**4. Current view of protein stability evaluation in different fields**

The common approaches to the determination of the protein and peptide digestibility in different areas of research varies significantly:


Methods of Protein Digestive Stability Assay – State of the Art 225

and some other, which are completely degraded in the experimental conditions, still induce immune response *in vivo* (Faeste et al. 2007; Fu 2002; Jensen-Jarolim & Untersmayr 2006).

One of the complicated moments during the immune response is that depending on the type of T-cell, the type of antibodies produced by the B-cell could be changed: Th2 cells produce a cocktail of cytokines that, among other actions, encourage plasma cells to switch to synthesis of IgE, while Th1 cells typically produce interleukin 2, interferon gamma, and tumor necrosis factor (TNF)-alpha/beta, that together downregulate IgE synthesis. Only one type of stimulation could be active for a given antigen, as Th1 cytokines suppress Th2 cell development, and *vice versa*. Structural investigations of several allergens have identified T epitopes implicated in the selective development of Th2-type lymphocytes. Some allergens, such as ovalbumin, have only a few epitopes that preferentially induce a Th2 response, whereas others, such as beta-lactoglobulin, have many. The potential to induce a polarized (Th1 or Th2) response is not exclusively an intrinsic property of particular epitopes; the quality of the T-cell response to any given epitope depends upon many factors, including the dose of protein or peptide given, the affinity of the peptide for MHC class II receptor, and the longevity of the class II/peptide complex. Together, these parameters act to influence the dynamic epitope density on the surface of APCs. It is generally agreed that parameters that serve to increase ligand density, including greater affinity to MHC class II, higher levels of ligand loading, or longer-lived complexes favor Th1-type responses (Huby

In addition to the possible anti-allergic action of Th1 epitopes, a significant variety into experimental results is introduced due to the experimental conditions differences. Several investigators term protein as «stable» after incubation times ranging from 8 to 60 min (Bannon et al. 2003; Fu 2002; Herman et al. 2006). As far as the unstable proteins usually survive less than 15 sec in the similar assay conditions, such conclusion seems to be quite correct. However, the 8-min survival could potentially be an indication of the protein instability, as the simplified conditions of simulated gastric and intestinal fluid do not take into account intestinal mucosa peptidases or minor luminal endopeptidases. Then, albeit standardized in the last period, the enzyme to substrate ratios vary between different research groups (Herman et al. 2006), and thus obtained stability or instability may not

Finally, the conditions inside the digestive tract during the potential allergen ingestion are of crucial importance. For example, during inflammation the permeability of digestive tract is increased (Perrier & Corthésy 2011); then, any elevation of the pH should result in hindrance of peptic degradation. Numerous situations with elevated stomach pH are known, e.g. in early childhood, in elderly, or in chronic atrophic gastritis. Moreover, there is a number of pathologies, like gastritis or ulcer, where acid neutralization or inhibition is an important therapeutic goal. Acid neutralization is state of the art during surgical care, corticosteroid or analgesic treatment. Moreover, anti-acids, H2-receptor blockers and proton pump inhibitors are increasingly consumed without prescriptions due to liberalization of

As a logical consequence of the aforementioned complications, immunogenicity assessment is typically done using a multistage strategy, generally including an digestive stability and antibody affinity screening assay, a specificity confirmation step, and in some cases a

the market by over-the-counter sale (Jensen-Jarolim & Untersmayr 2006).

et al. 2000).

adequately reflect the reality.


In fact, the stability towards protease degradation is not considered to be a basic method for the evaluation of potential protein allergenicity, albeit the resistant proteins are usually allergenic (Fu 2002). To better understand the data on the digestibility of any potential immunogenic peptide, a basic picture of the underlying mechanisms is necessary (for more details reader is referred to an excellent review by Huby et al. (Huby et al. 2000)).

Allergic sensitization to proteins involves the induction of IgE type antibodies production of sufficient magnitude by differentiated B-lymphocytes to facilitate the elicitation of an inflammatory reaction following subsequent exposure to the same (or a cross-reactive) allergen. For a B cell to differentiate into a plasma cell and produce antibodies, the B-cell receptor expressed on its surface must bind to specific B epitopes on the surface of the protein antigen. The antibodies subsequently produced by the plasma cell, into which the B cell differentiates, have the same specificity as the B-cell receptor, and are therefore able to bind specifically to the same B epitopes on the surface of the protein antigen. Efficient secretion of antibodies normally requires that the B cells receive help from T-helper cells that specifically recognize separate epitopes on the same protein antigen. Such recognition is mediated by the T-cell antigen receptor, which delivers stimulatory intracellular signals to the T cell. Allergenic proteins *per se* cannot be recognized by T cells; the proteins must first be processed and then presented by specialized antigen-presenting cells (APCs) (Gorovits 2010; Huby et al. 2000).

After the first contact with an antigen the T-helper-stimulated B-cells differentiate and become ready to produce antibodies. Thus, if such antigen is encountered for the second time, there will be some antibodies against it present in the blood flow, and the production of their large quantities will start immediately. The IgE antibodies, however, serve mainly as information transducers during allergic response. After antigen recognition they bind to the Fc receptors for IgE (FceR) on the surface of basophils and mast cells; for these cells to become activated and secrete histamine and other inflammatory mediators their receptors need to dimerize. Thus, the peptide is to be large enough to bind at least two antibodies, which will bring two FceR together. The minimal length of peptide chain, which could be recognized by IgE antibody (epitope), is 15 amino acids, and thus the minimum size of an immunogenic peptide should be around 30 amino acids or 5-6 kDa (Van Beresteijn et al. 1994). At the same time, the sensitization stage, when an antigen is recognized by T-cells and undifferentiated B-cells, requires only one epitope and thus a peptide size of around 3 kDa (Moreno 2007; Van Beresteijn et al. 1994).

The discussion above clearly indicates the valid conditions of digest products analysis: HPLC and MALDI are always informative enough, but SDS-PAGE is appropriate only if it includes molecular masses of 2-3 kDa, which unfortunately is not always the case (Huang et al. 2010; Yagami et al. 2000; Yu et al. 2011).

Even after all the requirements for the stability assay of a potential allergen are met, the results should be interpreted with caution, as some stable proteins are not allergens at all,

The studies on the properties of digestive enzymes rely on the purified enzymes or on

 Pharmacological studies tend to use complicated devices for the evaluation of the ability of a particular pharmacological form to degrade in specific compartment of

In fact, the stability towards protease degradation is not considered to be a basic method for the evaluation of potential protein allergenicity, albeit the resistant proteins are usually allergenic (Fu 2002). To better understand the data on the digestibility of any potential immunogenic peptide, a basic picture of the underlying mechanisms is necessary (for more

Allergic sensitization to proteins involves the induction of IgE type antibodies production of sufficient magnitude by differentiated B-lymphocytes to facilitate the elicitation of an inflammatory reaction following subsequent exposure to the same (or a cross-reactive) allergen. For a B cell to differentiate into a plasma cell and produce antibodies, the B-cell receptor expressed on its surface must bind to specific B epitopes on the surface of the protein antigen. The antibodies subsequently produced by the plasma cell, into which the B cell differentiates, have the same specificity as the B-cell receptor, and are therefore able to bind specifically to the same B epitopes on the surface of the protein antigen. Efficient secretion of antibodies normally requires that the B cells receive help from T-helper cells that specifically recognize separate epitopes on the same protein antigen. Such recognition is mediated by the T-cell antigen receptor, which delivers stimulatory intracellular signals to the T cell. Allergenic proteins *per se* cannot be recognized by T cells; the proteins must first be processed and then presented by specialized antigen-presenting cells (APCs) (Gorovits

After the first contact with an antigen the T-helper-stimulated B-cells differentiate and become ready to produce antibodies. Thus, if such antigen is encountered for the second time, there will be some antibodies against it present in the blood flow, and the production of their large quantities will start immediately. The IgE antibodies, however, serve mainly as information transducers during allergic response. After antigen recognition they bind to the Fc receptors for IgE (FceR) on the surface of basophils and mast cells; for these cells to become activated and secrete histamine and other inflammatory mediators their receptors need to dimerize. Thus, the peptide is to be large enough to bind at least two antibodies, which will bring two FceR together. The minimal length of peptide chain, which could be recognized by IgE antibody (epitope), is 15 amino acids, and thus the minimum size of an immunogenic peptide should be around 30 amino acids or 5-6 kDa (Van Beresteijn et al. 1994). At the same time, the sensitization stage, when an antigen is recognized by T-cells and undifferentiated B-cells, requires only one epitope and thus a peptide size of around 3

The discussion above clearly indicates the valid conditions of digest products analysis: HPLC and MALDI are always informative enough, but SDS-PAGE is appropriate only if it includes molecular masses of 2-3 kDa, which unfortunately is not always the case (Huang et

Even after all the requirements for the stability assay of a potential allergen are met, the results should be interpreted with caution, as some stable proteins are not allergens at all,

details reader is referred to an excellent review by Huby et al. (Huby et al. 2000)).

fractionated homogenates of the parts of gastrointestinal tract.

digestive tract.

2010; Huby et al. 2000).

kDa (Moreno 2007; Van Beresteijn et al. 1994).

al. 2010; Yagami et al. 2000; Yu et al. 2011).

and some other, which are completely degraded in the experimental conditions, still induce immune response *in vivo* (Faeste et al. 2007; Fu 2002; Jensen-Jarolim & Untersmayr 2006).

One of the complicated moments during the immune response is that depending on the type of T-cell, the type of antibodies produced by the B-cell could be changed: Th2 cells produce a cocktail of cytokines that, among other actions, encourage plasma cells to switch to synthesis of IgE, while Th1 cells typically produce interleukin 2, interferon gamma, and tumor necrosis factor (TNF)-alpha/beta, that together downregulate IgE synthesis. Only one type of stimulation could be active for a given antigen, as Th1 cytokines suppress Th2 cell development, and *vice versa*. Structural investigations of several allergens have identified T epitopes implicated in the selective development of Th2-type lymphocytes. Some allergens, such as ovalbumin, have only a few epitopes that preferentially induce a Th2 response, whereas others, such as beta-lactoglobulin, have many. The potential to induce a polarized (Th1 or Th2) response is not exclusively an intrinsic property of particular epitopes; the quality of the T-cell response to any given epitope depends upon many factors, including the dose of protein or peptide given, the affinity of the peptide for MHC class II receptor, and the longevity of the class II/peptide complex. Together, these parameters act to influence the dynamic epitope density on the surface of APCs. It is generally agreed that parameters that serve to increase ligand density, including greater affinity to MHC class II, higher levels of ligand loading, or longer-lived complexes favor Th1-type responses (Huby et al. 2000).

In addition to the possible anti-allergic action of Th1 epitopes, a significant variety into experimental results is introduced due to the experimental conditions differences. Several investigators term protein as «stable» after incubation times ranging from 8 to 60 min (Bannon et al. 2003; Fu 2002; Herman et al. 2006). As far as the unstable proteins usually survive less than 15 sec in the similar assay conditions, such conclusion seems to be quite correct. However, the 8-min survival could potentially be an indication of the protein instability, as the simplified conditions of simulated gastric and intestinal fluid do not take into account intestinal mucosa peptidases or minor luminal endopeptidases. Then, albeit standardized in the last period, the enzyme to substrate ratios vary between different research groups (Herman et al. 2006), and thus obtained stability or instability may not adequately reflect the reality.

Finally, the conditions inside the digestive tract during the potential allergen ingestion are of crucial importance. For example, during inflammation the permeability of digestive tract is increased (Perrier & Corthésy 2011); then, any elevation of the pH should result in hindrance of peptic degradation. Numerous situations with elevated stomach pH are known, e.g. in early childhood, in elderly, or in chronic atrophic gastritis. Moreover, there is a number of pathologies, like gastritis or ulcer, where acid neutralization or inhibition is an important therapeutic goal. Acid neutralization is state of the art during surgical care, corticosteroid or analgesic treatment. Moreover, anti-acids, H2-receptor blockers and proton pump inhibitors are increasingly consumed without prescriptions due to liberalization of the market by over-the-counter sale (Jensen-Jarolim & Untersmayr 2006).

As a logical consequence of the aforementioned complications, immunogenicity assessment is typically done using a multistage strategy, generally including an digestive stability and antibody affinity screening assay, a specificity confirmation step, and in some cases a

Methods of Protein Digestive Stability Assay – State of the Art 227

conditions Reference

2004)

2001)

2010)

2011)

(Akimov et al.

(Vardakou et al.

(Dhuique-Mayer et al. 2007)

(Bender 1958)

(Mokady et al.

(Darragh & Hodgkinson 2000; Faber et al. 2010)

(Foltz et al. 2009)

1969)

Hours (Thomas et al.

Hours (Tonglet et al.

Hours (Hansen 1975)

Hours (Curtis et al. 1978)

Hours; HPLC or MALDI

Hours + cell culture

cell culture facilities

Hours; requires

isotope labeling

microbiological facilities

Days-weeks; may require

Days(cannula); hours(probe); applicable for human

Minutes (computation), weeks to months (database and model development); requires large experimental body (data

on 100-200 proteins)

Table 2. A comparison of different method classes for protein digestibility evaluation

preparation; requires animal

Hours (single compartment) to days (whole digestive tract); requires a special device

analysis

Name Reproduced conditions Time to complete and special

Stomach lumen without mechanical forces after individual ingestion

Intestinal lumen without shed mucosal cells and mechanical forces after individual ingestion

a source of essential amino acids

intracellular digestion

intracellular intestinal digestion, transport

intracellular digestion of a given organ, transport

large intestine; ability to provide essential amino

in a given health condition

the point of probing without data on digestion products

any conditions depending on model

Luminal digestion with mechanical forces ant transitional times without mucosal enzymes

Amino acid score Ability of protein to be

Organ fragments Mucosal and

Caco-2 cell culture Luminal and

Isolated organs Luminal, mucosal and

Bacterial culture Microbial digestion in

acids

Nitrogen balance Whole digestion process

Cannula/probe Digestive process up to

QSAR Predicts degradation in

*In vitro* methods Simulated gastric

fluid

Simulated intestinal fluid

A device with immobilized enzymes

*Ex vivo* methods

*In vivo* methods

*In silico* methods

characterization step. Samples are first evaluated in a screening assay, for which an assay threshold has been set based on the variability of samples from a drug naïve target patient population. Screen positive samples are further evaluated in a confirmatory assay to verify whether the signal observed in the screening assay is a result of a specific response to the protein therapeutic treatment. Confirmed positive samples are then put into downstream methods for sequential characterization based on the comprehensive consideration of immunogenicity risk assessment and mechanism of action for the protein therapeutic (Gorovits 2010; Peng et al. 2011).

The studies of protein nutritive value usually rely on the *in vivo* experiments or on the data on protein amino acid composition. The point is that the most important aim of such studies is to tell, whether a given protein (perhaps after some processing) is suitable for supplying body needs for energy and essential amino acids. The question on how does an organism manage to consume that protein is often ignored, albeit it is tempting to speculate that the exact metabolic route from the whole protein to the amino acids and energy inside body cells could be important as well. One should note that as it was shown in section 2, there are several conceptually different steps in protein digestion (denaturation and splicing in large chunks in stomach, splicing in tetra- to dipeptides and individual amino acids in small intestine and various processing by the large intestine microbiota), whose efficiency could be independently affected by the protein structure and modifications, and thus during some pathological conditions a protein of a high nutritive value could suddenly lose it.

The *in vivo* evaluation of nutritive value consists of animal feeding with test protein with subsequent analysis of either the amount of absorbed or unabsorbed protein fractions, or the degree to which such meal is able to sustain animal growth. The experiments are usually quite long, expensive and involve complex manipulations with animals. The direct determination of protein amino acid composition is much simpler. The only complication of such approach is that the degree to which the constituent amino acids of a food protein are actually available to the body is determined by such factors as protein configuration, amino acid bonding, other constituents of the diet, and the physiological condition of the gastrointestinal tract of the individual person or animal involved. However, the correlation between prediction and performance is quite good: prediction for poor performance of proteins devoid or nearly devoid of an essential amino acid as well as of good performance for proteins containing all essential amino acids according to idealized patterns is excellent. However, fine-line predictions of intermediate quality are less accurate, and surprises are not uncommon (Kies 1981). Thus the *in vivo* models are still a preferred way for nutritive value analysis.

Enzymatic studies are usually concerned not with the stability of a some protein, but rather with the distribution, specificity and kinetics of a particular enzyme, and thus they usually implement some procedures of purification from animal tissues (Krzysik & Adibi 1977; Smith & Bergmann 1944; Sterchi & Woodley 1980a) or of expression of a recombinant enzyme in some prokaryotic or eukaryotic cells (Hauri et al. 1985; Lentze 1995). The applicability of such procedures is usually quite legitimate, albeit the biological significance and theoretical value of the obtained results could vary. The enzyme-to-substrate ratios in such studies usually do not resemble the *in vivo* situation, and thus the real significance of a given enzyme for the digestive process could not always be predicted. The differences of purification procedures complicate the integration of data from various research groups into

characterization step. Samples are first evaluated in a screening assay, for which an assay threshold has been set based on the variability of samples from a drug naïve target patient population. Screen positive samples are further evaluated in a confirmatory assay to verify whether the signal observed in the screening assay is a result of a specific response to the protein therapeutic treatment. Confirmed positive samples are then put into downstream methods for sequential characterization based on the comprehensive consideration of immunogenicity risk assessment and mechanism of action for the protein therapeutic

The studies of protein nutritive value usually rely on the *in vivo* experiments or on the data on protein amino acid composition. The point is that the most important aim of such studies is to tell, whether a given protein (perhaps after some processing) is suitable for supplying body needs for energy and essential amino acids. The question on how does an organism manage to consume that protein is often ignored, albeit it is tempting to speculate that the exact metabolic route from the whole protein to the amino acids and energy inside body cells could be important as well. One should note that as it was shown in section 2, there are several conceptually different steps in protein digestion (denaturation and splicing in large chunks in stomach, splicing in tetra- to dipeptides and individual amino acids in small intestine and various processing by the large intestine microbiota), whose efficiency could be independently affected by the protein structure and modifications, and thus during some

pathological conditions a protein of a high nutritive value could suddenly lose it.

The *in vivo* evaluation of nutritive value consists of animal feeding with test protein with subsequent analysis of either the amount of absorbed or unabsorbed protein fractions, or the degree to which such meal is able to sustain animal growth. The experiments are usually quite long, expensive and involve complex manipulations with animals. The direct determination of protein amino acid composition is much simpler. The only complication of such approach is that the degree to which the constituent amino acids of a food protein are actually available to the body is determined by such factors as protein configuration, amino acid bonding, other constituents of the diet, and the physiological condition of the gastrointestinal tract of the individual person or animal involved. However, the correlation between prediction and performance is quite good: prediction for poor performance of proteins devoid or nearly devoid of an essential amino acid as well as of good performance for proteins containing all essential amino acids according to idealized patterns is excellent. However, fine-line predictions of intermediate quality are less accurate, and surprises are not uncommon (Kies 1981). Thus the *in vivo* models are still a preferred way for nutritive

Enzymatic studies are usually concerned not with the stability of a some protein, but rather with the distribution, specificity and kinetics of a particular enzyme, and thus they usually implement some procedures of purification from animal tissues (Krzysik & Adibi 1977; Smith & Bergmann 1944; Sterchi & Woodley 1980a) or of expression of a recombinant enzyme in some prokaryotic or eukaryotic cells (Hauri et al. 1985; Lentze 1995). The applicability of such procedures is usually quite legitimate, albeit the biological significance and theoretical value of the obtained results could vary. The enzyme-to-substrate ratios in such studies usually do not resemble the *in vivo* situation, and thus the real significance of a given enzyme for the digestive process could not always be predicted. The differences of purification procedures complicate the integration of data from various research groups into

(Gorovits 2010; Peng et al. 2011).

value analysis.


Table 2. A comparison of different method classes for protein digestibility evaluation

Methods of Protein Digestive Stability Assay – State of the Art 229

utilize these proteins. However, a vast array of preliminary data is required for optimization

Thus, when considering various methods of protein digestibility assay, one should first of all define the limitations of the research at hand. If a high throughput screening is required? If there is a need for any data on digestion products? If absorption efficiency could vary and is essential? If a protein should serve as food or not? A possible decision tree for method optimization is presented on figure 1, albeit it should be considered more as a reference than

and validation of such models, and so they are relatively rare.

Fig. 1. A decision tree for protein or peptide digestibility assay choice

6, (November 2010), p. 690, ISSN 0132-3423

Adibi, S.A., (1976). Intestinal phase of protein assimilation in man. *American Journal of Clinical Nutrition*, Vol. 29, No. 2, (February 1976), pp. 205-215, ISSN 0002-9165 Akimov, M.G., Nazimov, I.V., Gretskaya, N.M., Deigin, V.I. & Bezuglov, V.V. (2010).

Investigation of peptide stability upon hydrolysis by of fragments of the organs of the gastrointestinal tract of rats. *Russian Journal of Bioorganic Chemistry*, Vol. 36, No.

a rule.

**6. References** 

one picture (*cf.,* for example, the discussion in the work by Sterchi et al. (Sterchi & Woodley 1980a)).

The aim of a substantial part of pharmacological studies of peptide- and protein-based therapeutics is often the evaluation of the disintegration, dissolution and drug release profiles of oral drug formulations (Vardakou et al. 2011). Thus another aspect of food digestion becomes important – the mechanical grinding and mixing forces, exerted on a pharmaceutical formulation by various digestive tract compartments. To simulate such conditions devices of various complexity are employed (Schasteen et al. 2002; Vardakou et al. 2011). The applicability of such devices for the drug form disintegration testing is unquestionable, however, usually they reproduce only the luminal phase of food digestion. Thus the evaluation of the ability of released protein or peptide to get to the blood flow intact requires additional methods.

A short reference of available method types is provided in table 2. It should be noted, that some *in vitro* methods could be combined and expanded to provide more relevant data. For example, single-enzyme digestions by stomach and small intestine could be performed sequentially, the Caco-2 cell culture is often overlaid by a simulated gastric fluid and possibly mucin (to protect cells and imitate the similar layer present *in vivo*). The computational methods are the most adaptive and powerful ones, however, to date their usability is limited to large screening studies due to huge preparatory experimental work.
