Human Lactoferrin Biopharming in New Zealand

Scientific Risk Assessement

Jack A. Heinemann

Constructive Conversations/Kōrero Whakaaetanga (Phase 2):

Biopharming, Risk Assessment and Regulation A research project funded by the Foundation for Research, Science and Technology

Research Report no. 15

2008

Constructive Conversations/Kōrero Whakaaetanga Rpt 15

2

Constructive Conversations/Kōrero Whakaaetanga Rpt 15 EXECUTIVE SUMMARY ........................................................................................................................5 CHAPTER 1: INTRODUCTION..............................................................................................................7 CHAPTER II: WHAT IS LACTOFERRIN?...........................................................................................9

II.1 ACTIVITIES .........................................................................................................................................9 II.2 STRUCTURE ......................................................................................................................................10

CHAPTER III: WHAT ARE THE INTENDED USES OF LACTOFERRIN PRODUCED IN TRANSGENIC ANIMALS AND PLANTS?..........................................................................................13

III.1 HUMAN MEDICINE ...........................................................................................................................13 III.2 ANIMAL MEDICINE ..........................................................................................................................13 III.3 AGRONOMY.....................................................................................................................................14

CHAPER IV: WHAT ARE THE POTENTIAL HAZARDS? .............................................................15 IV.1 ANTIMICROBIAL PROPERTIES AND ANTIBIOTIC RESISTANCE............................................................15

IV.1.1 Risks arising from the chelating properties of lactoferrin ......................................................16 IV.1.2 Risks arising from the use of rhLF as an antibiotic ................................................................17

IV.2 ANTI-VIRAL EFFECTS ......................................................................................................................18 IV.3 AMYLOIDOSIS .................................................................................................................................19 IV.4 ALLERGENICITY AND AUTOANTIBODIES .........................................................................................20 IV.5 TRANSCRIPTION FACTOR AND GENE ESCORT...................................................................................21

CHAPTER V: CONCLUSIONS .............................................................................................................23 BIBLIOGRAPHY.....................................................................................................................................24

3

Constructive Conversations/Kōrero Whakaaetanga Rpt 15 Abbreviations APHIS Animal and Plant Health Inspection Service Fe Iron Lf Lactoferrin bLf Bovine lactoferrin DNA Deoxyribonucleic acid GEOs Genetically engineered organisms (same as GMOs) GMOs Genetically modified organisms hLf Human lactoferrin kg kilograms LPS Lipopolysaccharides mg milligrams Mg2+ Magnesium PMP Plant-made pharmaceuticals PMIPs Plant-made industrial products rhLf Recombinant human lactoferrin T4SS Type IV secretion system USDA United Stated Department of Agriculture Disclaimer While every effort has been made to ensure that the information herein is accurate, the author does not accept any liability for error of fact or opinion which may be present, nor for the consequences of any decision based on this information. Information contained in this report may be reproduced, providing credit is given. Preliminary Biosafety Evaluation of Biopharming Recominant Human Lactoferrin in New Zealand January 2008 Acknowledgements The author acknowledges C. Amabile-Cuevas, J. Goven, B. Freese and S. Howard for helpful suggestions on this report. Center for Integrated Research in Biosafety University of Canterbury Christchurch New Zealand Ph: (64) (3) 364 2500 Fax: (64) (3) 364 2590

4

Constructive Conversations/Kōrero Whakaaetanga Rpt 15

Executive Summary Animals but especially plants and microbes are a source of a significant proportion of our medicines and industrial compounds. Arguably, their cultivation and preservation for this purpose is nothing new. However, the emergent genetic engineering of plants and animals to be used as recombinant biofactories for the production of therapeutic or industrial chemicals warrants thorough hazard identification and risk evaluation. Unlike the use of transgenic, or genetically modified, microorganisms for this purpose, plants and animals cannot be contained to the same degree at commercial scales of production. This report will focus on the risks to human health and the environment of using transgenic plants and animals as biofactories for recombinant human lactoferrin (rhLf). The economic and social implications are detailed in other reports in this series (Goven et al., 2008, Kaye-Blake et al., 2007). Transgenic organisms including plants and animals are being designed to produce pharmaceuticals, such as recombinant proteins that are secreted in milk. Presently, there are pre-commercial developments of rice plants and cows that can express rhLf. Transgenic sources may be harnessed to reduce costs and the potentials for undesirable contaminants. Transgenic “biofactories” producing pharmaceutical products (e.g., PMPs: plant-made pharmaceuticals) and industrial chemicals (e.g., PMIPs: plant-made industrial products), or foods of altered nutritional value, also can pose special risks to human and animal health. Pharmaceutical compounds have profound physiological effects. Molecules with these attributes are unlikely to act solely in the manner sought by clinicians. Human lactoferrin (hLf) is a protein with several known, distinct functions and others that remain only partially understood. The expression of this compound outside of its normal species and tissue ranges for the lifetime of transgenic organisms occupying different environments, creates a new combination of potentially undesirable outcomes. These will be identified and discussed as far as possible. Human lactoferrin1 is a whey protein found in human milk. It binds iron (Fe) and other metals, such as aluminium, chromium, cobalt, copper, gallium, magnesium and zinc. Fe-binding activity is associated with lactoferrin’s broad-spectrum anti-bacterial effects and is sufficient on its own to explain its activity as a bacteriostatic antibiotic for many bacteria. A derivative of the molecule, called lactoferricin, is bactericidal. That is, it is sufficient to kill certain bacteria through its effects on their membranes. Beyond this, lactoferrin has anti-viral, anti-fungal, immunomodulatory and antioxidant properties, demonstrated activity against some tumours, and other physiological roles. Several structural characteristics of the protein are particularly relevant to a safety assessment. First, the protein has secondary structures that could be important for predicting its potential to aggregate, and form a general cytotoxin (i.e., a compound that kills cells). Second, the protein is normally glycosylated. This affects predictions of its potential to be an allergen. RhLf may be used as a human medicine both in acute infections and as a prophylactic. It may also have uses as a veterinary medicine. Expression in animals and plants may affect microbes that normally inhabit these organisms and those that cause diseases in them. Because rhLf has species-specific effects on bacteria and viruses, its expression at agricultural scales may have implications for the development of resistance to lactoferrin and possibly other important antibiotics (e.g., aminoglycoside antibiotics) in pathogens of plants, animals or humans because of both the scale and

1 Throughout this report, reference to lactoferrin does not distinguish between human or bovine. When the distinction is pertinent, human (hLf) or bovine (bLf) is specified.

5

Constructive Conversations/Kōrero Whakaaetanga Rpt 15 diversity of non-target organisms exposed. In addition, some bacteria that cause disease use lactoferrin to acquire iron from their environments and thus specifically thrive in its presence. Lactoferrin has a tendency to aggregate and form amyloid fibrils. Any protein that makes amyloids may be infectious even if it does not cause known diseases such as those caused by prions (amyloid fibrils of other proteins). The tendency of a protein to form fibrils can be influenced both by its concentration and its biophysical environment. Both of these variables are uniquely affected when a protein is made in a genetically modified organism (GMO). The implications of fibrils are their potential infectious transmissibility and health implications for the GMO. Lactoferrin and especially rhLf may create new risks to people or animals susceptible to developing oral sensitivity to this potential allergen. Formation of autoantibodies to lactoferrin is a common observation in patients suffering from a number of diseases, although the relevance to the diseases or their complications has not been established. The propensity of humans to develop immune responses to bovine and goat milk suggests that the reverse might also be true, and this would be a potential welfare issue for animal biofactories. Finally, the possible effects of rhLf on wildlife that feed on biofactories are unknown. An intriguing property of lactoferrin is that it binds DNA, and transports DNA into cells with lactoferrin receptors. Lactoferrin binds specific DNA sequences that are expected to occur by chance at reasonably high frequencies in mammalian and plant genomes. These sequences also are important for gene expression regulation in the presence of lactoferrin, making lactoferrin a transcription regulator. Bacteria that import lactoferrin may also be able to acquire the DNA to which it is bound, thereby increasing horizontal gene transfer. Some risks of rhLF are unique to, or highly pronounced by, its production in plant or animal biofactories because biofactories increase both the scale of exposure to non-target organisms and the diversity of non-target organisms exposed. To determine the actual implications of plant or animal biofactories of rhLf will require regulatory authorities to recognise each of these potential harm pathways and either evaluate the possibilities of harm management or ask for relevant new research. In sum, the most immediate risk factors of producing rhLf in transgenic plants or animals derive from the protein’s toxic effects on microbes, fungi and viruses, and its potential to elicit an immune response in people when it is derived from GMOs. The potential of rhLF to aggregate into amyloid fibrils and to act as a gene transfer agent are poorly understood risk factors but they are both possibly of profound importance. Long-term studies will be required to better understand these risks.

6

Constructive Conversations/Kōrero Whakaaetanga Rpt 15

Chapter 1: Introduction Animals, but especially plants and microbes, are a source of a significant proportion of our medicines and industrial compounds (Rates, 2001). Arguably, their cultivation and harvest for these compounds is nothing new. Developing, cultivating and extracting medicines and industrial compounds from transgenic plants and animals used as biofactories is an experimental platform, and as such is new. This article will explore the potential production of human lactoferrin in transgenic plants and animals to provide some insight into the dimensions of risk this new platform may create. Transgenic, or genetically modified/engineered (GM/GE), organisms, including plants and animals are being designed to produce pharmaceuticals, such as recombinant proteins that are secreted in the milk of cows and goats (Kues and Niemann, 2004, Ma et al., 2003, Twyman et al., 2003). This use is commonly associated with the organism being a “biofactory” for such material2. For example, plants and animals have been used to produce human antibiodies for experimental use and testing. For this purpose, the native immunoglobulin genes in mice can be replaced with the analogous genes sourced from humans. Following normal developmental patterns for antibody genes, the mice produce “high-affinity antibodies with completely human sequences and differing only in glycosylation patterns” (p. 22 Kim et al., 2005). There are potential advantages to transgenic sources of proteins with medical and research applications. Currently, complex therapeutic molecules cannot be synthesized, or practically synthesized, outside of an organism (be it microbe, plant, fungus or animal). Transgenic sources may be harnessed to provide these molecules. Presently, GM microorganisms or mammalian tissue culture systems that are kept in contained facilities provide a number of drugs. For some compounds, transgenic plants or animals might increase flexibility and reduce costs of production and enhance purity by protecting against undesirable contaminants derived from microbes or mammals (Ma et al., 2003, Twyman et al., 2003). While these are important potential benefits, they are still far from certain. The present extrapolations are based on a few compounds and a limited number of existing production platforms available for comparison. There is no existing confirmation that large scale extraction and highly reliable purification of compounds from GMOs will be as inexpensive as claimed (Goven et al., 2008, Lonnerdal, 2002). GM “biofactories” producing pharmaceutical products (e.g., PMPs: plant-made pharmaceuticals) and industrial chemicals (e.g., PMIPs: plant-made industrial products), or foods of altered nutritional value, also can pose special risks to human and animal health. A case in point is TGN1412, an experimental monoclonal antibody made in transgenic mice. The experimental drug caused catastrophic multisystem failure in the six men, including a New Zealander, who received it in Phase I clinical trials (Schneider et al., 2006). The cause of TGN1412’s toxicity in humans compared with animal models is unknown, and the unfortunate trial has been criticized on procedural grounds. But the point remains that as new drugs become possible using transgenic techniques, so do new hazards. More subtle risks are also possible. By design, pharmaceutical compounds have profound physiological effects. Molecules with these attributes are unlikely to have only one effect, even if the primary activity is what has attracted the attention of clinicians. For example, many drugs designed to treat non-infectious or inborn afflictions also have an antimicrobial activity. This side-effect is associated with concerns about the growing frequencies of antibiotic-resistant bacteria that cause disease in humans and domestic animals (reviewed in Heinemann, 1999, Heinemann et al., 2000).

2 Referring to living organisms, particularly animals, as biofactories is offensive to some. The author openly acknowledges and respects those views and has attempted to make clear that in any shorthand use of that term in this article, there is no attempt to devalue the life of the transgenic organism.

7

Constructive Conversations/Kōrero Whakaaetanga Rpt 15 Moreover, the long-term use of these drugs to treat chronic conditions amplifies their potential effects on the microbial evolution of resistance to clinically-relevant antibiotics. Likewise, their expression throughout the lifetimes of GM plants and animals making compounds with antimicrobial activities could have a similar effect on a large environmental scale. In this report, the development of transgenic sources of human lactoferrin (hLf) is evaluated from a biosafety perspective. At least two relatively large scale developments of recombinant hLf (rhLf) are underway. The first is a rice PMP (Fox, 2006, Marvier and Van Acker, 2005) and the second is a rhLf made in cows (e.g., in-Pharma, 2007, NUTRA, 2007). hLf is a protein of many potential functions, including anti-microbial/viral as well as those that may profoundly influence human physiology. Although the lactoferrins are found in all mammals and are naturally produced by humans, particularly in breast milk, they have species and tissue-specific activities and effects. The risk profile for the protein is made more complex by species and tissue-specific differences that might occur during its synthesis in transgenic organisms.

8

Constructive Conversations/Kōrero Whakaaetanga Rpt 15

Chapter II: What is lactoferrin? Lactoferrin is a member of the transferrin family, which is approximately 300-500 million years old (Teng, 2002). It is a single-chain protein of approximately 690 amino acids and a molecular weight of 77kDa (Nuijens et al., 1996). Lf is a whey protein found in the milk and other biological fluids (including tears, pancreatic juice, mucous, saliva, bile, genital and nasal secretions and in circulating neutrophils) and other tissues of mammals (Bellamy et al., 1992, Lactoferrin Research, 2007, Teng, 2002). Human lactoferrin is present in human milk at up to 20-times the level of bovine lactoferrin (bLf) found in bovine milk, although this can change during bacterial infections (Hyvonen et al., 2006, Nuijens et al., 1996, Teng, 2002). In humans and mice but not cows, expression of the lactoferrin gene might be highly responsive to steroids (Teng, 2002). “Lactoferrin is expressed both ubiquitously and in a species-, tissue-, and cell-type-specific manner. Multiple signaling pathways are involved in the regulation of lactoferrin gene expression” (p. 13 Teng, 2002). Implications of species and tissue differences in lactoferrin concentration will be discussed in Section 3.

II.1 Activities Lactoferrin has multiple activities, for which there are varying amounts of characterising evidence (APHIS, 2007, Harper, 2004, Lactoferrin Research, 2007, Oh et al., 2001). Many of these activities have fostered interest in further developing lactoferrin as a therapeutic agent and thus in increasing capacity for production while reducing the costs of production. Likewise, because the protein can have potent activities, both the concentration of the protein and the context in which its activities are expressed can create potential hazards (Freese, 2007). Like other transferrins, lactoferrin binds iron (Fe). Lactoferrin has two metal binding sites for Fe3+ (Teng, 2002). The holo- and apo-forms are distinguished by whether the protein is or is not saturated with Fe, respectively. Lactoferrin also binds other metals, such as aluminium, chromium, cobalt, copper, gallium, magnesium (Mg2+) and zinc (Nuijens et al., 1996). Surface-exposed histidine residues in the protein also allow it to bind metals with low affinity. This activity is associated with a tendency for hLf to form polymers under low ionic strength conditions, and an ability to hydrolyze nucleic acids (Nuijens et al., 1996). Fe-binding activity is associated with lactoferrin’s broad-spectrum anti-bacterial effects and is sufficient on its own to explain its activity as a bacteriostatic antibiotic3 for many bacteria (Nuijens et al., 1996). In human milk, only about 8% of the hLf is saturated with iron, ensuring that free iron levels remain limiting for the growth of some kinds of bacteria (Nuijens et al., 1996). Some pathogenic bacteria, however, have receptors for lactoferrin and can exploit it as a substitute siderophore to acquire iron for themselves (Freese, 2007, Nuijens et al., 1996). This property of lactoferrin could have implications for the health and welfare of transgenic plants and animals that express rhLf at high concentrations throughout their lives. While bacteriostatic agents cause reversible inhibition of growth, lactoferrin also can be bactericidal. A derivative of the molecule, called lactoferricin (Zasloff, 2002), is sufficient to kill bacteria through actions on their membranes (Bellamy et al., 1992, Nuijens et al., 1996). Lactoferrin binds to lipopolysaccharides (LPS) (Nuijens et al., 1996), which are unique to Gram-negative bacteria, and either lactoferrin itself or its lactoferricin derivative then disrupts the membrane (Hancock and Chapple, 1999, Hyvonen et al., 2006, Zasloff, 2002).

3 Antibiotics are compounds produced by living organisms that kill or inhibit bacteria, and are a subclass of all compounds with antimicrobial activity.

9

Constructive Conversations/Kōrero Whakaaetanga Rpt 15 Lactoferrin also has anti-fungal and anti-viral properties (Nuijens et al., 1996). Although there is little direct evidence of how lactoferrin affects these pathogens (Lonnerdal, 2003), studies on the effects of viruses suggest that lactoferrin inhibits transmission, probably by interfering with receptor recognition (Groot et al., 2005, Nuijens et al., 1996, Viani et al., 1999). Beyond this, lactoferrin has immunomodulatory and antioxidant properties, demonstrated activity against some tumors, and other physiological roles (Fischer et al., 2006, Nuijens et al., 1996, Teng, 2002). As an immunomodulatory protein, lactoferrin contributes to the Th1/Th2 balance (Fischer et al., 2006). The Th cells produce cytokines and the subtypes (e.g. Th1, Th2) are distinguished by the pattern of cytokines they produce. Balance between these Th cell types is important for maintaining normal function. Allergic responses, for example, are thought to arise from an excessive Th cell response (Fischer et al., 2006). Lactoferrin’s ability to sequester free Fe could decrease the production of oxygen radicals through an Fe-catalyzed reaction and may account for observations that lactoferrin limits cell damage at the site of inflammation (Nuijens et al., 1996). The LPS-binding activity of lactoferrin also may compete with LPS-binding protein (LBP) which normally transfers LPS to CD14 (Elass-Rochard et al., 1998). CD14 is a protein on the surface of monocytes and macrophages. LPS-CD14 complexes initiate a signaling pathway that culminates in an inflammatory response (Hancock and Scott, 2000). However, Fe bound to surface histidines of lactoferrin may accentuate free radical formation under some conditions (Nuijens et al., 1996). Among its other physiological roles, hLf can be found at high concentrations as a coat protein of sperm and is differentially expressed during the estrus cycle, suggesting that the protein may have an important role in reproductive physiology (Teng, 2002). Finally, one of the more intriguing roles of lactoferrin may be as a transcription factor (He and Furmanski, 1995). Lactoferrin binds to specific DNA sequences and can be transported to the nucleus as part of a DNA-protein complex. “These observations suggest that lactoferrin released from neutrophils at inflammation sites may be taken up by other immune cells, transported to the nucleus, and bound to DNA, with consequent regulation of gene expression” (p. 290 Nuijens et al., 1996).

II.2 Structure Protein structure is defined at four levels: primary, secondary, tertiary and quaternary. Beyond this are post-translational modifications to proteins (Box 1). The primary and the secondary structures, and post-translational modifications, are pertinent to a biosafety assessment of recombinant human lactoferrin (rhLf). The primary structure of lactoferrin is determined by the order of amino acids of which it is composed. By comparison of primary structure, bLf and hLf are 68% identical. This is significant, because bLf is an allergen in some people that react to milk (Wal, 2001). Lactoferrin is also known to aggregate under some conditions (Buxbaum, 2003). When proteins unfold, they may re-fold into alternative forms and clump together with other like proteins in what are termed aggregates. Even normally benign proteins can be toxic when aggregated (Bucciantini et al., 2002, Ellis and Pinheiro, 2002). Lactoferrin takes on a distinct 3-dimensional structure when it folds into its native conformation. This conformation has secondary structural elements such as α-helices and β-pleated sheets (Nuijens et al., 1996). Re-folding that converts α-helices into β-sheets results in

10

Constructive Conversations/Kōrero Whakaaetanga Rpt 15 the formation of an aggregates structure called an amyloid fibril which can significantly alter both biophysical and biological properties of proteins. Fibril formation is of significance to biosafety because the environment is an important determinant of whether a protein aggregates. Therefore, recombinant proteins should be evaluated for their aggregation potential in the context of the new cellular environments in which they are found, and at the concentrations that may be achieved through the use of recombinant DNA techniques. There are no verified bioinformatics techniques to predict aggregation, so the protein must be studied in, or isolated from, the recombinant organism itself. Box 1: Post-translational modifications and glycoproteins The primary structure of proteins is the linear order of amino acids which compose them. Proteins are more than just a sequence of amino acids, however. They may also be modified by addition of different kinds of molecules to various amino acids. This is relevant to risk assessment because the modifications can alter protein structure and function, as well as change the potential for the protein to be a toxin or allergen. There are many forms of post-translational modifications. Most are the addition of molecules, but some modifications result from removing amino acids or re-folding a protein into an alternative three dimensional structure. The range of potential post-translational modifications varies by species, tissue and stage of development (Gomord et al., 2005). Modification has medical and food relevance because, for example, proteins modified in plants can be immunogenic in humans (e.g. Prescott et al., 2005) and may cause cross-reactivity to similar epitopes (i.e., immunogenic regions) that occur in proteins from animal sources. The same protein can exist in hundreds to thousands of different isoforms in the same cell at the same time, but each form may not exist at the same concentration. Thus, detecting different forms can be very difficult. More than 300 different types of chemical modifications are known, and are distributed among the following types: ubiquitination, halogenation, phosphorylation, farnesylation, glycosylation, glycoxidation, acetylation and methylation (Manzi et al., 2000, Zasloff, 2002). Over half of all proteins are glycosylated (Van den Steen et al., 1998). No modification is exclusive, so multiple isoforms of the “same” protein, distinguished by different combinations of modifications and groups of modifications, can co-exist (Lane and Beese, 2006, Norregaard Jensen, 2004). Glycoforms of a protein are sugar-modified variants of the same primary amino acid polymer. The three main post-translational protein modifications that use sugars are N- and O-linked glycosylation and glycosyl phosphatidyl inositol anchors (Van den Steen et al., 1998). Linkage to the polypeptide is made at serine, threonine and hydroxylysine amino acids (O-linked) or via the amide nitrogen of asparagine (N-linked) (Bardor et al., 1999, Mitra et al., 2006). The predominant and perhaps sole physiological form of lactoferrin is a glycoprotein (Box 1), but many different glycoforms (or isoforms) can co-exist (Nuijens et al., 1996). Depending on species, the dominant glycoform can be different but is always of the N- rather than O-linked type. For example, the N-acetyllactosaminic glycan is found in humans and mice but the N-oligomannosidic glycan is found in cows (Nuijens et al., 1996). rhLf produced in rice is glycosylated differently to that produced in humans, notably by the substitution of mannose and xylose carbohydrates on terminal modifications in rhLF for fucose or sialic acid in hLf (Lonnerdal, 2002). Glycosylation of rhLf also differs when it is expressed in different kinds of plants (Fujiyama et al., 2004). Some researchers have not identified activity differences between particular rhLF and hLf among their tests (e.g., Lonnerdal, 2002), but others report that rhLf and hLf show differences at least in metal binding (Nuijens et al., 1996). Moreover, glycosylated and unglycosylated forms have different susceptibilities to proteases and thus their relative abundance may affect their balance of immunomodulation and antimicrobial activity (Nuijens et al., 1996).

11

Constructive Conversations/Kōrero Whakaaetanga Rpt 15 Despite significant similarities at the genomic level (Nuijens et al., 1996, Teng, 2002), and highly conserved primary structures of the lactoferrin family of proteins (Teng, 2002), lactoferrins from different species or with different post-translational modifications can have different physiological functions, as discussed above. This demonstrates that seemingly minor structural differences can have important implications, either for the recipient of the protein as a therapeutic or for the plant or animal producing rhLf throughout its life. Finally, while a description of the physiological effects of lactoferrin expressed endogenously, although incomplete, is at least underway, knowledge of unintended environmental effects of the protein is almost completely lacking due to both a shortage of research and the large range of activities that would need to be monitored. Dietary lactoferrin is not completely degraded during digestion and the protease-generated smaller derivative peptides have significant biological activities (Nuijens et al., 1996). These activities will need to be considered in assessments of the impacts of predation on recombinant plants and animals.

12

Constructive Conversations/Kōrero Whakaaetanga Rpt 15

Chapter III: What are the intended uses of lactoferrin produced in transgenic animals and plants?

The variety of activities of lactoferrin have encouraged commercialisation based either on a novel application or a novel production platform (Table 1). The arguments behind sourcing this molecule from transgenic plants and animals come from expectations that GM plant and animal sources will make the costs of production commercially attractive (Crosby, 2003, Hancock and Scott, 2000, Sparrow et al., 2007), and possibly contribute to product safety (discussed in detail in Section 3).

III.1 Human medicine Lactoferrin’s combined anti-bacterial and anti-viral activities suggest it as a promising anti-diarrhoeal agent, because this condition can be caused by either bacteria or viruses. The development of rhLf as an anti-diarrheal has attracted criticism on a number of levels. Safety issues will be discussed in Section 3. A criticism not based on the safety of rhLf for those consuming it concerns the displacement by rhLf-based strategies for potentially more effective, sustainable and locally empowering strategies to address the problem of diarrhea, particularly in developing countries (FOE, 2007, Freese, 2007). Lactoferrin is also being tested as a therapeutic for conditions not caused by infection. Those suffering from diseases that increase gut permeability, such as Crohn’s disease, or from infections that result in transient increases in gut permeability, are prone to absorbing microorganisms, toxins and allergens in higher amounts (MacDonald and Monteleone, 2005). To determine whether rhLf could antagonise the activities increasing gut permeability, healthy volunteers were treated with indomethacin, a drug that increases the permeability of the small intestine. A significant biological effect on permeability was observed, which was protective against the indomethacin-induced enteropathy, but the mechanism of action is unknown (Troost et al., 2003). The use of rhLf for chronic conditions would likely create a demand for high levels of production. Table 1: actual and potential lactoferrin bio-pharms Company Platform Intended uses Meristem Therapeutics corn Treat gastrointestinal disorders

(Fox, 2006) Ventria Bioscience rice and barley Treat gastrointestinal disorders

(Fox, 2006, Marvier and Van Acker, 2005)

grass carp Viral resistance (Kapuscinski, 2005)

AgResearch bovine, goat Therapeutic (Anon, 2007, in-Pharma, 2007)

III.2 Animal medicine Veterinary applications are also envisioned. For instance, in a feeding trial using Asian catfish (Clarias batrachus), researchers found that bLf provided in feed at doses of 100 mg/kg enhanced non-specific immunity and disease resistance (Kumari et al., 2003). A study in China found that grass carp (Ctenopharyngodon idellus) made transgenic for rhLf took longer to display symptoms, and had a higher survival rate, than non-transgenic fish when infected with the Grass Carp Haemorrhage Virus (Zhong et al., 2002). rhLf, and possibly up-regulation of bLf (using transgenic techniques), are also

13

Constructive Conversations/Kōrero Whakaaetanga Rpt 15 being suggested for preventing or treating mastitis in cows (Hyvonen et al., 2006). In one such study on efficacy, transgenic cows producing rhLf were not completely protected from initial infection with a mastitis-causing strain of Escherichia coli, but symptoms were milder (Hyvonen et al., 2006).

III.3 Agronomy Recombinant hLf may also have agronomic qualities. Transgenic tobacco plants expressing rhLf were found to be more resistant to the wilt-causing bacterium Ralstonia solanacearum (Mourgues et al., 1998, Zhang et al., 1998). Recombinant bLf was used to chelate iron in pear plant tissues and confer some level of protection against the blight-causing bacterium Erwinia amylovara (Malnoy et al., 2003).

14

Constructive Conversations/Kōrero Whakaaetanga Rpt 15

Chaper IV: What are the potential hazards? Those involved in developing rhLf have suggested that it may have a number of medical and economic benefits. The realization of these benefits will depend upon inter alia the demonstration of the efficacy of rhLf in its proposed uses. Although potential benefits will be mentioned in this report, a critical analysis of efficacy will not be undertaken. This is because efficacy is difficult to predict without knowledge of the final product and its formulation, method of manufacture and distribution. A proper efficacy assessment therefore requires access to information that either as yet does not exist or is proprietary. This report emphasises the identification of potential hazards that may extend to the broader public. This is called risk forecasting, an exercise at the cutting edge of the research literature that serves to forewarn of risk where the science is not certain. What follows is a critical analysis of identified potential hazards that are specific to lactoferrin. They derive from both the activities and the structure of the molecule (described in Section I). RhLF is a member of a class of potential therapeutics and thus carries with it some generic risks as well. These risks are those that arise from unknown, unintended or unanticipated properties of the molecule, the molecule synthesised in a new organism, or the molecule present in a novel context. An example of this type of risk is illustrated by the effect of drugs not designed or intended for use as antimicrobial agents but which have antimicrobial activities and thus may contribute to the rise in number of infections caused by antibiotic-resistant bacteria (Box 2). To the extent possible, these generic risks will be addressed below. A general note of caution is warranted for this protein which has so many activities, few of which have been comprehensively characterised either biochemically or in the particular physiological context of different plants and animals and humans at different stages of their lives. There are indications from the literature that one application of rhLf would be to supplement bLf in infant formula (Lonnerdal, 2002). The United States National Academy of Science illustrated the validity of a precautionary stance especially on proteins that might be part of specialized foods, e.g., infant formula, or could contaminate these foods (NAS, 2004).

IV.1 Antimicrobial properties and antibiotic resistance Lactoferrin has antibacterial properties. The first of these properties arises from its ability to chelate Fe3+, and thus prevent some bacteria from obtaining sufficient quantities of iron necessary to thrive. Iron starvation is generally bacteriostatic; while the bacteria cannot reproduce, if the body or other microbes have not been able to clear the bacteria, they can be revived should they acquire iron. That lactoferrin has a bacteriostatic effect on some bacteria is not itself a problem. A number of clinical antibiotics are bacteriostatic and highly effective in people and animals with normal immune functions. The second property is also bacteriostatic. Lactoferrin has been associated with the inhibition of aggregation of some enteric pathogens, such as enteraggregative E. coli (Ochoa et al., 2006). Lactoferrin disrupts the type IV secretion system (T4SS), interfering with the export of certain proteins necessary for virulence and for uptake of the bacteria into human cells (Ferguson and Heinemann, 2002). However, the effect on virulence was not due to disruption of the T4SS per se as shown by the fact that even bacteria that do not rely on T4SS were inhibited by lactoferrin (Ochoa et al., 2006). The third of these properties is bactericidal. Inherent within the protein, or possibly released from the protein, is a smaller fragment that acts as a protein antibiotic. Digestion of hLf with a protease called

15

Constructive Conversations/Kōrero Whakaaetanga Rpt 15 pepsin produces a 47-amino-acid peptide called lactoferricin H; an analogous though smaller peptide is produced from bLf and is called lactoferricin B (Nuijens et al., 1996). Peptide antibiotics are common in animals of all kinds, but they do not all have the same mode of action (Hancock and Chapple, 1999). Thus, the production of rhLf may create no qualitative novelty if it has the same range of properties as either hLf or other antibiotic peptides. However, rhLf would by definition quantitatively extend the range of exposures to non-target or non-pertinent organisms and increase the concentrations of lactoferrin in some environments. These aspects would remain relevant to a safety assessment. Box 2: Generic issues of compounds toxic to microbes but not called clinical antibiotics The use of rhLf as a supplement in normal human foods such as yogurt, performance drinks and infant formula (Lonnerdal, 2003, USDA, 2004) or animal feeds (Humphrey et al., 2002), as has been proposed by both the industry and researchers, could significantly increase the duration and scale of lactoferrin in the environment. In this respect, these applications should be considered with reference to lessons learned from other therapeutic or antimicrobial agents. Even psychotherapeutic, anesthetic, antihypertensive, diuretic, and antihistaminic drugs, designed for clinical treatment of non-pathogen induced ailments, can have antimicrobial activities (Dastidar et al., 1986, Kristiansen, 1991, Kristiansen et al., 1989, Ray et al., 1980). The use of these drugs may promote the spread of resistance to conventional antibiotics. Bacteria expressing resistance to an antihistaminic drug, ambodryl, and a tranquillizer, promazine, for example, simultaneously demonstrated cross-resistance to penicillin, streptomycin, chloramphenicol, tetracycline, kanamycin and some combination of these drugs (Ray et al., 1980). Therapeutics for noninfectious diseases are often designed for extended use, maintaining long-term selection for microbial resistance and the opportunity for the evolution of multi-step, high level resistance in treated patients. Likewise, these drugs would be expressed throughout the lives of transgenic biofactories, at potentially large environmental scales, and thus create many times the selective pressure for antibiotic resistance. Unfortunately, any effect antibiotics, such as rhLf, have on the immune system will complicate resistance control. Immunosuppressive effects of antimicrobials may inhibit immune clearance of microbes thereby promoting the survival of resistant microbes. Immuno-potentiating effects, although not as intuitively contrary to our interests, will similarly be detrimental when combined with invasive procedures like organ replacements. Treatments for noninfectious illnesses also expose more non- or opportunistically-infectious flora to antimicrobial agents and may thereby select reservoirs of resistance determinants. Inadvertent selection for resistance to noninfectious-disease therapeutics predisposes infectious microorganisms to life in the presence of antimicrobial agents (Heinemann et al., 2000). IV.1.1 Risks arising from the chelating properties of lactoferrin Some bacteria can use the iron chelating property of lactoferrin to their own advantage. “Lactoferrin does not inhibit the growth of all iron requiring microorganisms; some pathogens of mucosal surfaces, such as Helicobacter pylori, Neisseria sp., Treponema sp., and Shigella sp. have receptors for lactoferrin (sometimes species specific) and acquire iron directly therefrom allowing their growth and pathogenicity in adverse host environments” (p. 287 Nuijens et al., 1996). This list extends to Bordetella pertussis (Agiato and Dyer, 1992), Gardnerella vaginalis (Jarosik and Land, 2000), Moraxella catarrhalis (Ekins et al., 2004) and Moraxella bovis (Yu and Schryvers, 2002). Among these bacteria are significant human and domestic animal pathogens, with Shigella strains being among some of the most virulent gastrointestinal pathogens (O'Ryan et al., 2005, Weir, 2002). Humans are the most significant reservoir for Shigella (Weir, 2002), however, it can persist for

16

Constructive Conversations/Kōrero Whakaaetanga Rpt 15 extended periods on plants that have been contaminated with human waste (Brandl, 2006) and thus potentially inoculate grazing animals. Regardless of current assessments of pathogenicity and reservoir, any species of bacteria can give rise to strains that are pathogenic, and any bacterium can be an opportunistic pathogen under the right conditions or in the right host, such as an immuno-compromised patient. For example, Burkholderia gladioli is a species that was first described in 1921 for its ability to cause a disease on flowers of the iris family (Bauernfeind et al., 1998). It would be nearly another 70 years before it was realized that this plant pathogen caused pulmonary infections in humans, and contributed to the mortality and morbidity of cystic fibrosis (Baxter et al., 1997). The distribution of microorganisms varies in different environments for a variety of reasons. The long-term expression of rhLf in these environments would be a unique factor that could influence microbial community dynamics because of the persistence of bacterial exposure, concentration of active molecule, and range of microorganisms exposed. These are all factors contributing to the emergence of antibiotic resistance in human pathogens (Amábile-Cuevas et al., 1995, de la Cruz et al., 2002, Heinemann, 1999, Heinemann et al., 2000, Levy, 1998, Swartz, 1994). This has been a guiding lesson from the introduction of other antibiotics into clinical use. The species and strains of bacteria dominating nocosomial infections in any decade are those that happened to be intrinsically less susceptible to the dominating antibiotic(s) in use in that decade (Emori and Gaynes, 1993, Swartz, 1994). If these results extend to other ecosystems, then it is reasonable to speculate that those bacteria least susceptible to, or that directly benefit from, rhLf, will invade or increase in presence within the recombinant organisms or the surrounding environment. A full assessment of the risks of expressing rhLf in recombinant organisms should therefore extend to a survey of the species of bacteria inhabiting the environments in which the recombinant would be found, the potential for species with a competitive advantage to invade or increase in number in these ecosystems, and the impact on human health and the environment if such a change were to develop. In addition, lactoferrin can bind other metal ions. Like iron, these metals may be vital co-factors for enzymes. Whether the ability of rhLf to sequester these other metal ions is of physiological relevance would require testing in the particular recombinant organism in which rhLf is produced. IV.1.2 Risks arising from the use of rhLF as an antibiotic The use of rhLf can be expected to release more lactoferricin H activity either into the recombinant organisms used as biofactories, the soil or other substrate exposed to the recombinant organism (e.g. waterways), predators of the recombinant organism, or into products meant to contain rhLf. As with other antibiotics, there is concern that the scale or range of novel environments in which rhLf would be present under commercial production might result in increasing the chances of resistance emerging within bacteria that cause disease, or resistance passing by horizontal gene transfer to bacteria that cause disease. While resistance to peptide antibiotics is reportedly rare, that is no guarantee that the use of rhLf will not cause resistance to emerge among important pathogens. Some argue that “[a]ntimicrobial peptides target a previously under-appreciated `microbial Achilles heel', a design feature of the microbial cellular membrane that distinguishes broad species of microbes from multicellular plants and animals” (p. 389 Zasloff, 2002), and this is reason to expect that peptide antibiotics, such as rhLf, will remain efficacious. Unfortunately, the same could have been said for commercially-produced antibiotics to which microbes have become resistant. For example, Streptococcus pneumoniae should never have evolved resistance to penicillin because the number of mutations necessary was a probabilistic impossibility (Heinemann and Traavik, 2004). Nevertheless, a significant proportion of these bacteria are now resistant to high levels of penicillin (e.g. Sahm et al., 2000).

17

Constructive Conversations/Kōrero Whakaaetanga Rpt 15 Caution is required when considering the use of rhLF as an antibiotic to avoid the same mistakes made with other antibiotics. Changes to the charge structure of the bacterial surface may be enough to impart significant tolerance to lactoferricin. “Bacteria have found ways to modulate their anionic cell wall polymers such as peptidoglycan, lipopolysaccharide, teichoic acid or phospholipids by introducing positively charged groups. Two of these mechanisms involving the transfer of D-alanine into teichoic acids and of L-lysine into phospholipids, respectively, have been identified and characterized in Staphylococcus aureus, a major human pathogen in community- and hospital-acquired infections” (p. 643 Weidenbaier et al., 2003). Clinical antibiotics also have ancient roots, but clinically-relevant resistance has only arisen since the large-scale use of these chemicals in medicine and agriculture (Collignon et al., 2005, Furuya and Lowy, 2005). The diversity of antimicrobial peptides, and their species-specific distribution, dilutes the frequency of exposure to individual peptides. This is not to dismiss the possibility that antimicrobial peptides will have a longer useful life in the clinic than other antibiotics might have, but rather to suggest that we should be cautious about concluding, based on the current lack of observed resistance, that commercial-scale production of rhLf will not induce resistance. This is especially true as pressure mounts to use lactoferrin in agricultural applications (e.g. Humphrey et al., 2002, Kumari et al., 2003). Indeed, the issues are much more complex than they initially appear. First, some microorganisms are intrinsically less susceptible to lactoferricins than are other bacteria. “Interestingly, the enteric bacteria Pseudomonas fluorescens, Enterococcus faecalis, and Bifidobacteria sp. were highly resistant to lactoferricin, with growth stimulated in the latter by native lactoferrin” (p. 288 Nuijens et al., 1996). S. aureus growth was not inhibited, possibly even was enhanced, by rhLf produced in potatoes (Chong and Langridge, 2000). As discussed above, the use of lactoferricin in new environments and organisms could create a niche for those microorganisms that are inherently more fit in its presence, and no microorganism can be discounted as a pathogen either to the plant or animal being used as a biofactory or to humans. Second, there is evidence that lactoferrin can induce resistance in bacteria to other, unrelated antibiotics. “Based on our results, we hypothesize that [rhLf] could be a host factor that decreases the efficacy of tobramycin in vivo” (p. 1615 Andres et al., 2005). This form of resistance has a limited persistence; however, it can last long enough to be of clinical relevance. By chelating cations, particularly Mg2+, lactoferrin can cause a change in the chemical composition of the outer membrane of some kinds of bacteria, and cause an associated change in the electrostatic potential required for antibiotic uptake. This physiological change results in a phenotypic but not genetic resistance to aminoglycoside antibiotics, that may nonetheless last for many generations (for a review, see Heinemann, 1999). Transient physiological resistance is gaining attention as an important form of resistance among disease-causing bacteria that is under-reported due to apparent susceptibility of the strains after they are isolated from a patient, cultured, and then tested for resistance to agents that failed in vivo (Wiuff et al., 2005). The use of recombinant biofactories therefore may be expected to increase the general environmental level of physiological adaptation to some front-line antibiotics, including some of last resort in human medicine. More directly, it may raise the level of effective resistance in microbes inhabiting the biofactories and could possibly reduce options for treating infections.

IV.2 Anti-viral effects The effect of lactoferrin on viruses is not as well understood. Both hLf and bLf inhibit some viruses, probably by interfering with the processes by which a virus identifies its target cell and gains entry into it (Arnold et al., 2002, Drobni et al., 2004, Nuijens et al., 1996, Swart et al., 1999). Lactoferrin’s

18

Constructive Conversations/Kōrero Whakaaetanga Rpt 15 effects have been documented on herpesvirus, adenovirus, rotavirus, poliovirus HIV, human cytomegalovirus, human papillomavirus and others (Arnold et al., 2002, Drobni et al., 2004). Since the focus of hLf research will traditionally have been viruses that infect humans, its use in other organisms, which would expose hLf to many different viruses, goes well beyond research coverage. Thus, it is impossible at present to fully describe the broader implications of rhLf outside of expression in people. The first environmental implication will be the possibility that viruses will evolve a resistance to the effects of lactoferrin. The concern here would be for those viruses that primarily infect non-human hosts to expand their range to include humans, because one natural barrier, lactoferrin, is no longer effective against them. The infectious range of viruses is well known to be much larger than the range of organisms in which they cause symptoms (for a recent review, see Heinemann, 2007). So opportunity and access may not be the critical barrier for viruses to switch to new hosts. Rather, intrinsic factors—such as receptor recognition—may prove to be important. The expression of rhLf in plants may not be an automatic solution. Viruses can change preference between plants and animals (references in Heinemann, 2007). A second environmental implication emerges from gene flow, particularly from transgenic plants. Transgenes, such as the one for rhLf, may pass to related plants by gene flow (Heinemann, 2007). Escape of the transgene by this path could significantly increase the magnitude of environmental exposure. Suppose, for instance, that rhLf inhibited a normal viral pathogen of a weed that received the transgene for rhLf. Release from this selective pressure could drive the spread of the recombinant weed both in and outside of the agro-ecosystem.

IV.3 Amyloidosis Lactoferrin aggregates are associated with several debilitating conditions. Lactoferrin amyloids are found as deposits in familial corneal amyloidosis (also known as gelatinous drop-like corneal dystrophy) (Stefani, 2004), and deposits are also found in seminal vesicles and the brain (Nilsson and Dobson, 2003). The latter was seen in Alzheimer’s disease, Down’s syndrome, Pick’s disease and amyotrophic lateral sclerosis patients (Nilsson and Dobson, 2003). Amyloid deposition in seminal vesicles is common among elderly males [21% of men that are more than 75 years of age have detectable deposits]” (p. 375 Nilsson and Dobson, 2003). However, there are inconsistencies in the reports linking lactoferrin and amyloidosis of the seminal vesicles (Linke et al., 2005), and the fibrils are of unknown clinical relevance in any case (Stefani, 2004). The inconsistency in reports may be explained if the fibrils were derived from fragments of lactoferrin that can aggregate (Nilsson and Dobson, 2003). Amyloids form when the α-helices of a normally folded protein convert into β-sheets. This converts a soluble form of the protein into an insoluble deposition (Buxbaum, 2003). Diseases associated or caused by aggregation of proteins into amyloid fibrils may be infectious (Sigurdson et al., 2002). Even normally benign proteins can be toxic when misfolded (Bucciantini et al., 2002, Ellis and Pinheiro, 2002). Toxic proteins can be produced by truncated mRNAs (Wilusz et al., 2001), themselves often the product of inserting recombinant DNA into a new genome. The conditions that promote amyloid formation are not fully known. Attempts to predict when a protein may form an amyloid are promising but still in their infancy (Fernandez-Escamilla et al., 2004, Pawar et al., 2005). What is clear is that the propensity to form amyloids (or other kinds of aggregates) is determined by biophysical properties of the immediate environment (Bucciantini et al.,

19

Constructive Conversations/Kōrero Whakaaetanga Rpt 15 2002, Gast et al., 2003, Pawar et al., 2005) and concentration of the protein (Buxbaum, 2003, Tartaglia et al., 2007), both of which are altered when proteins are produced in GMOs. The formation of aggregates is a long-term issue of particular relevance to the health and welfare of the animal producing rhLf, such as a cow, rather than an issue of transmissibility to humans via purified rhLf, or through food. However, food transmission cannot be completely ruled out considering that this is a route of transmission for the prion proteins, a class of amyloid fibrils (Caughey and Baron, 2006).

IV.4 Allergenicity and autoantibodies Cow’s milk causes a significant number of cases of food intolerance and allergic reactions in people (Schmidt et al., 1995, Woodford, 2007). The major proteins are β-lactoglobulin (BLG), α-lactoalbumin (ALA), bovine serum albumin (BSA), and the four proteins of the casein fraction (CAS). Allergens, molecules that induce an immune response leading to an allergic response, include the major proteins but minor proteins of cow’s milk, particularly lactoferrin and immunglobulins (Ig), cause a disproportionately high response in people, with “35–50% of patients [being] sensitized to those proteins and sometimes to those proteins only” (p. 35 Wal, 2001). The risk of oral sensitisation to a protein is highest in babies and infants, presumably because their guts are more permeable to the passage of large molecules (Freese, 2007, Lara-Villoslada et al., 2005). Likewise, those suffering from “leaky gut syndrome” have a higher risk for developing food allergies (Woodford, 2007). At stomach pHs ranging from 2.0 to 4.0, antigenic epitopes of whey proteins are much more likely to survive and potentially pass into circulation (Schmidt et al., 1995). It is known that infants have generally high stomach pH (approximately 3.0-4.0 and can be higher after food intake). While the pH of an adult stomach can raise to 5.0 after food intake, infants also have an immature gut barrier making them particularly susceptible to exposure to dietary proteins (Schmidt et al., 1995, Thomas et al., 2004). Breastfed babies digest about 90% of the hLf in breastmilk, leaving up to 10% of protein to be taken up by gut cells (Lonnerdal, 2003). In the context of formula, with different factors potentially affecting protein digestibility and at possibly different concentrations to breastmilk, there is the possibility of more rhLf surviving digestion. These considerations are different for bLf found in formula, because human cells do not have receptors for taking up bLf (Lonnerdal, 2003), and thus much less is expected to pass through the gut wall. Therefore, the use of rhLf, especially of unknown structural composition when produced in various GMOs (Fujiyama et al., 2004), and in the context of complex formula when babies are not breastfed, should be subject to extensive safety evaluations. The significant structural similarities between milk proteins of different species also increase the danger of cross-reactivity (Cantisani et al., 1997, Sicherer, 2001). Indeed, Cantisani et al. (1997) found IgE antibodies in atopic infants exposed only to human breast milk reacted against proteins in cow’s milk. The concern here would be that the mixing of hLf and bLf in the human food supply might result in producing autoantibodies, those directed against normal molecules. Despite the high sequence similarity between bLf and hLf, however, so far no cross-reactivity between bLf and hLf has been seen (Roozendaal and Kallenberg, 1999). Nevertheless, the propensity of people to develop allergies to milk of bovine or goat origin suggests that the use of rhLf might cause an allergic reaction in GM animals used as biofactories, creating an animal welfare issue. Antibodies to hLf have been found in patients suffering from a number of different diseases “including lupus erythematosus (5-10%), drug-induced lupus erythematosus (100%), rheumatoid arthritis complicated with vasculitis (up to 45%), HIV (65%), and ulcerative colitis (50%)” (p. 290

20

Constructive Conversations/Kōrero Whakaaetanga Rpt 15 Nuijens et al., 1996). While this observation has been of diagnostic value, there is no direct evidence at this time that hLf or autoantibodies have contributed to the onset or complications of these diseases (Nuijens et al., 1996). Nevertheless, dietary rhLf would be a concern to those already making such antibodies. While specific human health effects of hLf as an allergen are still speculative, the potential allergenicity of even minor proportions of rhLf that have been glycosylated differently cannot be determined without much more extensive testing. In this context, it is relevant to note that rice-produced rhLf has a different glycosylation pattern than hLf; all eight types of carbohydrate groups found in rice-expressed rhLf were plant-derived (Fujiyama et al., 2004). Regardless of whether a protein has a history of being an allergen in its native context, expression in another species can alter its potential to be an allergen. This was demonstrated recently when a bean protein with no history of causing allergies was produced using recombinant techniques in peas. The pea-produced protein elicited an immune response in mice (Prescott et al., 2005). Importantly, the plant-specific glycans detected in rhLF produced in rice elicit antibodies in humans (Saint-Jore-Dupas et al., 2007). Moreover, the general immuno-stimulatory effects of lactoferrins cannot be ignored as a more general issue when or if rhLf is released on a large scale in GMOs, particularly plants, used as biofactories. Wild animals will feed on plants grown outside of contained laboratory facilities and thus rhLf should be evaluated as an allergen to ecologically important and culturally-prized animals. Recent court rulings in the United States are worth considering in this context, because they lead New Zealand in the field testing of PMPs. In August 2006, a US Federal Court found that the United States Department of Agriculture (USDA), Animal and Plant Health Inspection Service (APHIS) illegally issued permits to a number of companies to conduct field trials of PMPs. The USDA failed even to consider the potential impacts of these PMPs on herbivores or further up the food chain before issuing the permits. Thus, the court ruled that in issuing the permits, the USDA “violated both the [Endangered Species Act and the National Environmental Policy Act] in issuing the four permits” (p. 4 CFS v. USDA, 2006).

IV.5 Transcription factor and gene escort hLf has an affinity for specific DNA sequences between 9 and 13 nucleotides in length. Both apo and holo hLf protect the consensus sequences GGCACTT(G/A)C, TAGA(A/G)GATCAAA and ACTACAGTCTACA from DNaseI, but the holo form has more affinity (He and Furmanski, 1995). When these consensus sequences were incorporated into the promoter regions of special indicator genes, hLf caused them to be transcribed, the first step in gene expression (He and Furmanski, 1995). Lactoferrin-DNA complexes are reportedly taken up by human cells and then transported to the nucleus, thus giving hLf the potential to augment horizontal gene transfer as a DNA escort (Fleet, 1995). Indeed, this activity forms the basis of a 2001 United States patent which explicitly states: “The present invention relates generally to the use of DNA binding proteins to effect or enhance entry of DNA into cells for gene transfer or gene therapy. It more particularly relates to the production of chimeric, recombinant, or synthetic proteins containing DNA binding elements for the purposes of effecting or enhancing gene transfer or gene therapy. It also relates to the use of lactoferrin for enhancing entry of DNA into cells” (US Patent 6191257, 2007). A sequence-directed DNA uptake in mammalian cells is reminiscent of a similar phenomenon amongst naturally competent bacteria (Claverys and Martin, 2003). It correlates with the expression of a set of proteins, called com for competence. In contrast to the mechanism above, DNA is taken into the cell in single-stranded form where it is thought to serve as a substrate for single-strand specific DNA binding recombinases that initiate a search for similar sequences of DNA in the recipient

21

Constructive Conversations/Kōrero Whakaaetanga Rpt 15 genome. Many bacteria that are not known to be naturally competent have homologues of com genes, which might indicate that they do have the potential to take up DNA from the extracellular environment under conditions that have not been investigated. It remains uncertain how many bacteria or mammalian cells may take up DNA, or under what circumstances and in which environments they may be induced to take up DNA (Heinemann and Billington, 2004). This leads to a novel and important speculation. Lactoferrin’s roles as iron chelating and DNA-binding protein gives it the dual functionalities requisite to a gene transfer agent amongst bacteria. As indicated above, some bacteria have evolved to exploit holo-lactoferrin as an alternative to producing siderophores, thus specifically taking lactoferrin up from the environment as an iron-protein complex (Braun and Braun, 2002). In vivo, lactoferrin would likely be bound to DNA (or possibly RNA) that circulates in animal blood and other fluids, and possibly in plants (Anker and Stroun, 2000). Microbes in and on transgenic plants and animals used as biofactories may therefore be efficiently transformed by DNA that contains the lactoferrin binding sequence(s) as the iron and DNA binding proteins are actively transported across the different bacterial membranes in some species (Braun and Braun, 2002). The probability of a match to one 13-nucleotide sequence (e.g., ACTACAGTCTACA) is 1.5 x 10-8 (or one in one hundred million), while it is 6 x 10-8 to a 12mer sequence and 4 x 10-6 (or about four in a million) to a 9mer sequence. Thus, a particular 13 or 12mer sequence will occur about 20 times in the human genome just by chance. There are three consensus sequences, so the number of lactoferrin binding sites in a genome the size of humans and cows is expected to be around 6000, and rice 1,700, by chance alone. The three binding sequences are not absolute, but consensus. So variations of them will still bind lactoferrin. Therefore, the occurrence of potential lactoferrin binding sequences would be much higher.

22

Constructive Conversations/Kōrero Whakaaetanga Rpt 15

Chapter V: Conclusions Lactoferrin is a protein found naturally in mammals, but has species-specific differences and activities (Nuijens et al., 1996). Lactoferrin has many proposed applications, including use as a therapeutic for infectious and non-infectious diseases, immunomodulation, and gene therapy. Since lactoferrin is a relatively minor milk protein, production in biofactories might be a cost-effective way to dramatically increase production, especially of rhLf. However, this claim is relatively untested and may become more difficult to defend as fermentation systems improve. The very attributes that make lactoferrin an interesting and exciting molecule for commercial production also are relevant to broader-scale risks, particularly when the molecule is manufactured in plants or animals outside of contained laboratory walls. The activities of lactoferrin can neither be suppressed within the GM biofactory itself nor in the various plants, animals, microbes and viruses that might be exposed to it through their normal or occasional association with the GM biofactory. Of the risks covered in this report, the most immediate risk factors of producing rhLf in transgenic plants or animals derive from the protein’s toxic effects on microbes, fungi and viruses, and its potential to elicit an immune response in people, particularly infants, when it is derived from GMOs. In saying this, it is important to recognise that the implications of proteins that form amyloids are far from fully known and that the effects of such proteins have been spectacularly underestimated in the past, such as with BSE in the 1990s. Also, the ability of lactoferrin to serve as a gene vector is deserving of further study to provide assurance that its impacts will be minimal in the environment. The risks arising from hLf’s anti-infective properties are: 1) the possibility that large-scale exposure to many different kinds of non-target organisms will increase the chance of resistance arising somewhere, and then being transferred to organisms that are normal pathogens of people, animals or plants; 2) exacerbation of infections due to pathogens whose growth may be promoted by lactoferrin in those treated with it; 3) a change in host range of a lactoferrin-tolerant microbe/virus leading to emergent disease agents; and 4) lactoferrin-induced tolerance to other antibiotics in microbes that are normal human pathogens. The risks arising from hLf as a gene vector are that it will increase horizontal gene transfer. This gene transfer is not necessarily species-specific, so it might contribute to the movement of genes to both microbes and other organisms. These risks are accentuated uniquely by the use of recombinant techniques and biofactories as a method of production because they increase both the scale of exposure to non-target organisms and the diversity of non-target organisms exposed. Clinical antibiotics have had their useful lives limited by their overuse, both in and outside of medicine. Lactoferrin’s usefulness as an antibiotic might also be cut short, but because of its method of production rather than by an overuse of the product. To determine the actual implications of plant or animal biofactories of rhLf will require regulatory authorities to recognise each of these potential harm pathways and either evaluate the possibilities of harm management or ask for relevant new research.

23

Constructive Conversations/Kōrero Whakaaetanga Rpt 15

Bibliography Agiato, L. A. and Dyer, D. W. (1992). Siderophore production and membrane alterations by Bordetella pertussis in response to iron starvation. Infect. Immun. 60, 117-123. Amábile-Cuevas, C. F., Cardenas-Garcia, M. and Ludgar, M. (1995). Antibiotic Resistance. Am. Sci. 83, 320-329. Andres, M. T., Viejo-Diaz, M., Perez, F. and Fierro, J. F. (2005). Antibiotic Tolerance Induced by Lactoferrin in Clinical Pseudomonas aeruginosa Isolates from Cystic Fibrosis Patients. Antimicrob. Agents Chemother. 49, 1613-1616. Anker, P. and Stroun, M., eds. (2000). Circulating Nucleic Acids in Plasma or Serum, 1 edn (New York, New York Academy of Science). Anon. http://www.stuff.co.nz/4336629a11.html Date of Access: 25 December 2007. APHIS (2007). Issuance of permits to grow genetically engineer rice producing recombinant human lactoferrin, lysozyme or serum albumin. USDA. http://www.aphis.usda.gov/brs/aphisdocs/06_27801r_ea.pdf. Arnold, D., Di Biase, A. M., Marchetti, M., Pietrantoni, A., Valenti, P., Seganti, L. and Superti, F. (2002). Antiadenovirus activity of milk proteins: lactoferrin prevents viral infection. Antivir. Res. 53, 153-158. Bardor, M., Faye, L. and Lerouge, P. (1999). Analysis of the N-glycosylation of recombinant glycoproteins produced in transgenic plants. Trends Plant Sci. 4, 376-380. Bauernfeind, A., Schneider, I., Jungwirth, R. and Roller, C. (1998). Discrimination of Burkholderia gladioli from other Burkholderia species detectable in cystic fibrosis patients by PCR. J. Clin. Microbiol. 36, 2748-2751. Baxter, I., Lambert, P. and Simpson, I. (1997). Isolation from clinical sources of Burkholderia cepacia possessing characteristics of Burkholderia gladioli. J. Antimicrob. Chemother. 39, 169-175. Bellamy, W., Takase, M., Yamauchi, K., Wakabayashi, H., Kawase, K. and Tomita, M. (1992). Identification of the bactericidal domain of lactoferrin. Biochim. Biophys. Acta 1121, 130-136. Brandl, M. T. (2006). Fitness of human enteric pathogens on plants and implications for food safety. Annu. Rev. Phytopathol. 44, 367-392. Braun, V. and Braun, M. (2002). Active transport of iron and siderophore antibiotics. Curr. Opin. Microbiol. 5, 194-201. Bucciantini, M., Giannoni, E., Chiti, F., Baroni, F., Formigli, L., Zurdo, J., Taddei, N., Ramoni, G., Dobson, C. M. and Stefani, M. (2002). Inherent toxicity of aggregates implies a common mechanism for protein misfolding diseases. Nature 416, 507-511. Buxbaum, J. N. (2003). Diseases of protein conformation: what do in vitro experiments tell us about in vivo diseases? Trends Biochem. Sci. 28, 585-592.

24

Constructive Conversations/Kōrero Whakaaetanga Rpt 15 Cantisani, A., Giuffrida, M. G., Fabris, C., Bertino, E., Coscia, A., Oggero, R., Monti, G., Stroppiana, P. and Conti, A. (1997). Detection of specific IgE to human milk proteins in sera of atopic infants. FEBS Lett. 412, 515-517. Caughey, B. and Baron, G. S. (2006). Prions and their partners in crime. Nature 443, 803-810. CFS v. USDA (2006). Center for Food Safety et al. v. Mike Johanns, Secretary, U.S.Department of Agriculture et al. United States District Court for the District of Hawaii. http://www.earthjustice.org/library/legal_docs/hawaii-biopharm-order-81096.pdf. Chong, D. K. X. and Langridge, W. H. R. (2000). Expression of full-length bioactive antimicrobial human lactoferrin in potato plants. Transgen. Res. 9, 71-78. Claverys, J.-P. and Martin, B. (2003). Bacterial 'competence' genes: signatures of active transformation, or only remnants? Trends Microbiol. 11, 161-165. Collignon, P., Wegener, H. C., Braam, P. and Butler, C. D. (2005). The routine use of antibiotics to promote animal growth does little to benefit protein undernutrition in the developing world. Food Safety 41, 1007-1013. Crosby, L. (2003). Commercial production of transgenic crops genetically engineered to produce pharmaceuticals. www.biopharminternational.com. Dastidar, S. G., Mondal, U., Niyogi, S. and Chakrabarty, A. N. (1986). Antibacterial property of methyl-DOPA and development of antibiotic cross-resistances in m-DOPA mutants. Ind. J. Med. Res. 84, 142-147. de la Cruz, F., Garcia-Lobo, J. M. and Davies, J. (2002). Antibiotic resistance: how bacterial populations respond to a simple evolutionary force. In Bacterial Resistance to Antimicrobials, K. Lewis, A. A. Salyers, H. W. Taber, and R. G. Wax, eds. (New York and Basel, Marcel Dekker), pp. 19-36. Drobni, P., Naslund, J. and Evander, M. (2004). Lactoferrin inhibits human papillomavirus binding and uptake in vitro. Antivir. Res. 64, 63-68. Ekins, A., Khan, A. G., Shouldice, S. R. and Schryvers, A. B. (2004). Lactoferrin receptors in Gram-negative bacteria: Insights into the iron acquisition process. BioMet. 17, 235-243. Elass-Rochard, E., Legrand, D., Salmon, V., Roseanu, A., Trif, M., Tobias, P. S., Mazurier, J. and Spik, G. (1998). Lactoferrin Inhibits the Endotoxin Interaction with CD14 by Competition with the Lipopolysaccharide-Binding Protein. Infect. Immun. 66, 486-491. Ellis, R. J. and Pinheiro, J. T. (2002). Danger - misfolding proteins. Nature 416, 483-484. Emori, T. G. and Gaynes, R. P. (1993). An overview of nosocomial infections, including the role of the microbiology laboratory. Clin. Microbiol. Rev. 6, 428-442. Ferguson, G. C. and Heinemann, J. A. (2002). Recent history of trans-kingdom conjugation. In Horizontal Gene Transfer, M. Syvanen, and C. I. Kado, eds. (London and San Diego, Academic Press), pp. 3-17.

25

Constructive Conversations/Kōrero Whakaaetanga Rpt 15 Fernandez-Escamilla, A.-M., Rousseau, F., Schymkowitz, J. and Serrano, L. (2004). Prediction of sequence-dependent and mutational effects on the aggregation of peptides and proteins. Nat. Biotechnol. 22, 1302-1306. Fischer, R., Debbabi, H., Dubarry, M., Boyaka, P. and Tome, D. (2006). Regulation of physiological and pathological Th1 and Th2 responses by lactoferrin. Biochem. Cell Biol. 84, 303-311. Fleet, J. C. (1995). A new role for lactoferrin: DNA binding and transcription activation. Nutr. Rev. 53, 226. FOE (2007). Does Africa need a genetically modified solution to diarrhea? Friends of the Earth. Fox, J. L. (2006). Turning plants inot protein factories. Nat. Biotechnol. 24, 1191-1193. Freese, B. (2007). A grain of caution: a critical assessment of pharmaceutical rice. Center for Food Safety. www.centerforfoodsafety.org. Fujiyama, K., Sakai, Y., Misaki, R., Yanagihara, I., Honda, T., Anzai, H. and Seki, T. (2004). N-linked glycan structures of human lactoferrin produced by transgenic rice. Biosci. Biotechnol. Biochem. 68, 2565-2570. Furuya, E. Y. and Lowy, F. D. (2005). Antimicrobial-resistant bacteria in the community setting. Nat. Rev. Microbiol. 4, 36-45. Gast, K., Modler, A. J., Damaschun, H., Krober, R., Lutsch, G., Zirwer, D., Golbik, R. and Damaschun, G. (2003). Effect of environmental conditions on aggregation and fibril formation of barstar. Eur. Biophys. J. 32, 710-723. Gomord, V., Chamberlain, P., Jefferis, R. and Faye, L. (2005). Biopharmaceutical production in plants: problems, solutions and opportunities. Trends Biotechnol. 23, 559-565. Goven, J., Hunt, L., Shamy, D. and Heinemann, J. A. (2008). Animal biopharming in New Zealand drivers, scenarios and practical implications. Constructive Conversations/Kōrero Whakaaetanga Research Report 12. Groot, F., Geijtenbeek, T. B. H., Sanders, R. W., Baldwin, C. E., Sanchez-Hernandez, M., Floris, R., van Kooyk, Y., de Jong, E. C. and Berkhout, B. (2005). Lactoferrin Prevents Dendritic Cell-Mediated Human Immunodeficiency Virus Type 1 Transmission by Blocking the DC-SIGN--gp120 Interaction. J. Virol. 79, 3009-3015. Hancock, R. E. W. and Chapple, D. S. (1999). Peptide Antibiotics. Antimicrob. Agents Chemother. 43, 1317-1323. Hancock, R. E. W. and Scott, M. G. (2000). The role of antimicrobial peptides in animal defenses. Proc. Natl. Acad. Sci. USA 97, 8856-8861. Harper. http://www-fst.ag.ohio-state.edu/People/HARPER/Functional-foods/Milk%20Components/LACTOFERRIN.html Date of Access: 2 December 2007. He, J. and Furmanski, P. (1995). Sequence specificity and transcriptional activation in the binding of lactoferrin to DNA. Nature 373, 721-724.

26

Constructive Conversations/Kōrero Whakaaetanga Rpt 15 Heinemann, J. A. (1999). How antibiotics cause antibiotic resistance. Drug Discov. Today 4, 72-79. Heinemann, J. A. (2007). A typology of the effects of (trans)gene flow on the conservation and sustainable use of genetic resources. Bsp35. UN FAO. http://www.fao.org/biotech/doc.asp. Heinemann, J. A., Ankenbauer, R. G. and Amábile-Cuevas, C. F. (2000). Do antibiotics maintain antibiotic resistance? Drug Discov. Today 5, 195-204. Heinemann, J. A. and Billington, C. (2004). How do genomes emerge from genes? ASM News 70, 464-471. Heinemann, J. A. and Traavik, T. (2004). Problems in monitoring horizontal gene transfer in field trials of transgenic plants. Nat. Biotechnol. 22, 1105-1109. Humphrey, B. D., Huang, N. and Klasing, K. C. (2002). Rice Expressing Lactoferrin and Lysozyme Has Antibiotic-Like Properties When Fed to Chicks. J. Nutr. 132, 1214-1218. Hyvonen, P., Suojala, L., Orro, T., Haaranen, J., Simola, O., Rontved, C. and Pyorala, S. (2006). Transgenic Cows That Produce Recombinant Human Lactoferrin in Milk Are Not Protected from Experimental Escherichia coli Intramammary Infection. Infect. Immun. 74, 6206-6212. in-Pharma. http://www.in-pharmatechnologist.com/news/ng.asp?id=60993-pharming-agresearch-lactoferin Date of Access: 11 December 2007. Jarosik, G. P. and Land, C. B. (2000). Identification of a Human Lactoferrin-Binding Protein in Gardnerella vaginalis. Infect. Immun. 68, 3443-3447. Kapuscinski, A. R. (2005). Current scientific understanding of the environmental biosafety of transgenic fish and shellfish. Rev. Sci. Tech. Off. Int. Epiz. 24, 309-322. Kaye-Blake, W., Saunders, C. and Ferguson, L. (2007). Preliminary economic evaluation of biopharming in New Zealand. Constructive Conversations/Kōrero Whakaaetanga Research Report 10 and Agribusiness and Economics Research Unit, Lincoln University, Report 296. www.lincoln.ac.nz/story9430.html? Kim, S. J., Park, Y. and Hong, H. J. (2005). Antibody engineering for the development of therapeutic antibodies. Mol. Cells 20, 17-29. Kristiansen, J. E. (1991). Antimicrobial Activity of Nonantibiotics. ASM News 57, 135-139. Kristiansen, J. E., Justesen, T., Hvidberg, E. F. and Andersen, L. P. (1989). Trimipramine and other antipsychotics inhibit Campylobacter pylori in vitro. Pharmacol. Toxicol. 64, 386-388. Kues, W. A. and Niemann, H. (2004). The contribution of farm animals to human health. Trends Biotechnol. 22, 286-294. Kumari, J., Swain, T. and Sahoo, P. K. (2003). Dietary bovine lactoferrin induces changes in immunity level and disease resistance in Asian catfish Clarias batrachus. Vet. Immunol. Immunopath. 94, 1-9. Lactoferrin Research. http://www.lactoferrinresearch.org/ Date of Access: 4 December 2007.

27

Constructive Conversations/Kōrero Whakaaetanga Rpt 15 Lane, K. T. and Beese, L. S. (2006). Thematic review series: Lipid Posttranslational Modifications. Structural biology of protein farnesyltransferase and geranylgeranyltransferase type I. J. Lipid Res. 47, 681-699. Lara-Villoslada, F., Olivares, M. and Xaus, J. (2005). The Balance Between Caseins and Whey Proteins in Cow's Milk Determines its Allergenicity. J. Dairy Sci. 88, 1654-1660. Levy, S. B. (1998). The challenge of antibiotic resistance. Sci. Am. 278, 32-39. Linke, R. P., Joswig, R., Murphy, C. L., Wang, S., Zhou, H., Gross, U., Rocken, C., Westermark, P., Weiss, D. T. and Solomon, A. (2005). Senile seminal vesicle amyloid is derived from semenogelin I. J. Lab. Clin. Med. 145, 187-193. Lonnerdal, B. (2002). Expression of Human Milk Proteins in Plants. J Am Coll Nutr 21, 218S-221. Lonnerdal, B. (2003). Nutritional and physiologic significance of human milk proteins. Am J Clin Nutr 77(supl), 1537S-1543S. Ma, J. K.-C., Drake, P. M. W. and Christou, P. (2003). The production of recombinant pharmaceutical proteins in plants. Nat. Rev. Genet. 4, 794-805. MacDonald, T. T. and Monteleone, G. (2005). Immunity, inflammation, and allergy in the gut. 307, 1920-1925. Malnoy, M., Venisse, J.-S., Brisset, M.-N. and Chevreau, E. (2003). Expression of bovine lactoferrin cDNA confers resistance to Erwinia amylovora in transgenic pear. Mol. Breeding 12, 231-244. Manzi, A. E., Norgard-Sumnicht, K., Argade, S., Marth, J. D., van Halbeek, H. and Varki, A. (2000). Exploring the glycan repertoire of genetically modified mice by isolation and profiling of the major glycan classes and nano-NMR analysis of glycan mixtures. Glycobiology 10, 669-689. Marvier, M. and Van Acker, R. C. (2005). Can crop transgenes be kept on a leash? Front. Ecol. Environ. 3, 99-106. Mitra, N., Sinha, S., Ramya, T. N. C. and Surolia, A. (2006). N-linked oligosaccharides as outfitters for glycoprotein folding, form and function. Trends Biochem. Sci. 31, 156-163. Mourgues, F., Brisset, M.-N. and Chevreau, E. (1998). Strategies to improve plant resistance to bacterial diseases through genetic engineering. Trends Biotechnol. 16, 203-210. NAS (2004). Infant Formula: Evaluating the Safety of New Ingredients. US National Academy of Sciences. Nilsson, M. R. and Dobson, C. M. (2003). In vitro characterization of lactoferrin aggregation and amyloid formation. Biochem. 42, 375-382. Norregaard Jensen, O. (2004). Modification-specific proteomics: characterization of post-translational modifications by mass spectrometry. Curr. Opin. Chem. Biol. 8, 33-41. Nuijens, J., Berkel, P. and Schanbacher, F. (1996). Structure and biological actions of lactoferrin. J. Mam. Gland Biol. Neopl. 1, 285-295.

28

Constructive Conversations/Kōrero Whakaaetanga Rpt 15 NUTRA. http://www.nutraingredients.com/news/ng.asp?id=64905-pharming-lactoferrin Date of Access: 11 December 2007. O'Ryan, M., Prado, V. and Pickering, L. K. (2005). A millennium update on pediatric diarrheal illness in the developing world. Sem. Ped. Infect. Dis. 16, 125-136. Ochoa, T. J., Brown, E. L., Guion, C. E., Chen, J. Z., McMahon, R. J. and Cleary, T. G. (2006). Effect of lactoferrin on Enteroaggregative E. coli (EAEC). Biochem. Cell Biol. 84, 369-376. Oh, S.-M., Hahm, D. H., Kim, I.-H. and Choi, S.-Y. (2001). Human Neutrophil Lactoferrin trans-Activates the Matrix Metalloproteinase 1 Gene through Stress-activated MAPK Signaling Modules. J. Biol. Chem. 276, 42575-42579. Pawar, A. P., Dubay, K. F., Zurdo, J., Chiti, F., Vendruscolo, M. and Dobson, C. M. (2005). Prediction of “Aggregation-prone” and “Aggregationsusceptible” Regions in Proteins Associated with Neurodegenerative Diseases. J. Mol. Biol. 350, 379-392. Prescott, V. E., Campbell, P. M., Moore, A., Mattes, J., Rothenberg, M. E., Foster, P. S., Higgins, T. J. V. and Hogan, S. P. (2005). Transgenic Expression of Bean alpha-Amylase Inhibitor in Peas Results in Altered Structure and Immunogenicity. J. Agric. Food Chem. 53, 9023-9030. Rates, S. M. K. (2001). Plants as source of drugs. Toxicon 39, 603-613. Ray, S., Dastidar, S. G. and Chakrabarty, A. N. (1980). Antibiotic Cross-Resistance Patterns of Ambodryl and Promazine Resistant Mutants. Br. J. exp. Path. 61, 465-470. Roozendaal, C. and Kallenberg, C. G. M. (1999). Are anti-neutrophil cytoplasmic antibodies (ANCA) clinically useful in inflammatory bowel disease (IBD)? Clin. Exp. Immunol. 116, 206-213. Sahm, D. F., Jones, M. E., Hickey, M. L., Diakun, D. R., Mani, S. V. and Thornsberry, C. (2000). Resistance surveillance of Streptococcus pneumoniae, Haemophilus influenzae and Moraxella catarrhalis isolated in Asia and Europe, 1997-1998. J. Antimicrob. Chemother. 45, 457-466. Saint-Jore-Dupas, C., Faye, L. and Gomord, V. (2007). From planta to pharma with glycosylation in the toolbox. Trends Biotechnol. 25, 317-323. Schmidt, D. G., Meijer, R. J. G. M., Slangen, C. J. and van Beresteijn, E. C. H. (1995). Raising the pH of the pepsin-catalysed hydrolysis of bovine whey protein increases the antigenicity of the hydrolysates. Clin. Exp. Aller. 25, 1007-1017. Schneider, C. K., Kalinke, U. and Lower, J. (2006). TGN1412 - a regulator's perspective. Nat Biotechnol. 24, 493-496. Sicherer, S. H. (2001). Clinical implications of cross-reactive food allergens. J. Aller. Clin. Immun. 108, 881-890. Sigurdson, C. J., Wisniewski, T. and Frangione, B. (2002). Infectivity of amyloid diseases. Trends Mol. Med. 8, 411-413. Sparrow, P. A. C., Irwin, J. A., Dale, P. J., Twyman, R. M. and Ma, J. K. C. (2007). Pharma-planta: road testing the developing regulatory guidelines for plant-made pharmaceuticals. Transgen. Res. 16, 147-161.

29

Constructive Conversations/Kōrero Whakaaetanga Rpt 15 Stefani, M. (2004). Protein misfolding and aggregation: new examples in medicine and biology of the dark side of the protein world. Biochim. Biophys. Acta 1739, 5-25. Swart, P. J., Harmsen, M. C., Kuipers, M. E., Van Dijk, A. A., Van Der Strate, B. W. A., Van Berkel, P. H. C., Nuijens, J. H., Smit, C., Witvrouw, M., De Clercq, E., et al. (1999). Charge modification of plasma and milk proteins results in antiviral active compounds. J. Pept. Sci. 5, 563-576. Swartz, M. N. (1994). Hospital-acquired infections: diseases with increasingly limited therapies. Proc. Natl. Acad. Sci. USA 91, 2420-2427. Tartaglia, G. G., Pechmann, S., Dobson, C. M. and Vendruscolo, M. (2007). Life on the edge: a link between gene expression levels and aggregation rates of human proteins. Trends Biochem. Sci. 32, 204-206. Teng, C. T. (2002). Lactoferrin gene expression and regulation: an overview. Biochem. Cell Biol. 80, 7-16. Thomas, K., Aalbers, M., Bannon, G. A., Bartels, M., Dearman, R. J., Esdaile, D. J., Fu, T. J., Glatt, C. M., Hadfield, N. and Hatzos, C. (2004). A multi-laboratory evaluation of a common in vitro pepsin digestion assay protocol used in assessing the safety of novel proteins. 39, 87-98. Troost, F. J., Saris, W. H. M. and Brummer, R. J. M. (2003). Recombinant human lactoferrin ingestion attenuates indomethacin-induced enteropathy in vivo in healthy volunteers. Eur J Clin Nutr 57, 1579-1585. Twyman, R. M., Stoger, E., Schillberg, S., Christou, P. and Fischer, R. (2003). Molecular farming in plants: host systems and expression technology. Trends Biotechnol. 21, 570-578. US Patent 6191257. http://www.patentstorm.us/patents/6191257-fulltext.html Date of Access: 17 December 2007. USDA (2004). USDA/APHIS Environmental Assessment In response to permit application (04-302-01r) received from Ventria Bioscience for field-testing of rice, Oryza sativa, genetically engineered to express human lactoferrin. United States Department of Agriculture. http://www.aphis.usda.gov/brs/aphisdocs/04_30201r_ea.pdf. Van den Steen, P., Rudd, P. M., Dwek, R. A. and Opdenakker, G. (1998). Concepts and Principles of O-Linked Glycosylation. Crit. Rev. Biochem. Mol. Biol. 33, 151-208. Viani, R. M., Gutteberg, T. J., Lathey, J. L. and Spector, S. A. (1999). Lactoferrin inhibits HIV-1 replication in vitro and exhibits synergy when combined with zidovudine. AIDS 13, 1273-1274. Wal, J. M. (2001). Structure and function of milk allergens. Allergy 56, 35-38. Weidenbaier, C., Kristian, S. A. and Peschel, A. (2003). Bacterial resistance to antimicrobial host defenses - an emerging target for novel antiinfective strategies? Curr. Drug Targ. 4, 643-649. Weir, E. (2002). Shigella: Wash your hands of the whole dirty business. CMAJ 167, 281-. Wilusz, C. J., Wang, W. and Peltz, S. W. (2001). Curbing the nonsense: the activation and regulation of mRNA surveillance. Genes Dev. 15, 2781-2785.

30

Constructive Conversations/Kōrero Whakaaetanga Rpt 15 Wiuff, C., Zappala, R. M., Regoes, R. R., Garner, K. N., Baquero, F. and Levin, B. R. (2005). Phenotypic Tolerance: Antibiotic Enrichment of Noninherited Resistance in Bacterial Populations. Antimicrob. Agents Chemother. 49, 1483-1494. Woodford, K. (2007). Devil in the Milk (Nelson, Craig Potton). Yu, R.-H. and Schryvers, A. B. (2002). Bacterial lactoferrin receptors: insights from characterizing the Moraxella bovis receptors. Biochem. Cell Biol. 80, 81-90. Zasloff, M. (2002). Antimicrobial peptides of multicellular organisms. Nature 415, 389-395. Zhang, Z., Coyne, D. P., Vidaver, A. K. and Mitra, A. (1998). Expression of Human Lactoferrin cDNA Confers Resistance to Ralstonia solanacearum in Transgenic Tobacco Plants. Phytopathol. 88, 730-734. Zhong, J., Wang, Y. and Zhu, Z. (2002). Introduction of the human lactoferrin gene into grass carp (Ctenopharyngodon idellus) to increase resistance against GCH virus. Aquaculture 214, 93-101.

31