This document supersedes the "Recommendations" dated May 1990. No substantive changes have been made

A. INTRODUCTION

Enzyme preparations are produced in varying degrees of purity from animal, plant, and microbial sources. They may consist of whole killed cells, parts of cells, or cell-free extracts. They may also contain carriers, solvents, preservatives, and antioxidants. The enzyme preparations may be formulated as liquid, semi-liquid, or dry solid preparations. Food enzyme preparations have traditionally been added directly to food during processing. For many applications, the components of the preparation remain in the processed food product. In recent years, enzymes immobilized on solid supports have gained importance. Immobilized enzyme preparations may range from those that contain a highly specific, purified enzyme, to those that contain whole killed cells or structurally intact viable cells. For some enzymatic processes, co-immobilization of enzymes and cells may be advantageous. Immobilized enzyme preparations are not intended to become food components.

Enzyme preparations are regulated either as "secondary" direct food additives under Title 21 of the Code of Federal Regulations (CFR), Part 173, or are affirmed as GRAS (Generally Recognized As Safe) substances in 21 CFR Part 184. The regulatory status of food additives or substances affirmed as GRAS is established through the petition process. According to Section 409(b)(1) of the Federal Food, Drug, and Cosmetic Act (the Act), anyone may file a petition proposing the issuance of a regulation. Section 409(b)(2) of the Act prescribes the statutory requirements for food additive petitions. The requirements for food additive petitions are discussed in greater detail in 21 CFR 171.1.

The Act does not provide specific statutory requirements for GRAS affirmation petitions. Although it exempts substances generally recognized as safe from the term "food additive," it states that general recognition of safety must be demonstrated through scientific procedures or common use of the ingredient in food prior to January 1, 1958 (Section 201(s)). The eligibility requirements for classification of a substance as GRAS are described in 21 CFR 170.30 and the requirements for GRAS affirmation petitions are set forward in 21 CFR 170.35. ⁽¹⁾

The recommendations contained in this document are intended to aid petitioners in assembling the chemical and technological data currently considered appropriate for a food additive or GRAS affirmation petition for an enzyme preparation. They cover data needs in the following areas: identity, manufacturing process, purity, use, analytical methodologies, technical effects, and probable human exposure. The recommendations do not address other data needs, such as those pertaining to microbiological, toxicological and environmental considerations. The recommendations will be updated as needed to reflect new developments in the manufacture and use of food enzyme preparations.

Not all petitions will need to be supported by the same quantity and quality of data. The data and information needs outlined below should not be regarded as absolute. Questions regarding the tailoring of the data package for specific cases may be discussed with agency personnel⁽¹⁾ prior to submitting a petition.

B. IDENTITY

In the "Identity" section of a petition, the enzyme preparation should be characterized as completely as possible. The characterization should cover the source of the enzyme preparation, characteristics pertaining to the enzyme itself such as chemical and common name, classification, and chemical and physical properties, as well as characteristics pertaining to the whole enzyme preparation such as composition and purity. The full list of items to be considered under "Identity" is given below in the preferred order in which they should appear in a petition.

1. Source of the enzyme preparation should be identified.

2. Chemical name and EC (Enzyme Commission) classification should be given according to the "Recommendations of the Nomenclature Committee of International Union of Biochemistry" (current edition).

3. Common and/or trade names should be provided.

4. Chemical Abstract Service (CAS) Registry Number should be given where possible. CAS Registry Numbers may be obtained by contacting CAS Client Services, Box 3343, 2540 Olentangy River Road, Columbus, Ohio 43210-0334, U.S.A.

5. Properties. Enzymatic function, substrate specificity , and mode of action should be described. Other properties, such as molecular weight, isoelectric point, kinetic properties, and specific activity should be included, if available or necessary to support the enzyme identity. Enzymatic activity of the enzyme preparation should be measured and a unit of activity defined. Enzymatic activity should be characterized further with respect to, for example, temperature, pH, and inorganic ions, if such characteristics are relevant to the intended use of the enzyme preparation. Permanent or transient structural modifications intentionally introduced into the enzyme protein by chemical or genetic means in order, for example, to improve stability or enzymatic activity, or to facilitate purification or binding to a support, should also be described.

   The scope of properties to be included in a petition may have to be broader if, for example, an enzyme normally produced in one organism were expressed in another organism through molecular cloning. Characterization of the enzyme would be especially important in the case where DNA coding for the enzyme has been transferred between genetically distant organisms, e.g., from mammalian species into bacteria. Because of differences between eukaryotes and prokaryotes in certain molecular mechanisms, the recombinant eukaryotic enzyme can differ from the original one. Therefore, physico-chemical and functional characteristics of the recombinant enzyme should be thoroughly compared with those of its original counterpart. Comparison of the two enzymes should be based on at least several characteristics such as enzymatic activity, kinetic parameters, amino acid and amino sugar composition, full or partial amino acid sequence, molecular weight, isoelectric point, and gel-migration, chromatographic or other similar properties. If any differences are identified their significance should be assessed.

6. Composition. Food enzyme preparations are usually complex mixtures. To characterize the enzyme preparation the following information should be provided:
   
   a) Content of proteins and nucleic acids or, alternatively, proteins plus nucleic acids (i.e., Kjeldahl protein), carbohydrates, fats, water, total solids, and ash measured for at least five representative batches of the commercial-strength preparation, as well as for the preparation used in toxicological studies. Diluents, carriers, and stabilizers should be identified.

   b) The percentage of the enzyme preparation that is enzyme protein.

   c) Information on other major enzymatic activities that may be present in the preparation.

   e) Total Organic Solids (TOS) content of the commercial preparation and of that used in the toxicological studies.

   TOS is the sum of the organic compounds in the preparation that originated from the animal, plant or microbial substrate from which the enzyme preparation was obtained ("The 1978 Enzyme Survey Summarized Data," NRC/NAS, U.S. Department of Commerce, NTIS, 1981). TOS should be calculated as follows:

   TOS (%) = 100 - A - W - D
   where
   A = % ash
   W = % water
   D = % diluents or carriers

   For immobilized enzyme preparations all components should be listed along with their percentages on a dry-solids basis. The enzymatic component should be given as TOS and characterized with respect to its composition as described in the preceding paragraph.

C. MANUFACTURING PROCESS

The source material from which the enzyme preparation has been obtained should be identified and characterized. It should comply with the General Requirements for enzyme preparations set forth in the Food Chemicals Codex, 3rd (1981) edition. If a microorganism is used as a source, it should be taxonomically and genetically identified. If the microorganism was modified by introducing DNA from other sources, these sources should also be identified and taxonomically and genetically characterized, if appropriate. All procedures and steps involved in the construction of the production microorganism should be thoroughly described. The production microorganism should be characterized with respect to the introduced DNA, genetic stability, and growth properties. The fermentation process should be described, including all steps and controls necessary to maintain the proper growth conditions, cultural purity, and genetic stability. All components of the fermentation medium should be identified. The isolation of the enzyme from the cellular material or from the growth medium (depending on whether the enzyme is intracellular or excreted) should also be described, including all chemical and physical treatments and quality controls.

If an enzyme preparation is obtained from animal or plant materials, including tissue culture, the source materials should be characterized and the procedures for enzyme isolation and purification should be thoroughly described.

For immobilized enzyme preparations, the immobilizing agent should be characterized with respect to its composition, purity and chemical and physical properties, and the method of immobilization should be provided.

D. PURITY AND FOOD-GRADE SPECIFICATIONS

The commercial enzyme preparation should conform at a minimum to the Food Chemicals Codex specifications for enzyme preparations. Conformance should be demonstrated by analyzing at least five production batches of the enzyme preparation. Impurities specific to the source of the enzyme or to the manufacturing process should also be identified and measured. Should a limit upon such impurities be required, it will be incorporated into the food-grade specifications. New or modified specifications may also be proposed.

Enzyme preparations obtained from microbial sources should not contain antibiotics, toxins (e.g., enterotoxins, mycotoxins) and transformable DNA coding for toxins and/or proteins that inactivate therapeutic antibiotics. In those instances where there may be uncertainty as to the presence of these substances, appropriate tests should be performed to demonstrate that they are not present at biologically significant levels.

Immobilized enzyme preparations must be prepared with fixing agents that are either listed in 21 CFR 173.357(a)(2) or are generally recognized as safe in food. Other substances must be evaluated and regulated through the petition process.

E. INTENDED USE AND USE LEVELS

F. RESIDUES OF IMMOBILIZED ENZYMES

This approach is used for estimating migration of various components of food-contact materials (e.g., polymeric packaging) into food and may be useful in evaluating migration of some components of immobilized enzyme preparations into food⁽²⁾. The following food-simulating solvents are recommended: 8% ethanol to simulate aqueous, acidic, and alcoholic foods, and 50% ethanol to simulate fatty foods. Extractions may be conducted using an experimental laboratory-scale column, provided the column is equivalent in terms of the operating conditions to the industrial-scale column. To assure this equivalency, the flow rate of solvent in the experimental column expressed as space velocity (SV) should be identical to that used in the industrial-scale column⁽³⁾. The solvent should be passed through the column at the highest temperature anticipated for use of the enzyme preparation. The migration pattern of the enzyme preparation into the solvent may vary depending on the type of the preparation. To establish this pattern we recommend passing the solvent through the column ("flow-through" system) for 10 days. Samples for analyses should be taken in three replicates on days 1, 3, 5, 7, and 10 and analyzed for total nonvolatile extractives (TNEs) and, if appropriate, for individual migrants. Most likely, the concentrations of migrants in the solvent will change initially and then stabilize during the later stages of the experiment. The stabilized values are assumed to remain unchanged over the enzyme's lifetime and should be used for estimation of human exposure to these migrants.

Alternatively, a measured volume of the solvent may be circulated through the column in a closed system. In this approach, a smaller volume of the solvent may be used as compared with the volume used in the equivalent "flow-through" system and, therefore, the extracted impurities would be more concentrated and measured more accurately. Two migration patterns are likely to emerge in the closed system. In one, the concentration of a migrant would initially increase, then plateau and remain unchanged until the experiment is completed. This pattern would correspond in the equivalent "flow-through" system to an initial surge of migration followed by a drop in migration to an undetectable level. Because the initial migration would affect only a small fraction of the treated food it may be considered negligible and the concentration of the migrant in food may be assumed to be below the analytical detection limit. In the second pattern, the concentration of a migrant would be likely to increase initially in a nonlinear fashion followed by a linear increase. The linear increase would manifest itself in the equivalent "flow-through" system as a constant concentration. From the linear increase of migrant concentration in the closed system, the migration rate of the migrant can be calculated. This migration rate is assumed to remain constant over the enzyme's lifetime and can be used to calculate the concentration of the migrant in the food that would be passed through the column in the equivalent "flow-through" system. An example of the calculation is given in the Appendix. In instances of uncertainty, the petitioner should submit a protocol for extraction studies for our evaluation before undertaking the actual measurements.

G. TECHNICAL EFFECTS

The method or methods for using the enzyme preparation in food should be described and the effects of the enzyme on food evaluated. It is essential to provide evidence demonstrating that the enzyme performs its intended function in food. The minimum level of the enzyme preparation required to accomplish the intended technical effect should be determined or estimated. For an immobilized enzyme preparation, the usable lifetime of the preparation at the conditions of use should be given, as well as the amount of food processed per kilogram of the enzyme preparation over its usable lifetime.

Any undesirable effects of the enzyme preparation on food, as well as factors that may affect the functional properties of the enzyme in food should be discussed.

H. ANALYTICAL METHODOLOGIES

All methods used to establish the identity and to evaluate the purity of the enzyme preparation should be described. However, standard analytical methods, such as those in the Food Chemicals Codex for specification tests, need only be referenced. Descriptions of methods used to analyze food for the presence of the enzyme and the components of the enzyme preparation (see E. "INTENDED USE AND USE LEVELS" and F. "RESIDUES OF IMMOBILIZED ENZYMES") should also be provided, including the principles, procedures, and the equipment used. Validation tests should be performed for these methods where appropriate, for example, in analyses for migrants from immobilized enzyme preparations.

Validation tests are typically performed using three samples, each in triplicate, of food or a food simulant, each fortified ("spiked") with the substance to be analyzed (e.g., fixing agent expected to migrate into food) at a different level. The recommended spiking levels are one-half of, equal to, and two times the levels of the component of the enzyme preparation expected to be found in food. Percent recovery should be calculated using the formula:

((b-a)/c) x 100.

where "a" is the measured level in the unspiked solution, "b" is the measured level in the spiked solution and "c" is the spiking level. Percent recoveries of the added substance should range between 60% and 110% for levels below 0.1 ppm and between 80% and 110% for levels above 0.1 ppm.

All results should be presented in a clear and concise manner and accompanied by all raw data, including validation data and examples of instrument recordings. Calculations and computational methods should be adequately described.

I. EXPOSURE

The petition should contain an estimate on a chronic daily basis of the probable human exposure to the enzyme preparation, i.e., an estimated daily intake (EDI). The estimate should be based preferably on actual levels, or, alternatively, on use levels (for the "worst-case" exposure) of the enzyme preparation expressed as TOS (see E. "INTENDED USE AND USE LEVELS") and on food intake information. Sources of food intake data should be referenced. Both average and high-intake (preferably 90th percentile) EDIs should be determined.

For immobilized enzyme preparations, EDIs for the residual enzymatic component as TOS and for other potential migrants, including contaminants (such as molecules of a monomer in a polymeric support), should be provided (see F. "RESIDUES OF IMMOBILIZED ENZYMES").

APPENDIX

ESTIMATION OF IMMOBILIZED-ENZYME RESIDUES IN FOOD BY EXTRACTION WITH A FOOD-SIMULATING SOLVENT.

The purpose of the experiment is to establish the migration pattern of the immobilized-enzyme components into a food-simulating solvent and use this pattern to estimate the concentrations of the migrants in food. The experiment is performed with a laboratory-scale column operationally equivalent to an industrial column. The solvent is circulated through the column in a closed system for ten days. Samples of the solvent are withdrawn at days 1, 3, 5, 7, and 10 and analyzed with suitable analytical methodologies for total nonvolatile extractives (TNEs) and individual migrants. Make-up solvent should be added in place of that withdrawn and appropriate corrections in the measured concentrations of migrants should be introduced for the remaining samples.

In the example below, the concentration of a migrant increases nonlinearly at the beginning of the experiment, e.g., between day 0 and day 3, and then linearly between days 3 and 10. For estimating the concentration of the migrant in food processed in an open "flow-through" system, the following additional assumptions have been made:

1. the solvent volume is 250 mL.
2. the solvent is circulated through the column at a rate of 60 mL per hour (1 mL/min).
3. the concentrations of the migrant in the solvent measured on days 3, 5, 7, and 10 are 150 ppm, 450 ppm, 750 ppm, and 1050 ppm, respectively.

Thus, the rate of increase in the migrant concentration is 150 ppm per day. For a solvent whose density is 1 g/mL, this daily increase of 150 ppm is equivalent to 37.5 mg of the migrant that has migrated into 250 mL of the solvent in a single day ((250 mL x 150 mg) / 1000 mL = 37.5 mg). If the column were operated at "flow-through" conditions, the above amount, i.e., 37.5 mg, would migrate into a volume passing through the column in 24 h, i.e., 1440 mL (60 mL/h x 24 h = 1440 mL). Thus, the resulting concentration of the migrant would be 26 ppm ((37.5 mg / 1440 mL) x 1000 = 26 mg/L). According to the initial assumption, the industrial column is equivalent to the experimental one in terms of the operating conditions. Thus, the above migrant concentration may be expected to be similar to that in the enzyme-treated food provided densities of the solvent and food are approximately 1 g/mL. If this is not the case, appropriate corrections should be incorporated in the calculations.

Footnotes

1. More information pertaining to petitions may be obtained from the Office of Premarket Approval (HFS-200), Food and Drug Administration, 200 C Street, S.W., Washington, DC 20204. (See updated contact information)

2. Recommendations for Chemistry Data for Indirect Food Additives. Copies of this document may be obtained from the Office of Premarket Approval (HFS-200), Food and Drug Administration, 200 C Street, SW, Washington, DC 20204. (See updated contact information)

3. The space velocity is the volume of liquid passing through a given volume of immobilized enzyme in 1 h divided by the latter volume (Immobilized Enzymes, 1978, Ichiro Chibata, ed., Halsted Press, a Division of John Wiley and Sons, New York, N.Y.).