dann, G. P., Shi, T., Wang, L., Gao, X., Su, D., Nicora, C. D., Shukla, A. K., Moore, R. J., Liu, T., Camp, D. G., 2nd, Smith, R. D., & Qian, W. J. (2012). Simple sodium dodecyl sulfate-assisted sample preparation method for LC-MS-based proteomics applications. Analytical Biochemistry, 84(6), 2862-2867.

240. Seddon, A. M., Curnow, P., & Booth, P. J. (2004). Membrane proteins, lipids and detergents: Not just a soap opera. Biochimica et Biophysica Acta - Biomembranes, 1666(1–2), 105-117.

241. Prive, G. G. (2007). Detergents for the stabilization and crystallization of membrane proteins. Methods, 41(4), 388-397.

242. Proc, J. L., Kuzyk, M. A., Hardie, D. B., Yang, J., Smith, D. S., Jackson, A. M., Parker, C. E., & Borchers, C. H. (2010). A quantitative study of the effects of chaotropic agents, surfactants, and solvents on the digestion efficiency of human plasma proteins by trypsin. Journal of Proteome Research, 9(10), 5422-5437.

243. Arachea, B. T., Sun, Z., Potente, N., Malik, R., Isailovic, D., & Viola, R. E. (2012). Detergent selection for enhanced extraction of membrane proteins. Protein Expression and Purification, 86(1), 12-20.

244. Jamur, M. C., & Oliver, C. (2010). Permeabilization of cell membranes. Immunocytochemical Methods and Protocols, Third Edition, 588, 63-66.

245. Goldschmidt, M. C., & Wyss, O. (1967). The role of tris in EDTA toxicity and lysozyme lysis. Journal of General Microbiology, 47(3), 421-431.

246. Blobel, G., & Potter, V. R. (1966). Nuclei from rat liver - isolation method that combines purity with high yield. Science, 154(3757), 1662-1665.

247. Chappell, J. B., & Perry, S. V. (1954). Biochemical and osmotic properties of skeletal muscle mitochondria. Nature, 173(4414), 1094-1095.

248. Mishra, S., & Mishra, R. (2015). Molecular integrity of mitochondria alters by potassium chloride. International Journal of Proteomic, 2015, 647-659.

249. Peng, Z. G., Hidajat, K., & Uddin, M. S. (2005). Selective and sequential adsorption of bovine serum albumin and lysozyme from a binary mixture on nanosized magnetic particles. Journal of Colloid and Interface Science, 281(1), 11-17.

250. Mary, A. R., & Heather, M. L. (1995). Monoclonal antibodies: Production, engineering and clinical application.

251. Qualtiere, L. F., Anderson, A. G., & Meyers, P. (1977). Effects of ionic and nonionic detergents on antigen-antibody reactions. Journal of Immunology, 119(5), 1645-1651.

252. Dimitriadis, G. J. (1979). Effect of detergents on antibody-antigen interaction. Analytical Biochemistry, 98(2), 445-451.

150

253. Barnikol, W. (2005). Patent: Mammalian haemoglobin compatible with blood plasma, cross-linked and conjugated with polyalkylene oxides as artificial medical oxygen carriers, production and use thereof. Patent #: US 6,956,025 B2.

254. Bah, C. S. F., Bekhit, A. E. A., Carne, A., & McConnell, M. A. (2016). Composition and biological activities of slaughterhouse blood from red deer, sheep, pig and cattle. Journal of the Science of Food and Agriculture, 96(1), 79-89.

255. Donnelly, E. B., Delaney, R. A. M., & Hurley, N. (1978). Studies on slaughter animal blood plasma: I. Composition of bovine and porcine plasma. Irish Journal of Food Science and Technology, 2(1), 31-38.

256. Brunisholz, R. A., & Rickli, E. E. (1981). Primary structure of porcine plasminogen. Isolation and characterization of CNBr-fragments and their alignment within the polypeptide chain. European Journal of Biochemistry, 119(1), 15-22.

257. Castellino, F. J., & Ploplis, V. A. (2005). Structure and function of the plasminogen/plasmin system. Journal of Thrombosis and Haemostasis, 93(4), 647-654.

258. Vangelder, W., Huijskesheins, M. I. E., Hukshorn, C. J., Dejeujaspars, C. M. H., Vannoort, W. L., & Vaneijk, H. G. (1995). Isolation, purification and characterization of porcine serum transferrin and hemopexin. Comparative Biochemistry and Physiology B- Biochemistry & Molecular Biology, 111(2), 171-179.

259. Agah, A., Montalto, M. C., Kiesecker, C. L., Morrissey, M., Grover, M., Whoolery, K. L., Rother, R. P., & Stahl, G. L. (2000). Isolation, characterization, and cloning of porcine complement component C7. Journal of Immunology, 165(2), 1059-1065.

260. Hiss, S., Knura-Deszcka, S., Regula, G., Hennies, M., Gymnich, S., Petersen, B., & Sauerwein, H. (2003). Development of an enzyme immuno assay for the determination of porcine haptoglobin in various body fluids: Testing the significance of meat juice measurements for quality monitoring programs. Veterinary Immunology and Immunopathology, 96(1-2), 73-82.

261. Eckersall, P. D., Saini, P. K., & McComb, C. (1996). The acute phase response of acid soluble glycoprotein, alpha(1)-acid glycoprotein, ceruloplasmin, haptoglobin and C- reactive protein, in the pig. Veterinary Immunology and Immunopathology, 51(3-4), 377- 385.

262. Wassell, J. (2000). Haptoglobin: Function and polymorphism. Journal of Clinical Laboratory Analysis, 46(11-12), 547-552.

263. Levy, A. P., Asleh, R., Blum, S., Levy, N. S., Miller-Lotan, R., Kalet-Litman, S., Anbinder, Y., Lache, O., Nakhoul, F. M., Asaf, R., Farbstein, D., Pollak, M., Soloveichik, Y. Z., Strauss, M., Alshiek, J., Livshits, A., Schwartz, A., Awad, H., Jad, K., & Goldenstein, H. (2010). Haptoglobin: Basic and clinical aspects. Antioxid Redox Signal, 12(2), 293-304.

151

264. Goldenstein, H., Levy, N. S., & Levy, A. P. (2012). Haptoglobin genotype and its role in determining heme-iron mediated vascular disease. Pharmacological Research, 66(1), 1-6.

265. Dobryszycka, W., & Krawczyk, E. (1979). Microheterogeneity of mammalian haptoglobins in isoelectric focusing. Comparative Biochemistry and Physiology Part B: Biochemistry and Molecular Biology, 62(1), 111-113.

266. Yang, S. J., & Mao, S. J. (1999). Simple high-performance liquid chromatographic purification procedure for porcine plasma haptoglobin. Journal of Chromatography B- Analytical Technologies in the Biomedical and Life Sciences, 731(2), 395-402.

267. Canaves, S. Water-injected meat: the next Chinese food scandal? (2009). http://blogs.wsj.com/chinarealtime/2009/03/13/water-injected-meat-the-next-chinese- food-scandal/. Accessed: March 12, 2017.

268. Warning to religious groups after traces of pig's blood found in cigarette filters. (2010). http://www.dailymail.co.uk/health/article-1262322/Warning-religious-groups-traces- pigs-blood-cigarette-filters.html. Accessed: March 12, 2017.

269. Matt, M. Concerns grow in Europe over threat from deadly pig virus (2014). http://www.bbc.com/news/science-environment-27256466. Accessed: March 12, 2017.

270. Hillary, E. There is blood and bladders in your wine (2015). https://munchies.vice.com/en/articles/theres-blood-and-bladders-in-your-wine. Accessed: March 13, 2017.

271. Brooke, B. Muslims demand texas pig farmer move so they can build a Mosque (2016). http://toprightnews.com/when-muslims-wanted-to-build-a-mosque-on-his-land-this-pig- farmer-took-action/. Accessed: March 12, 2017.

272. Wells, S. D. Meat industry uses chemical 'meat glue' to trick consumers into eating food scraps (2014). http://www.naturalnews.com/047448_meat_glue_transglutaminase_food_scraps.html. Accessed: March 13, 2017.

273. Daphne D. Ramos, Luz H. Villalobos-Delgado, Enrique A. Cabeza, Irma Caro, Ana Fernández-Diez, & Mateo, J. (2013). Mineral composition of blood sausages: A two case study. In M. Innocenzo (Ed.), Food Industry, (pp. 93-111): InTech, Croatia.

274. Guise, B. Traditional black pudding (2016). http://www.englishbreakfastsociety.com/black-pudding.html. Accessed: March 12, 2017.

275. Farmerie, B. Blood work (2011). http://www.foodarts.com/news/features/17037/blood- work. Accessed: March 12, 2017.

276. Enriqueta, D.-P. (1953). Recipes of the Philippines: ED Perez.

152

277. Semidetko, D. (2014). Traditional Finnish cuisine: organizing the event “days of Finnish and Russian cuisines”. Bachelor, University of Aplied Science.

278. Proglobulin. https://d118ospkkl5uqf.cloudfront.net/Sonac/Sonac%20leaflets%20V2/dar151250_SON_ productleaflet_Progolbulin_LR.pdf. Accessed: March 12, 2017.

279. Lugay, J. C., Haas, G. J., & Beale, R. J. (1978). Patent: Pet food acceptability enhancer. Patent #: 4089978.

280. Porcine hemoglobin powder. http://1004120.en.makepolo.com/products/Porcine- Hemoglobin-Powder-p13215854.html. Accessed: March 12, 2017.

281. Fibrimex. https://d118ospkkl5uqf.cloudfront.net/Sonac/Sonac%20leaflets%20V2/dar150883_SON_ productdatasheet_Fibrimex_V2.pdf. Accessed: March 12, 2017.

282. Hazato, T., & Kase, R. (1986). Isolation of angiotensin-converting enzyme-inhibitor from porcine plasma. Biochemical and Biophysical Research Communications, 139(1), 52-55.

283. Park, E., Won, M., Lee, H. H., & Song, K. B. (1996). Angiotensin converting enzyme inhibitory pentapeptide isolated from supernatant of pig plasma treated by trichloroacetic acid. Biotechnology Techniques, 10(7), 479-480.

284. Levy, O. (2000). Antimicrobial proteins and peptides of blood: Templates for novel antimicrobial agents. Blood, 96(8), 2664-2672.

285. Fred, N., Katarina, S., & Glamsta, E.-L. (2010). Analysis of antimicrobial peptides from porcine neutrophils. Journal of Microbiological Methods, 83(1), 8-12.

286. Wang, J. Z., Zhang, H., Zhang, M., Yao, W. T., Mao, X. Y., & Ren, F. Z. (2008). Antioxidant activity of hydrolysates and peptide fractions of porcine plasma albumin and globulin. Journal of Food Biochemistry, 32(6), 693-707.

287. Lee, S. H., & Song, K. B. (2009). Article isolation of a calcium-binding peptide from enzymatic hydrolysates of porcine blood plasma protein. Journal of the Korean Society for Applied Biological Chemistry, 52(3), 290-294.

288. Lee, S. H., & Song, B. K. (2009). Purification of an iron-binding nona-peptide from hydrolysates of porcine blood plasma protein. Process Biochemistry, 44(3), 378-381.

289. Brantl, V., Gramsch, C., Lottspeich, F., Mertz, R., Jaeger, K. H., & Herz, A. (1986). Novel opioid-peptides derived from hemoglobin - hemorphins. European Journal of Pharmacology, 125(2), 309-310.

290. Leonil, J., Gagnaire, V., Molle, D., Pezennec, S., & Bouhallab, S. (2000). Application of chromatography and mass spectrometry to the characterization of food proteins and derived peptides. Journal of Chromatography A, 881(1-2), 1-21.

153

291. Montiel-Sosa, J. F., Ruiz-Pesini, E., Montoya, J., Roncales, P., Lopez-Perez, M. J., & Perez-Martos, A. (2000). Direct and highly species-specific detection of pork meat and fat in meat products by PCR amplification of mitochondrial DNA. Journal of Agricultural and Food Chemistry, 48(7), 2829-2832.

292. Rodriguez, M. A., Garcia, T., Gonzalez, I., Hernandez, P. E., & Martin, R. (2005). TaqMan real-time PCR for the detection and quantitation of pork in meat mixtures. Meat Science, 70(1), 113-120.

293. China's incredible shrinking hog herd. https://research.rabobank.com/far/en/sectors/animal-protein/Chinas-Incredible-Shrinking- Hog-Herd.html. Accessed: March 12, 2017.

294. Dead-pig tide and the ongoing danger of China epidemics. http://ipolitics.ca/2013/03/11/dead-pig-tide-and-the-ongoing-danger-of-china-epidemics/. Accessed: March 12, 2017.

295. Pork loin roast (skin off). http://meatatbillys.com.au/product/pork-loin-roast-skin-1kg/. Accessed: March 12, 2017.

154

BIOGRAPHICAL SKETCH

EDUCATION 08/2015 - present: M.S in Food and Nutrition, Florida State University, USA. GPA: 3.935/4.0

09/2011 - 06/2015: B.S in Food Quality and Safety, Nanjing Agricultural University, P.R China. GPA: 3.90/4.00 Ranking: 4/72

PROFESSIONAL EXPERIENCE 08/2015 - present: Research Assistant, Department of Human Science, Florida State University.

PUBLICATIONS 1. Wenjun Che, Xingyi Jiang, Kai Qian, Shanshan Sun, Jinlan Wu. 2014. Simultaneous Determination of Three Antibacterial Drugs in Oral Care Products by HPLC. Chemistry Industry (China), 2014(10): 19-21. 2. Haoyuan Shu, Leiqing Pan, Kuier Zhao, Xingyi Jiang, Kang Tu, Jikun Chen. 2015. Advance in Research of Antibacterial Materials in Food Packaging. Food Science (China), 2015, 36(5): 260-265. 3. Xiaoting Yin, Kuier Zhao, Xingyi Jiang, Leiqing Pan, Kang Tu. 2015. Effect of Ultrasonic Treatment Combined with Nano-Packaging on The Quality of Fresh-cut Lettuce. Food Science (China), 2015, 36(2): 250-254 4. Zongzhuan Shen, Beibei Wang, Nana Lv, Yifei Sun, Xingyi Jiang, Rong Li, Yunze Ruan, Qirong Shen. 2015. Effect of The Combination of Bio-organic Fertiliser with Bacillus Amyloliquefaciens NJN-6 on The Control of Banana Fusarium Wilt Disease, Crop Production and Banana Rhizosphere Culturable Microflora. Biocontrol Science and Technology, 25(6): 716-731.

POSTER PRESENTATIONS 1. Han Mu, Xingyi Jiang, Qinchun Rao, College of Human Science, Florida State University, 2017. Characterization of a monoclonal antibody specific to α-livetin, a hen egg allergen. 2. Mustafa Samiwala, Xingyi Jiang, Qinchun Rao, College of Human Science, Florida State University, 2017. Development and characterization of nanoparticle substrates for the detection of food adulterants. 3. Danielle Fuller, Xingyi Jiang, Qinchun Rao, Undergraduate Research Symposium, Florida State University, 2017. Effect of pH on thermostability of porcine hemoglobin. 4. Qinchun Rao, Behnam Keshavarz, Xingyi Jiang, Mustafa Samiwala, Han Mu, Discovery Parade, 2017. Fighting food fraud.

155

5. Xingyi Jiang, Qinchun Rao, College of Human Science, Florida State University, 2016. Identification of a monoclonal antibody which is specific for porcine blood.

ORAL PRESENTATIONS 1. Xingyi Jiang, Qinchun Rao, College of Human Science, Florida State University, 2017. Immunodetection of porcine blood in animal meat.

PROFESSIONAL HONORS AND AWARDS 1. Pao-Sen Chi Memorial Scholarship Endowment, Florida State University, 2017. 2. The first place winner of poster presentation on Research and Creativity Day, College of Human Science, Florida State University, 2016. 3. Jean A. Reutlinger and Lillian H. Munn Scholarship, Florida State University, 2016. 4. PRC National Nutritionist Certificate Holder. 5. Volunteered Quality Safety Officer, 2013-2015, Nanjing China. 6. Nanjing Agricultural University, Scholarships, multiple times from 2011 to 2015. 7. Third Prize of Tomorrow’s Engineer Forum, Nanjing Agricultural University, β01β.

NON-PROFESSIONAL POSITIONS AND AWARDS 1. Vice Minister, Liaison Department, College of Food Science and Technology, Nanjing Agricultural University. 2. Excellent Student Cadre, College of Food Science and Technology, Nanjing Agricultural University. 3. Third Prize of New Oriental Cup College Student Pubic Speaking Contest, Nanjing Area. 4. Honored as one of the Outstanding Students Leaders, Nanjing Agricultural University. 5. Exchange student at UC-Davis funded by Nanjing Agricultural University, 2 of 420.

PROFESSIONAL MEMBERSHIPS 1. Institute of Food Technologists 2. American Chemical Society 3. Chinese American Food Society 4. Glenn Society

156