FERMENTATION OF FRUIT PRODUCTS.
The invention relates to the fermentation of vegetable and/or fruit products, more in particular to the fermenta¬ tion of crushed/broken fruits from plants of the family of the Solanaceae, in particular tomato, capsicum (paprika and/or pepper) , Chili pepper and egg plant, preferably from Lycopersicum esculentum (=tomato) , and any cultivar there- of.
This invention also includes the fermentation of mixtures of crushed/broken fruits as e.g. of tomato and pepper or mixtures of tomato and vegetables .
More in particular the invention relates to the preparation of tomato paste, tomato pulp, tomato juice or other tomato based products. Tomato paste is an important commercial product and is used as an ingredient for soups, sauces and ketchup. The largest part of the world tomato crop is processed into tomato paste and the present invention which relates inter alia to the preparation of tomato paste is therefore commercially important.
A typical tomato paste process comprises: tomatoes -> washing -> crushing/breaking -> heating -> pul¬ ping/sieving -> juice -> concentration -> paste, but many variations are known. Breaking can be effected at tempera¬ tures of about 90°-95°C (hot break) or at low temperatures of about 40°-60°C (cold break) . Cold break favours degra¬ dation of cell wall material by pectolytic enzymes and the apparent viscosity, which is an important quality attribu¬ te, is increased. Adjustment of the Ph by addition of citric acid and degassing are steps which are often inclu- ded to improve the end quality of the paste. The above processing steps cause physical, chemical and enzymatic
changes to occur in the tomato material which influence the rheological, other physical properties and organoleptic properties of the end product.
Enzymatic modification of tomato suspensions has been investigated (thesis F. W. C. den Ouden, Agricultural
University Wageningen, The Netherlands 1995) . The effects of pectin degrading enzymes caused tomato cells and parti¬ cles to disintegrate into smaller particles and the values of rheological parameters of the suspension were generally, sometimes after an initial increase, found to decrease and moreover objectionable serum separation on top of fluid products increased which causes less consumer appeal . Serum separation is interrelated with thickness and a higher thickness tendency decreases serum separation. The tendency towards serum separation can conveniently be estimated on a laboratory scale by centrifuging the material .
Lebensm. -Wiss. u. Technol . , 22, 65-67 (1989) discloses the preservation of whole ripe small ripe tomatoes (8-10g) by means of covering them with brine and subsequent subjecting them to lactic acid fermentation as to obtain a keepable fermented product that can be consumed in salad.
EP-A-0 308 064 (Kagome Kabushiki Kaisha) discloses to improve the flavour of a beverage based on tomato by lactic acid fermentation using particular lactobacilli strains. Thus tomato beverages are prepared with a "compound but unified flavour" by fermenting a processed tomato product, preferably together with a milk product with Lactobacillus bulgaricus and/or Lactobacillus helveticus- More preferably the fermentation is also carried out in the presence of Streptococcus thermophilus. It stated that only by using at least Lactobacillus bulgaricus and/or Lactobacillus helve¬ ticus that generation of so-called "off flavour" can be controlled during the lactic acid fermentation of a proces-
sed tomato product or its mixture and it is stated that beverages with a compound but unified flavour can be ob¬ tained efficiently.
It is clear from the prior art that lactic acid fermen¬ tation of tomato based products has been used in order to obtain keepable products. It is also clear that tomato products with an improved "unified" flavour can be obtained by lactic acid fermentation with very specific lactobacil- li. At least some of these lactobacilli as e.g. L. brevis are heterofermentative and convert sugar into lactic acid, acetic acid, carbon dioxide and ethanol. Most lactic acid bacteria aim at the production of lactic acid and not at the production of extra cellular polysaccharides. However, fruit products like tomato juice, tomato paste and products derived therefrom, apple juice, etc usually also suffer from other disadvantages such as e.g. serum separation. Generally consumers like a keepable, thick, rich product showing no serum separation and having a good flavour. Food additives like colours, acidifying agents e.g. citric acid and thickening agents e.g. modified starch are, however, not generally appreciated. The present invention aims to provide fruit products with a favourable combination of the above features.
In a first embodiment of the invention a process for pre¬ paring an improved fruit product such as juice, paste or pulp is provided comprising the steps of : i) breaking and/or crushing the fruit, ii) adjusting the pH to 4.0 - 7.0, preferably to 5.5 - 6.5, iii) sterilising and/or pasteurizing the broken/crushed fruit, iv) adding a culture of a Lactobacillus or Lactococcus strain producing when growing an extracellular
polysaccharide at a rate of at least 0.8 g/1 at 20°C, a pH of 5.8 within 24 hours in a medium according to De Man, J.C., M. Rogosa and M.E. Sharpe (J. Appl . Bacteriol . 23:130-135, 1960) , . followed by fermenting, and v) pasteurisation and/or sterilisation of the fermen ted fruit material.
More preferable the rate of producing extracellular polys¬ accharide (EPS) mentioned above is at least 1.0 g/1, most preferably at least 1.2 g/1.
Breaking and/or crushing the fruit is conveniently carried out after first washing and blanching or scalding the fruit, the tomatoes are then broken and/or crushed using e.g. a chopper or vacuum crusher. There are the possibili- ties of a "hot break" or a "cold break" as set out above. The broken and/or crushed fruits may then be refined i. e. extracted or sieved to remove peels, seeds and possibly stems. However, it is also possible to carry out the refi¬ ning step later in the process. Suitable equipment is e.g. an extractor of the screw type or of the paddle type. The juice obtained is then optionally deaerated and/or salted. The pH of the fruit mass is then adjusted to 4.0 - 7.0, preferably to 5.5 - 6.5 by the careful addition of a basic substance usually food grade sodium hydroxide. The exact pH value selected is usually determined by the optimal pH value for the growth of the particular Lactobacillus or Lactococcus strain to be employed. When it is intended to use e.g. Lactobacillus sake 0-1 (CBS 532.92) of which the optimal pH value is known to be 5.8 an initial pH value somewhat above 5.8 is selected so that the Lactobacillus grows well during the fermentation period. For other suit¬ able Lactobacilli and/or Lactococci different values will generally apply. Dependent on the nature of the Lactobacilli and/or Lacto- cocci actual fermenting may take place under different
conditions. Again in the case of Lactobacillus sake 0-1 (CBS 532.92) fermenting takes preferably place at a tempe¬ rature between 15 and 35°C and under conditions without forced supply of oxygen for a period of 5 - 30 hours. Generally the ranges of the fermentation temperature is somewhat wider viz. from 10 - 35°C anaerobic conditions may not be required. Fruits employed in the practice of this invention generally have a dry matter content below 10% (tomatoes 5-7.5%) of which about half consists of reducing sugars mostly D-fructose and/or D-glucose. Lactobaccilli and Lactococci convert these sugars when growing into lactic acid and polysaccharide. The percentage of soluble solids is conveniently expressed according to the Brix scale (i.e. calculated as sugar) and refractometers there- fore often have in addition to the refractive scale a Brix scale.
After fermenting the Lactobacilli and/or Lactococci are inactivated usually by heat treatment i.e. by pasteurisation and/or sterilisation. Thereafter the fermen- ted product may be deaerated and packed. Quite often pac¬ king takes place before pasteurisation/sterilisation. When aiming to produce a fruit paste as e.g. tomato paste con¬ centration of the fermented liquid e.g. to a strength of around 7.5. Brix is desirable. Concentration is conveni- ently effected in tanks with heating coils or in vacuum pans.
In a preferred embodiment of the invention such a process is provided in which the fruit is from a plant of the family of the Solanaceae, preferably from Lycopersicum esculentum (=tomato) , or any cultivar thereof. A particu¬ larly preferred group of tomato cultivars for the practice of this invention are the so-called "Pomodori" of Italy. For special effects it maybe desirable to use mixtures of e.g. tomatoes and paprika, or tomatoes and vegetables or
their juices, preferably at least 50 wt% of tomatoes, more preferably at least 80 wt% of tomatoes are used.
In another preferred embodiment of the invention the Lac- tobacillus or Lactococcus strain producing when growing an extracellular polysaccharide is selected from the group
Lactobacillus sake 0-1 (CBS 532.92) , Lactobacillus paraca- sei (LMG 9193tl and Lab 97) and Lactococcus lactis cremori π (LAB 338) . LMG and LAB are abbreviations which indicate that the strains have been deposited at collections kept at Ghent University, Belgium. The use of mutants including those obtained by DNA-recombinant technology or classical mutagenolysis derived from the above Lacobacilli and Lac¬ tococci which are functionally equivalents of those iden- tified above as to EPS and lactic acid formation is also covered by the present invention. The above identified microorganisms form during growth not only lactic acid, but also polysaccharides which thicken the fruit mass. As far as e.g. Lactobacillus sake 0-1 (CBS 532.92) is concerned reference is made to Appl . and Environm. Microbiol . , 6
(August 1995) , pp. 2840-2844 in which inter alia the exo- polysaccharide formed by liactobacillus sake 0-1 which comprises glucose and rhamnose units is more fully identi¬ fied. The use of a Lactobacillus or Lactococcus strain producing an extracellular polysaccharide containing rham¬ nose units is preferred inter alia because this may lead to particularly favourable flavour effect in the processed end product.
In a further preferred embodiment of the invention deae- ration of the broken and/or crushed fruit is effected prior to pasteurisation and/or sterilisation.
In a further preferred embodiment of the invention sieving is effected prior to pasteurisation and/or sterilisation.
In a further preferred embodiment of the invention fermen¬ tation is carried out at a temperature between 10 and 50°, preferably 20 and 40°C in the absence of supplied oxygen or air.
In a further preferred embodiment of the invention prior to fermentation with the Lactobacillus or Lactococcus strain a saccharide, preferably sucrose is added to the broken and/or crushed fruit material. More preferably the amount of sucrose in the broken/crushed fruit material is adjusted to a level of 15-25 g/kg fruit material. Especial¬ ly when using Lactobacillus paracasei (LMG 9193tl) the addition of sucrose is beneficial.
In a further embodiment of the invention the process from step ii) onwards is preferably carried out under aseptic conditions. Under these circumstances sterilisation to obtain the end product may be superfluous.
In a still further embodiment of the invention an enzymatic treatment of the fruit product is inserted between steps iv and v of the process, preferably using oxidoreductases such as peroxidases and/or glucose-oxidases .
Furthermore, the invention provides a fermented food pro¬ duct obtainable by a process as described above. These products compare favourably with those known sofar as to the properties mentioned above. More in particular they are improved as to serum separation, and thickness, and some- times also as to colour and taste etc. Moreover their content of labelled food additives as e.g. thickening agents and citric acid can be minimised or their use even completely avoided.
The invention is further illustrated by the following examples. All parts and percentages in this specification and claims are taken on a weight basis unless otherwise indicated.
Example 1.
About 2 kg of fresh tomatoes (bought in a local supermar¬ ket) were stripped from stalks and washed with tap water. Subsequently the tomatoes were crushed for 3 min (at posi¬ tion 2-5) using a type Kenwood major (Kenwood Ltd, UK) electronic kneader / mixer. After crushing the seeds and skins were removed by centrifugation and sieving using a type Braun (Braun, Germany) household centrifuge and a lab- oratory test sieve type ASTM 11, 30 mesh (Endecots Ltd, UK) respectively, resulting in a tomato juice (suspension) having a dry matter content of 4.8%, a pH value of 3.93 and a Brix value of 4.5. After adjusting the pH value to 6.3 with food grade sodium hydroxide the juice was pasteurised (2 min - 100°C) , cooled to 28°C and inoculated with
Lactobacillus sake 0-1 (CBS 532.92) and to a starting cell concentration of 2 x 106 cells per gram of juice (determi¬ ned as colony forming units) . Subsequently the inoculated tomato juice was fermented and processed as described below in more detail in example 5.
Table 1. Effect of fermentation on viscosity and sensoric properties of tomato juice.
Rheological measurements (run time)
To demonstrate the effect of fermentation on the rheologi¬ cal properties of tomato paste a fermented sample was compared with an unfermented sample in a standardised viscosity test using a type EZ™, equivalent "Zahn" vis- cosity cup (Gardco, USA) , and which is said to exceed ASTM D4212, which cup has a diam. 5 mm orifice.
Example 2.
The same procedure as in Example 1 was used, however, after centrifugation the tomato paste was heated for 2 minutes at 98 - 100°C and concentrated to a Brix value of 7.0 using a type Buchi R-124 Rotavapor vacuum evaporator operating at 65°C and a pressure of 15 - 20 kPa.
Table 2. Effect of fermentation on serum development in tomato paste at x 540 g force ( dm = 7.8 %) .
R eo ogy accord ng to met o as escr be n examp e 1 The organoleptic properties of the fermented product were excellent.
Example 3.
About 2 kg of fresh tomatoes (type Italian Pomodori, bought at a local wholesaler) were stripped from stalks and washed with tap water. Subsequently the tomatoes were crushed for 3 minutes (at position 2-5) using a Kenwood major (Kenwood Ltd, UK) electronic kneader/mixer. After crushing the seeds and skins were removed by centrifugation and sieving using a Braun (ex Braun, Germany) household centrifuge, resulting in a tomato paste having a pH value of 4.29 and a Brix value of 5.0. After adjusting the pH value to 6.3 with food
grade sodium hydroxide the paste was heated for 2 minutes at 98 - 100°C and concentrated to a Brix value of 7.0 using a type Buchi R-124 Rotavapor vacuum evaporator operating at 75°C and a pressure of 15 - 20kPa. Subsequently,the tomato paste was pasteurised (2 min - 100°C) cooled to 28°C and inoculated with Lactobacillus sake 0-1 (CBS 532.92) and a starting cell concentration of 2 x 106 cells per gram of paste (determined as colony forming units) . Thereafter the inoculated tomato paste was fermented for 24 hours at 28°C. After fermentation the tomato paste was pasteurised for 30 minutes at 80°C, cooled to room temperature and used for organoleptic - and rheological analysis as described in Example 5. The product according to the invention proved to have very good organolepcic properties.
Table 3. Effect of fermentation on serum development in tomato paste at x 540 g force (Brix value 7.1)
R eo ogy accor ng to t e met o escr e n Examp e 1.
Example 4.
About 2 kg of fresh tomatoes (type Italian Pomodori, bought at a local wholesaler) were manually stripped from stalks and skins (after 1 minute immersion in boiling water) and washed with tap water. Subsequently the tomatoes were cut into 4 pieces and heated in a microwave oven (type Amano, 750W) for 10 minutes till the temperature reached 90°C. After heating the tomato pieces were processed into a paste using a Hobart N 50 (ex Hobart, Holland) kneader/mixer and sieved through a diam. 0.9-1.1 mm orifice using a type
Hobart 200S (ex Hobart, Holland) pilot sieving unit. After adjusting the pH value to 6.3 the tomato paste was further processed and fermented as described in the previous exam¬ ple (Ex. 3) .
Table 4. Effect of fermentation on serum development in tomato paste at x 540 g force (Brix value 7.0)
R eo ogy according to t e met od described n Examp e 1
Example 5,
Commercially available tomato paste (type Pummaro, ex STAR, Milan, Italy) . Thus 100 g of Pummaro product was aseptical- ly transferred into a sterile 300 ml size glass bottle and inoculated with 0.5% of a washed cell suspension of the lactic acid bacterium type Lactobacillus sake 0-1 [_£B_S_
532.92) and to a starting cell concentration of 2 x 106 cells per gram (determined as colony forming units) . Before inoculation the pH of the tomato matrix was adjusted to a value of 6.4 +/- 0.1 by mixing in approx. 0.5% food grade NaOH (10.8 mol/1) .
Subsequently the inoculated tomato paste was fermented for 24 hours at 28°C during which time samples were taken for carrying out analysis. After the fermentation was completed the matrix was pasteurised for 30 minutes at 80°C and cooled to room temperature.
Rheological measurements:
To demonstrate the effect of fermentation on the rheologi¬ cal properties of tomato paste a fermented sample was
compared with an unfermented sample in a standardised centrifugation test using serum layer development as an indicator. Thus 10 g of (un) fermented tomato paste was transferred each into a 16 x 100 mm culture tube and cen- trifuged for 0, 3 and 6 minutes at 540 g using a type
Hettig 30F Universal centrifuge. After centrifugation the height of the serum layer was measured in mm:
Table 5. Effect of fermentation (as also indicated by viable cell count) on serum development in tomato paste at x 540 g force (dm = 8.59%) .
11 viable count in cfu/g before pasteurisation.
Example 6.
Same as example 5 above except that during fermentation samples have been analysed for serum development, pH and viable counts.
Table 6. Effect of fermentation time on tomato paste
Example 7.
The same procedure as described in example 6 was followed, however, now using Lactococcus lactis cremoris LAB 338 .. Result : serum layer in 10 g fermented product after 9 minutes at 540 g (for procedure see Examples 1 and 5) within range of 2 - 4 mm. Control sample (unfermented) same as in example 1.