Introduction
The past two decades has seen considerable interest in the area of functional foods and bioactive peptides from both scientists and the potential consumer. It is now recognised that some food proteins and food-derived peptides are able to elicit beneficial physiological effects and therefore have the potential to influence health and may help reduce the risk of chronic disease. Hence there is considerable commercial interest in the development of products with potential health benefits, and consequently there are already a number of such products commercially available in different countries around the world such as Japan, USA and Europe(Reference Korhonen1).
Although much of the work to date has focused on milk, cheese and other dairy products as sources of bioactive proteins and peptides, interest in other sources of food-derived bioactive peptides is continuing to grow and gain momentum. Various fish species such as sardines, tuna, bonito and salmon have been found to be viable sources of bioactive peptides as have other animal products such as blood, eggs and gelatine. Plants are also potential sources, with studies reporting bioactive peptides derived from pulses, wheat, rice, soya, pumpkin and even mushrooms. Clearly the reported sources of bioactive peptides are numerous as are the range of biological effects attributed to their consumption.
Much less prevalent is detailed information from animal and human studies demonstrating the efficacy and safety of bioactive peptides. Although many of the products are expected to be safe because they have been consumed for centuries and therefore have GRAS (generally recognised as safe) status there are a number of issues which must also be considered, such as the potential for allergenicity or intolerance. It is important that the efficacy of the peptides is determined so that scientifically substantiated claims can be made along with the upper safe limits of products. A significant amount of information is still required to demonstrate that the relevant bioactive peptides are in fact produced and absorbed in sufficient amounts to elicit a beneficial response. This raises another potential issue, if bioactive peptides are to be generated within a food product prior to consumption, for example due to fermentation, it must be shown that the peptides survive the digestive processes and are absorbed in quantities sufficient to cause a physiological effect. Likewise, if the peptide is expected to be generated during digestion, it must be shown to be produced and to be absorbed in adequate quantities to elicit an effect. Independent of the method of production of the bioactive peptide the quantity of product required to be consumed must also be realistic. For the safety and confidence of the consumer it is important that the methods used to measure these factors are well documented and preferably standardized.
The present contribution aims to summarise evidence concerning the generation, absorption and health benefits of bioactive proteins and peptides. In addition, the question of whether it is feasible to incorporate measures of health optimisation as an adjunct to classical protein quality description is discussed.
Intact bioactive proteins and peptides with direct effects on the intestine
Milk is a fluid specifically designed to meet the nutritional and developmental needs of the new-born and undoubtedly has a role beyond nutrition, acting in the infant luminally in the gut at the mucosal surface and after absorption acting systemically to affect gut maturation(Reference Kelly and Coutts2, Reference Meisel3). A number of intact proteins, particularly those from milk and those occurring at higher concentrations in colostrum such as immunoglobulins, lactoferrin and growth factors have been shown to directly exert their in vivo effects by acting directly in the intestine(Reference Moller, Scholz-Ahrens and Roos4). In fact it has been reported that intestinal epithelial cells show receptor-mediated binding of lactoferrin(Reference Hu, Mazurier and Sawatzki5) and exogenous opioid peptides appear to target the intestinal brush border membrane(Reference Guesdon, Pichon and Tomé6).
At least some of the major bioactive milk proteins such as lactoferrin and immunoglobulins have been shown to be at least partially resistant to digestive enzymes(Reference Drescher, Roos and Pfeuffer7–Reference Korhonen and Marnila9) making direct effects feasible. While some of these proteins are also precursors for bioactive peptides it is their biological function in their intact form that will be discussed in this section.
The industrial production of lactoferrin which is estimated at over 70 tons annually(Reference Korhonen and Pihlanto10) is a reflection of the commercial interest in this product, largely due to its bioactive properties. Lactoferrin is a naturally occurring iron-binding glycoprotein found in milk, and is widely considered to be an important anti-microbial component for protecting the host against microbial infections(Reference Baker, Baker and Koon11). However, the prophylactic effects of lactoferrin against septicaemia, which have been demonstrated in vivo are due to more than antimicrobial effects. Lactoferrin has been shown to modulate inflammatory responses by preventing cytokine release from monocytes and also by regulating the proliferation, differentiation and activation of immune cells(Reference Baveye, Elass and Mazurier12, Reference Elass, Baveye and Fernig13).
Lactoperoxidase, the most abundant enzyme in milk has an antimicrobial role and has synergistic effects with immunoglubulins and lactoferrin(Reference Korhonen and Pihlanto10). Epidermal growth factor (EGF) and transforming growth factor-β (TGF-β) are both present in bovine colostrum and studies in human infants, suckling rats, lambs and pigs suggest that EGF at least can survive the gastrointestinal tract in neonatal animals(Reference Britton, George-Nasscimento and Koldovsky14–Reference Schusdziarra, Schick and De La Fuente17). EGF exhibits multiple effects including exerting a trophic effect on epithelia, leading to acceleration of cell maturation and stimulation of cell proliferation(Reference Weaver18). TGF-β has been implicated in epithelial cell growth and differentiation and aids in the repair of injured tissue(Reference Donnet-Hughes, Duc and Serrant19). Diets containing TGF-β have been shown to have favourable effects in the treatment of various forms of inflammatory bowel disease including: inducing remission and promoting mucosal healing(Reference Donnet-Hughes, Duc and Serrant19–Reference Fell, Paintin and Arnaud-Battandier21), reducing intestinal inflammation(Reference Beattie, Schiffrin and Donnet-Hughes20, Reference Ozawa, Miyata and Nishimura22), reducing leukocytes, acute phase reactants(Reference Schriffin, El Yousfi and Faure23) and expression of interferon-γ(Reference Beattie, Schiffrin and Donnet-Hughes20, Reference Fell, Paintin and Arnaud-Battandier21). The importance of TGF-β in preventing the development of inflammatory bowel disease (IBD) in animal models has been demonstrated in studies which show that both systemic administration and induction of TGF-β have protective effects against induced colitis, and administration of anti-TGF-β antibodies appears to oblate this effect(Reference Neurath, Fuss and Kelsall24, Reference Sanchez de Medina, Daddaoua and Requena25). Results such as these clearly suggest there may be a possible role for bioactive peptides in enteral diets used as therapy for IBD.
Since the intestinal concentration of some dietary peptides is high, it is possible that even though their affinity for cellular receptors is relatively low a physiologically significant effect may still occur, particularly since several peptides may act synergistically(Reference Meisel3, Reference Pellegrini26, Reference Teschemacher27). Moughan et al. (Reference Moughan, Fuller and Han28) have reported that some food-derived peptides are able to stimulate the secretion of protein in the gut lumen while others either inhibit the reabsorption of amino acids or affect both activities. A large number of reports (reviewed in Moughan et al. (Reference Moughan, Fuller and Han28)) describe the effects of food-derived bioactive peptides on gut function by regulating the gastric emptying rate and potentially satiety, affecting tissue growth and increasing the secretory and absorptive capacity of the gut. Many of the peptides identified with these effects are opioid agonists and antagonists which will be discussed in a later section.
Release of bioactive peptides
Although some food proteins are able to elicit their effects by acting directly in their intact form, generally it is peptides (usually between 3 to 20 amino acids in length) derived from the parent protein that are of most interest. There are three main mechanisms by which bioactive peptides can be released; the first two being the result of normal digestive processes: 1) degradation via digestive enzymes 2) digestion via microbial enzymes primarily in the large intestine or 3) in vitro hydrolysis during food processing, usually by microbial fermentation or proteolytic digestion. Acid hydrolysis, although less expensive is more difficult to control and can result in damage to certain amino acids and is therefore used less frequently.
Production of bioactive peptides via normal digestive processes relies on the active form of the peptide firstly being generated and remaining intact at its site of absorption or action, which can occur throughout the intestine. In many instances this means that the protein or the peptide itself must be at least partially resistant to proteolysis, as normally most food proteins are totally digested during passage through the small intestine. Several proteins such as lactoferrin and immunoglobulins have been shown to partly resist hydrolysis in the small intestine(Reference Moller, Scholz-Ahrens and Roos4, Reference Drescher, Roos and Pfeuffer7, Reference Roos, Mahe and Benamouzig8), and tripeptides with the sequence PP at their C-terminus have also been shown to be resistant to digestion by peptidases(Reference Yoshimoto, Fischl and Orlowski29, Reference Mock, Green and Boyer30). Following casein consumption casein phosphopeptides (CPP) have been found in the intestinal contents of rats(Reference Naito and Suzuki31, Reference Kitts, Yuan and Nagasawa32) and in rat faeces(Reference Kasai, Honda and Kiriyama33). Casein (native and acid casein) has also been shown to have a role in protecting certain peptides from digestion potentially by blocking the active site of proteolytic enzymes(Reference Kelly and Coutts2).
Due to the different cleavage site specificities of microbial enzymes the peptides produced due to their action are likely to be different to those resulting from cleavage with digestive enzymes such as trypsin. This includes both peptides deriving from the large intestine and also from bacterial fermentation processes such as during cheese ripening. A potential issue with pre-hydrolysed peptides is the risk that active peptides may be hydrolysed by host peptidases during the digestion, absorption processes, although it is also possible that active peptides may be generated from partially hydrolysed and inactive proteins. Issues such as these reinforce the importance of carrying out in vivo studies for demonstrating efficacy, as results from in vitro studies using hydrolysates cannot be extrapolated to imply beneficial effects in vivo.
Absorption of bioactive peptides
The majority of bioactive peptides described to date cause systemic effects and therefore must either be absorbed from the intestine, act directly on the intestinal tract or via receptors and cell signalling in order to elicit their action. While it is known that the intestine is able to absorb di- and tripeptides relatively easily(Reference Adibi and Morse34, Reference Hara, Funabiki and Iwata35) there is little information in the literature documenting the absorption and kinetics of absorption of higher molecular weight bioactive peptides. There is evidence using radioactively labelled peptides suggesting that material infused into the jejunum is rapidly cleared from the systemic circulation(Reference Bloch, Wright and Bishara36). However, there is also a report of at least two long peptides derived from casein being detected in the plasma of adults following yoghurt or milk consumption(Reference Chabance, Marteau and Rambaud37). In addition, caseinmacropeptide has been found in the blood of rats after oral administration(Reference Fosset, Fromentin and Gietzen38). Although a significant amount of research is still required to document the fate of specific bioactive peptides, it appears that certain food-derived peptides are able to be absorbed, with the extent of absorption determined by the nature of the peptide, and tending to decrease with increasing chain length.
Assuming that the bioactive peptide is absorbed, the concentrations reaching the target cells or receptors must be sufficient to cause a quantifiable and sustained response. Studies in rats have shown that the fermented milk product Calpis™, which contains the tripeptides VPP and IPP is able to lower blood pressure(Reference Nakamura, Masuda and Takano39, Reference Masuda, Nakamura and Takano40), and that these two peptides can be detected in aortal tissue of spontaneously hypertensive rats six hours after the oral administration of a single dose of Calpis™. In vitro studies using the human intestinal cell-line CaCo-2 have shown that a significant amount of the peptide VPP was transported intact across the monolayer, but was not detected in the cells, leading the authors to conclude that paracellular diffusion was the main mechanism of transport of this peptide across the monolayer(Reference Satake, Enjoh and Nakamura41).
Diverse effects of bioactive proteins and peptides
The range of physiological effects attributed to the consumption of food-derived bioactive peptides and proteins is diverse including immunomodulatory, anti-hypertensive, anti-carcinogenic, anti-microbial and anti-inflammatory effects amongst others(Reference Korhonen1, Reference Moller, Scholz-Ahrens and Roos4). Taken at face value, consumption of foods rich in bioactive peptides, particularly those with multiple functions has the potential to have a huge impact on health and wellbeing, being able to optimise health and therefore prevent infection and disease. The ability to prevent and treat chronic disease via consumption of functional foods could vastly decrease health care costs which are expected to rise dramatically in the future given the aging population. However, a major stumbling block is that the veracity of the evidence supporting the effects of different bioactive peptides varies considerably.
Antihypertensive
Hypertension is a potentially serious condition which increases risk of cardiovascular disease and stroke. Peptides with potential anti-hypertensive effects are among the most extensively studied in the functional foods arena. The majority have been tested for their ability to inhibit the activity of angiotensin-I converting enzyme (ACE), thus lowering the production of the potent vasoconstrictor angiotensin II. Whilst many peptides have been subjected to in vitro studies a number have also been tested in hypertensive animal models and in human subjects. Tripeptides from bonito muscle (IKP)(Reference Yokoyama, Chiba and Yoshikawa42) and chicken muscle (IKW)(Reference Fujita, Yokoyam and Yoshikawa43) as well as the previously mentioned IPP and VPP from fermented milk(Reference Nakamura, Masuda and Takano39, Reference Masuda, Nakamura and Takano40) have all been shown to lower blood pressure in spontaneously hypertensive rats. In a study using spontaneously hypertensive rats the consumption of Calpis™ led to a reduction in angiotensin-I converting enzyme activity in aortal tissue compared to the control group given saline(Reference Masuda, Nakamura and Takano40). In an eight week placebo controlled study in hypertensive humans, daily consumption of 100 ml of Calpis™ was shown to significantly lower blood pressure after both four and eight weeks, compared to the control group in which blood pressure did not change(Reference Hata, Yamamoto and Ohni44). Peptides also exist which have been shown to lower blood pressure in spontaneously hypertensive rats by causing vasodilation mediated via either prostacyclin or the bradykinin B1 receptor(Reference Yoshikawa, Fujita and Matoba45).
A number of fermented milk drinks, yoghurts and hydrolysed milk preparations which contain either the ACE-inhibitory peptides IPP and VPP or VPP, TTMPLW and RY are commercially available in Japan, Spain, USA and Europe(Reference Gilani, Xiao and Lee46), with at least seven of these products having Food of Specific Health Use (FOSHU) approval in Japan.
Cholesterol Lowering Bioactive Peptides
Elevated blood cholesterol has been identified as a major risk factor for cardiovascular disease, and significant research activity is ongoing into methods for the prevention and lowering of plasma cholesterol levels, including the potential use of bioactive peptides and proteins. Consumption of a number of food proteins, particularly plant proteins such as soyabean protein, is known to contribute to a lowering of serum cholesterol levels(Reference Carroll47). There is a considerable body of evidence from controlled clinical studies showing that replacing animal protein with soya protein results in beneficial effects on serum lipids including a lowering of total cholesterol, low-density lipoprotein (LDL) cholesterol and triglycerides, but not effecting the levels of the so called ‘good’ cholesterol high-density lipoprotein cholesterol (HDL)(Reference Anderson, Johnstone and Cook-Newell48).
Several mechanisms have been suggested for the cholesterol lowering effects of food proteins, including prevention or blocking of the reabsorption of bile acids and/or cholesterol synthesis(Reference Sugano, Yamada and Yoshihida49), inhibition of cholesterol synthesis, and stimulation of LDL receptor (LDL-R) receptor transcription(Reference Cho, Juillerat and Lee50). It appears that several or all of these mechanisms may operate based on the proteins or peptides being tested. Following digestion of soyabean protein, high molecular weight core peptides remain, and these are able to prevent the reabsorption of bile acids and therefore are able to lower cholesterol(Reference Sugano, Yamada and Yoshihida49). In mice, lower molecular weight peptides such as α-lactotensin (β-lactoglobulin fragment) and a peptide from soyabean glycinin have both been shown to reduce serum cholesterol levels, although neither increases excretion of faecal cholesterol or bile acids(Reference Yoshikawa, Fujita and Matoba45). In vitro studies by Cho et al. (Reference Cho, Juillerat and Lee50) have shown that the cholesterol lowering activity of a soyabean protein hydrolysate is due at least in part to an upregulation of LDL-R transcription.
Opioid Peptides
As mentioned a number of reports in the literature describe the effect of opioid peptides on the gastrointestinal tract, where they may have localised effects, effects mediated via gut hormones or systemically-mediated effects following their absorption. Since there are several opioid receptors, each responsible for specific physiological effects, the effects elicited by peptides with opioid activity can be diverse, including affecting appetite, respiratory depression, behaviour (e.g. handling of stress) and gastrointestinal motility. Of the receptors, the μ-receptor affects emotional behaviour, pain sensation and suppression of intestinal motility, the δ-receptor emotional behaviour and the κ-receptor sedation and the regulation of satiety signals and hence food intake.
The sources of food-derived peptides with opioid activity are diverse, including from cereal(Reference Morley, Levine and Yamada51), bovine-, human-, ovine- and water buffalo-milk(Reference Meisel and Frister52–Reference Teschemacher, Koch and Brantl55), haemoglobin from bovine blood (haemorphins)(Reference Brantl, Gramsch and Lottspeich56), gluten and gliaden from wheat, zein from maize, hordein from barley and soya α-protein and cytochrome b(Reference Meisel and Schlimme57). Generation of opioid peptides can be achieved by digesting the parent protein with a variety of digestive enzymes, either alone or combination: e.g. pepsin, pepsin followed by trypsin, or chymotrypsin alone(Reference Pihlanto-Leppälä58); however to date not all have been shown to be released in the gut following oral administration.
A number of bioactive peptides derived from milk proteins are opioid agonists, binding to opioid receptors and exhibiting morphine-like effects. Peptides in this category include: casomorphins, α-lactorphin (from α-lactalbumin), β-lactorphin (from β-lactoglobulin), serorphin from serum albumin. Other peptides such as lactoferroxins and casoxins are opioid antagonists and are able to depress the agonist activity of enkephalin(Reference Yoshikawa, Tani and Yoshimura59, Reference Yoshikawa, Tani and Chiba60). Generally the peptides released from α- and β-casein elicit agonist responses whilst those from κ-casein give rise to antagonist responses(Reference Clare and Swaisgood61).
Opioid Agonists
The range of physiological effects attributed to milk-derived opioid agonist peptides are diverse and include inhibiting diarrhoea, prolonging gastrointestinal transit time by inhibiting intestinal peristalsis and motility, modulating amino acid transport, and prolonging analgesia(Reference Clare and Swaisgood61, Reference Meisel62). The β-casomorphins were the first identified and are the most studied opioid agonist peptides from food proteins, with β-casomorphin-11(Reference Meisel and Frister52, Reference Meisel63) and β-casomorphin-7(Reference Svedberg, de Haas and Leimenstoll64, Reference Singh, Rosen and Chang65) being characterized as in vivo digestion products in both humans and minipigs(Reference Clare and Swaisgood61). Casomorphins have also been shown to be produced during cheese ripening due to the proteolytic activity of certain bacteria such as Pseudomonas aeroginosa and Bacillus cereus (Reference Hamel, Kielwein and Teschemacher66). Studies in rats indicate that β-casomorphins may have a role in modulating dietary fat intake, with β-casomorphin 1-7 stimulating intake of a high fat diet, and suppressing intake of a high carbohydrate diet in satiated rats(Reference Lin, Umahara and York67). Bovine β-casomorphins have also been shown to have depressive effects on the central respiratory system, causing a slowing of respiratory frequency and tidal volume in rats and rabbits(Reference Hedner and Hedner68).
In adult humans, absorption of casomorphins has not been observed, therefore it is considered that their effects are limited to the gastrointestinal tract(Reference Teschemacher, Koch and Brantl55). In infants however, there is evidence suggesting that β-casomorphins can be transported from the blood to both the brain stem(Reference Pasi, Mahler and Lansel69) and the cardiovascular compartment(Reference Teschemacher, Koch and Brantl55). Passive transport across the intestinal mucosa occurs in neonates and may provide an analgesic effect on the nervous system resulting in calmness and sleep in infants(Reference Teschemacher, Koch and Brantl55). In pregnant/lactating women only (not non-pregnant women) β-casomorphins or their precursors have been shown to pass through the mammary gland and have been detected in the plasma. Based on this, it has been proposed that β-casomorphins could potentially modulate the release of prolactin and oxytocin during lactation(Reference Clare and Swaisgood61).
The amino acid sequences of the tetrapeptides α-lactorphin and β-lactorphin differ by only one amino acid, yet their physiological effects appear to be quite different. In vitro,α-lactorphin exerts a weak inhibitory effect on contractions of guinea pig ileum, while β-lactorphin causes a non-opioid stimulatory effect(Reference Pihlanto-Leppälä58). Since both α-lactorphin and β-lactorphin are able to displace 3H-naloxone from its binding site their actions appear to be elicited via the μ-receptor. Although theoretically the digestion of 1 L of milk should give rise to sufficient lactorphin concentrations to achieve the effects observed in vitro (Reference Pihlanto-Leppälä58), hydrolysate studies have shown that the yield is actually only 5-14 % of that calculated(Reference Pihlanto-Leppälä58). This, combined with the fact that the generation of α-lactorphin and β-lactorphin during gastrointestinal digestion is yet to be proven raise questions as to whether physiological effects due to α-lactorphin and β-lactorphin derived from milk consumption are likely(Reference Pihlanto-Leppälä58).
Enzymatic digestion of wheat gluten has led to the discovery of several opioid peptides which have amino acid sequences quite different to those from other sources(Reference Fukudome and Yoshikawa70–Reference Fukudome and Yoshikawa71). Studies into the effects of two of these gluten exorphins; A5 and B5 have shown that oral and intravenous administration to rats led to a stimulation of postprandial insulin release(Reference Fukudome, Shimatsu and Suganuma72).
Haemorphins have been shown to bind and stimulate opioid receptors in a number of in vitro binding studies(Reference Brantl, Gramsch and Lottspeich56, Reference Glämsta, Marklund and Hellman73–Reference Piot, Zhao and Guillochon74). As well as exhibiting agonistic effects, the haemorphins can also act as opioid receptor antagonists(Reference Zadina, Kastin and Kersh75). Haemorphin-6 and LVV-haemorphin-6 have been shown to have a higher affinity for the δ-binding site than β-casomorphins and other opioid peptides(Reference Glämsta, Marklund and Hellman73). Haemorphins appear to have analgesic properties similar to classical opioid peptides(Reference Davis, Gillespie and Porreca76) and may also have a role in the modulation of acute inflammatory responses(Reference Nyberg, Sanderson and Glämsta77). In addition, hemorphins have been shown to have ACE inhibitory activity(Reference Nyberg, Sanderson and Glämsta77, Reference Ivanov, Karelin and Philippova78).
However, despite this body of evidence a functional role for the opioid peptides is yet to be demonstrated. One possible role for casomorphins could be to slow the passage of digesta through the gut, thus enabling maximum production of other bioactive peptides and also increasing the time for these peptides to assert their action(Reference Schanbacher, Talhouk and Murray79).
Opioid Antagonists
The number of food-derived opioid antagonists appears to be lower than opioid agonists. The casoxins: A, B, C and D are derived from bovine κ-casein and/or human αS1-casein, while lactoferroxins A, B and C are generated from human lactoferrin(Reference Clare and Swaisgood61). These antagonists tend to have preferences for the μ- and κ-type of opioid receptors(Reference Clare and Swaisgood61). There is a significant body of evidence from both animal and human studies(Reference Shahidi and Zhong80) demonstrating that ingestion of opioid antagonists leads to appetite regulation and reduced feeding by regulating insulin secretion by the pancreas(Reference Kitts81), so these opioid peptides may well have a role in preventing and/or treating obesity.
Osteoprotective proteins and peptides
Osteoporosis is a chronic, yet largely preventable disease, as bone mineralisation can be improved with sufficient supplies of soluble or available calcium. Milk is a well known source of calcium and in addition contains casein which increases the absorption of calcium in the intestine and is therefore osteoprotective. Caseinophosphopeptides (CPP), formed during the digestion of casein in the gastrointestinal tract, are also able to chelate calcium, and since this complex remains soluble there is the potential for enhanced calcium absorption across enterocytes in the distal intestine. CPP carries a large number of negative charges due to a high acidic amino acid content, and this renders the peptides resistant to further proteolysis, and perhaps explains why CPP have been detected in the intestinal contents of animals fed intact casein and purified β-casein(Reference Kitts, Yuan and Nagasawa32, Reference Kasai, Honda and Kiriyama33, Reference Meisel and Frister52, Reference Naito, Kawakami and Inamura82, Reference Sato, Noguchi and Naito83).
Results from both animal and human studies investigating the effects of casein based diets on calcium absorption have shown conflicting results. Yuan et al. (Reference Yuan and Kitts84) showed that calcium solubility, absorption and hence bioavailability was enhanced in animals fed casein based diets compared to soya based diets, however a long term study in mini-pigs only showed beneficial effects from consumption of CPP under specific conditions such as vitamin D deficiency(Reference Scholz-Ahrens, de Vrese and Barth85). Results from a trial in humans where CPP was incorporated into a rice-based infant food led to an increase in both calcium and zinc absorption, although subsequent incorporation into a whole grain infant cereal did not achieve the same result, possibly indicating matrix effects(Reference Hansen, Sandstöm and Jensen86). In a trial in post-menopausal women Narva et al. (Reference Narva, Karkkainen and Poussa87) showed there was no effect of CPP-enriched milk on several calcium and calcium related parameters.
Whey components have also been shown to have beneficial effects on bone metabolism. A fraction from whey known as milk basic protein or MBP has been shown to promote bone formation and inhibit bone resorption in both in vitro and in vivo studies. Two double-blind, placebo-controlled human studies:one carried out in 35 healthy young women average age 21·3 years(Reference Uenishi, Ishida and Toba88) and one completed in 27 healthy menopausal women average age of 50·5 years(Reference Aoe, Koyama and Toba89) both showed that MBP supplementation (40 mg/day) for a period of 6 months led to increased bone mineral density compared to the control group. Changes in the concentrations of biochemical indices of bone metabolism over these study periods, namely an increase in the bone formation marker serum osteocalcin in the younger women(Reference Uenishi, Ishida and Toba88) and decrease in the bone resorption marker urinary cross-linked N-teleopeptides of type-I collagen in both studies suggest the MBP is acting by enhancing bone formation and reducing bone resorption(Reference Uenishi, Ishida and Toba88, Reference Aoe, Koyama and Toba89). In vitro studies indicate that MBP may have a direct effect on osteoclasts thus suppressing bone resorption(Reference Toba, Takada and Yamamura90).
The potential use of bioactivity as an additional measure of protein quality: Issues to be considered
If the physiological effects elicited by the consumption of foods containing bioactive proteins and peptides are to be considered as a factor contributing to dietary protein quality then it is imperative that measurable health benefits are demonstrated, and these must be able to be quantitated in a standardised manner. Standardised methods will also be needed to determine adequate doses of the bioactive protein or peptide, the amounts of the bioactive in the food product and the safety of the product. Methods will also be needed to show that the bioactive peptide or protein can be detected in the tissue being affected, and at levels capable of eliciting a biological effect. These factors are also those likely to be required by regulatory authorities concerned both with food safety and granting permission for marketing with recognised health claims.
Demonstration of measurable health effects
One problem relating to the demonstration of measurable health effects is that many of the studies describing beneficial effects have been performed in vitro. It is therefore important that in vivo trials are conducted to assess true efficacy. In fact it has been suggested that in many instances an in vitro approach to testing is not valuable at all, one reason being that bioavailability issues due to several factors including low absorption and rates of metabolism are not accounted for(Reference Foltz, van der Pijl and Duchateau91). Another issue with in vitro testing was highlighted by Fujita et al. (Reference Fujita, Yokoyama and Yoshikawa92) when they demonstrated that some peptides reported to have ACE inhibitory activity were acting by competing with the assay's synthetic substrate rather than actually inhibiting the enzyme. Cost and ethical issues are among the reasons for carrying out in vitro studies to screen for potential bioactive peptides, however, given that the peptides actually generated by digestion may be different to those being tested in vitro it is possible that bioactive peptides might also be missed using this approach.
Another issue relating to the demonstration of measurable health benefits is that while for some effects such as cholesterol lowering and blood-pressure lowering the measures and methods are clear, this is potentially a major issue in other areas of bioactive research e.g. immunomodulatory proteins and peptides. For whilst there is a significant amount of literature(Reference Rutherfurd-Markwick and Moughan93) on the immunomodulatory effects of food-derived bioactive peptides there is considerable debate as to whether the changes in parameters such as lymphocyte proliferation, phagocytosis and natural killer cell activity observed with in vitro and ex vivo studies will in fact lead to a health benefit such as resistance to disease or infection. Some studies have demonstrated the ability of certain peptides and proteins to enhance antibody production against vaccines. However, to date there are only limited numbers of studies showing anti-infection properties of certain bioactive peptides and proteins. It has also been argued that stimulation of the immune system could actually be deleterious leading to chronic inflammation and other undesired effects. Thus the effects of bioactive peptides in terms of a measureable outcome relating to health remains unclear, which would make assessing their contribution to dietary protein quality by optimising health difficult at the current time.
Dose response and safety
With bioactive peptides being derived from foods there is the potential misconception that by extrapolation their consumption is safe. While this may well be the case if products are consumed at normal levels, there is no guarantee that the increased consumption that is highly likely to occur if the consumer believes there to be a significant health benefit, will not result in levels beyond a safe limit, which could result in deleterious effects. Trials should be performed to show the dose response and safety of long term ingestion particularly in relation to areas such as allergenicity, toxicity and mutagenicity, especially if health related claims are to be made. Ideally animal studies should be followed by trials in humans including intervention studies where appropriate.
Standardised methods are required to accurately quantify and characterise bioactive peptides from all sources, be they of plant or animal origin(Reference Gilani, Xiao and Lee46). These methods must be sensitive enough to allow detection in vivo at a relevant target site. For example if the peptide acts systemically then it must be detectable in human blood in its biologically active form in quantities sufficient to elicit a measurable response. Since mechanisms of action may differ from one peptide to another the half-life of the peptide in the blood must also be measured as a guide to both efficacy and safety. Clearly peptides with long half-lives consumed frequently could potentially run the risk of exceeding safe upper limits, while those with very short half-lives may be degraded before they can cause a response.
Methods for quantifying and characterising bioactive peptides
Potential methods for identifying food-derived peptides in blood and other biological samples as well as in food matrices are being developed. The complexity of protein hydrolysates, which can contain hundreds of peptide sequences, of which the bioactive peptide may represent only a minor constituent, combined with the complex nature of food matrices has meant that the task of identifying, characterising and quantifying food-derived bioactive peptides has been a difficult and often labour-intensive one. The advent of soft ionization mass spectrometry (MS) techniques such as fast atom bombardment (FAB), electrospray ionisation (ESI) and matrix-assisted laser desorption ionisation (MALDI) has heralded a new era for protein characterisation. MS techniques are highly selective and sensitive, with the result that they have been successfully used to not only identify and characterise, but also when combined with other techniques such as HPLC (i.e. LC/MS or LC/MS/MS), to quantify food-derived bioactive peptides in both foods and biological fluids(Reference Del Mar Contreras, Lopez-Exposito and Hernandez-Ledesma94). There are however, several potential problems inherent with MS techniques such as repeatability and matrix effects which result in suppression of ionisation, and although these can usually be overcome by the inclusion of an internal standard other methods such as ELISA or RIA are frequently used for quantitation purposes.
Since an adequate dosage of bioactive peptides is essential in order to elicit a measureable biological effect it has become important for product manufacturers to be able to accurately quantify the levels of the active peptide(s) in the final product. Various techniques for quantifying ACE-inhibitory peptides in functional foods have been described by a number of authors with detection limits as low as 0·01 ng/ml(Reference Del Mar Contreras, Lopez-Exposito and Hernandez-Ledesma94). Methods for detecting lactoferricin in biological fluids and wheat gluten-derived opioid peptides in cerebrospinal fluid are amongst others that have also been published(Reference Del Mar Contreras, Lopez-Exposito and Hernandez-Ledesma94).
Other techniques for isolating, characterising and quantifying bioactive peptides include solid-phase extraction, followed by size-exclusion chromatography and reversed-phase HPLC. The successful identification and quantitation of food-derived peptides from a fish gelatine hydrolysate has been described using the method above(Reference Sato, Iwai and Aito-Inoue95). In this particular example two volunteers consumed 20 g of the hydrolysate and collagen peptides were detected in the plasma 60 min later. Interestingly, both the quantity and the structure of the identified peptides differed between the individuals suggesting that the digestive and/or absorption processes differed. This finding reinforces the importance of identifying the amino acid sequences of bioactive peptides and determining optimum levels for eliciting beneficial effects and the upper levels for safety, as clearly depending on each individuals' metabolism, the concentration of bioactive peptides found in the blood may vary markedly from one individual to another.
Conclusions
The area of food bioactive proteins and peptides is a rapidly growing field of study and no doubt more bioactive peptides and proteins along with additional physiological effects will be discovered in the future. A number of factors can potentially impact on the ability of a bioactive peptide or protein to elicit an effect. The amounts of bioactive peptides generated must be sufficient to elicit a response, but should not be so high as to be unsafe, thus causing negative effects. To do this, methods must be developed and standardised that enable not only the measurement of health benefits but also of the level of bioactive peptides which are absorbed into the bloodstream. Once standardised, such methods may provide a new perspective and an additional mechanism for describing dietary protein quality which is currently not encompassed by the use of measures such as protein digestibility-corrected amino acid score (PDCAAS). This would take the measure of protein quality beyond one of nitrogen balance and more towards optimum health, encompassing the specific physiological roles of bioactive proteins and peptides as well the key regulatory roles of certain amino acids. In addition such information gained using standardised techniques will prove invaluable to regulatory authorities and should aid in the development of regulations and policy regarding acceptable scientific validation of efficacy and safety claims(Reference Gilani, Xiao and Lee46).
Acknowledgements
The author has no conflict of interest to declare. K.R.M wrote the paper and has sole responsibility for the content. This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.
