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 α2 2-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, α2 2) 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