FR2731587A1 - Removing somatic cells from food or biological media - Google Patents

Abstract

Removal of somatic cells from food or biological media involves subjecting the medium to at least one tangential microfiltration on a membrane having cut-off threshold at least 10 (pref. 10-50) microns and recovering the treated prod. as microfiltrate or permeate. Also claimed are products obtd. by the process, specifically milk contg. less than 10,000 SS per ml (and opt. also retaining at least 85% of its original content of fat globules or having a practically unmodified total initial flora); prods., esp. cheeses, obtd. by conversion of such milk are also claimed.

Description

 The present invention, which was carried out with the technical assistance of Messrs. Maubois and from the Dairy Technology Research Laboratory of the National Institute for Agricultural Research (INRA), belongs to the field of food and biotechnology industries. It relates more particularly to a method for eliminating somatic cells present in food or biological media. The invention also relates to the products that can be obtained by such a method.

 Somatic cells are present in many food and biological products and environments. Among food environments, milk must be mentioned first, without limitation. Blood and cell culture media are also examples of fluids or biological media within the scope of the invention. In the following, the process of the invention has been described in detail with reference to the treatment of milk. This is indeed a particularly interesting example of liquid food that can be subjected to said process.

It is known that somatic milk cells have many disadvantages. Thus, we have already noted the relationship between the cellular content of milk and diseases of the breast of the female dairy. according to
Kasthi et al. (Path Microbiol., (1966), 29, 630), the presence in the milk of a large number of cells may have an inhibitory action on the growth of certain bacteria in the milk. As a result, the somatic cells of the milk exert an unfavorable influence on the milk processing because of the heat-resistant enzymes they contain, these enzymes being at the origin of the degradation of the lipid and protein components of the milk. A recent article (Griffiths
MW Sci.Food agric. (1989), 47, 133-158) studies the thermal resistance of Listeria bacteria and concludes that the presence of Zie Listeria cells in leucocytes gives them heat resistance. As a result, the somatic cells protect the pathogenic bacteria, of the Listeria monocytogenes type, in particular, against the thermization and / or pasteurization treatments to which the milks are subjected in the food industry.

 The bibliographical references cited above are illustrative of the state of the art, but it is clear from this brief study that the presence of somatic cells in milk can have many adverse consequences, including diminishing notable, the yields of cheese making.

 It should also be noted the emergence of increasingly stringent regulations in terms of food hygiene. For example, France has a decree of 18 March 1994 on the quality of milk. This text prohibits the use of milks containing more than 500,000 somatic cells per ml for processing into pasteurized milk products and those containing more than 400,000 cells per ml for processing into raw milk products. There is therefore a very serious current need to increase the sanitary quality of milks by greatly lowering the somatic cell levels they contain.

 It has already been proposed, in particular to improve the quality of milk, to subject it to filtration or centrifugation: see for example Genin G. Milk (1932), 12, 839 and Giroux J. Milk ( 1937), 17, 354. These known techniques do not allow to selectively remove somatic cells from milk, especially whole milk. In the latter case, in fact, it is necessary to preserve the fat globules of the milk. Filtration or centrifugation techniques have the main disadvantage of eliminating such fat globules at the same time as somatic cells. And, in addition, the elimination of somatic cells is not completely satisfactory, especially on an industrial scale.

 The object of the present invention is to overcome the disadvantages of the prior art as briefly mentioned above.

 It is particularly concerned with a method for eliminating somatic cells from the food or biological media which contain them, on the one hand in a more extensive manner than in the prior art and, on the other hand, selectively.

 In the case of milk treatment, the invention further relates to a process for obtaining a milk meeting the most stringent requirements of sanitary quality, that is to say, whose somatic cell level is very low. According to another particularly important aspect, the invention relates to a method for treating whole milk, in particular raw whole, which makes it possible to selectively separate the somatic cells that it contains, while preserving the useful banal flora in the milk, especially with a view to the processing of milk, for example cheese.

 Another object of the invention is an industrially applicable method for eliminating somatic cells without destroying them.

 The various technical problems solved by the invention as well as others will also be apparent from the description which follows.

 In its general form, the subject of the invention is a method for eliminating somatic cells present in food or biological media, comprising subjecting the medium to be treated to at least one tangential microfiltration on a membrane having a cutoff threshold of greater than 10. pm, preferably between 10 and 50 pm and recover, as microfiltrate (or permeate), the desired treated product.

The invention makes a particular application of the microfiltration technique. Microfiltration is a process known per se for separating particles whose size is greater than that which would separate them by ultrafiltration. To fix ideas,
Ultrafiltration makes it possible to work on compounds or particles up to about 0.1 gm, whereas microfiltration is used for particle sizes of the order of 1 "m in general According to the microfiltration technique, the medium to be treated is brought into contact under pressure with a membrane of appropriate cutoff threshold.The working pressure is moderate, for example of the order of 0.1 to 0.5 bar.

 In principle, the tangential microfiltration technique is also known. The person skilled in the art can refer in this respect to the documentation distributed by Alfa Laval, in particular that relating to the microfiltration unit type MFS-38.

 The invention takes advantage of the known technique of tangential microfiltration to solve the technical problems described in the present description, by selecting high-cut-off membranes for this purpose. This cut-off threshold is at values greater than 10 m, preferably between 10 and 50 μm and even more preferably of the order of 10 to 15 μm, for example approximately 12 μm.

 The tests reported below show that the membranes with a cut-off threshold of less than or equal to 3 μm are unsatisfactory, especially in the case of the treatment of whole milk. Indeed, such membranes, or are not capable of satisfactorily retaining somatic cells or, if they manage to retain them, the separation performed is nonselective, because they also retain the milk fat of the initial milk. Those skilled in the art, noting such results, would be encouraged to use skimmed milk in order first to remove the somatic cells that it contains, and then to enrich the permeate with the milk fat. can be satisfactory because the centrifugal skimming operation leads to a significant training, in practice of the order of 50 to 65%, somatic cells in the fat phase. Thus, when it is introduced into the microfiltration permeate, a significant amount of somatic cells will still be present in the milk for processing, including cheese.

In contrast, the method of the invention, which uses high cut-off membranes, not only makes it possible to obtain a milk that is free of somatic cells, but also a microfiltered milk whose biochemical composition is close to that of the initial milk. As an illustration,
The invention makes it possible to obtain a milk containing less than 10,000 somatic cells per ml, without modifying the total flora that it initially contained. Likewise, the fat content is practically preserved in the microfiltrate, which corresponds to a selective separation of somatic cells and fat globules.

The critical parameter of the process of the invention is the implementation of high cut-off membranes, as mentioned above. The particular nature of the membrane is irrelevant as long as it meets the cut-off requirement. Other technical considerations may be involved in practice for the final choice of the membrane: mechanical resistance, no clogging, ease of cleaning etc. Preference is given to mineral or ceramic membranes, for example alumina. But one could also implement organic membranes having the desired cutoff thresholds. The best results have been obtained with an alpha alumina membrane, available on the market under the name Membralox from the company.
SCT and having a cutoff threshold of about 12 "m.

 The actual implementation of the tangential current microfiltration is carried out in a known type of installation. The product to be treated is supplied by a feed pump to flow tangentially through channels consisting of or including membranes. Part of the product, the permeate or microfiltrate, passes through the membrane and is collected in an enclosure. Media that does not pass through the membrane, or retentate, is recycled to the circulation pump and mixed with the incoming product to be microfiltered again. In practice, the installations are therefore in the form of a loop system. Such loops can be connected in parallel to constitute higher capacity installations. The tests reported below were carried out in a microfiltration unit type MFS-38 from Alfa Laval.

 The process of the invention is suitable for continuous or discontinuous use. In the latter case, a determined quantity of the medium to be treated is brought into contact with the membranes of the microfiltration unit until a permeate or microfiltrate having the desired properties is obtained, in particular with regard to the concentration maximum in somatic cells. But the tangential microfiltration installations are also suitable for continuous implementation, according to which the medium to be treated is brought into contact with the membranes in a continuous manner, the operating conditions of the installation being chosen so as to enable the collection, also continuously a microfiltrate having the desired properties. Those skilled in the art have at their disposal the means for adjusting the microfiltration units to obtain such a result under industrial conditions.

In the case of the treatment of milk, it has been found, during the implementation of the process of the invention, that it is advantageous to use a positive circulation pump, in particular a volumetric lobe pump, instead of a conventional centrifugal pump. With the typical installation
MFS-38 supra, it was also found that the most efficient flow rate was 4 to 5m3 per hour, with a volume concentration factor (CVF) of about 25. An important parameter to consider is the microfiltration temperature. In the case of milk, it has been found that the best temperature was of the order of 500 ° C., since at lower temperatures, especially close to 150 ° C., there was a rapid clogging of the membranes. The optimum adaptations of the various parameters available to the skilled person to implement tangential microfiltration, have been illustrated in the following examples, which relate to the application of the method of the invention to the treatment of milk.

 To be able to be subjected to the process of the invention, it suffices that the media to be treated are pumpable, so as to be brought into tangential contact with the membranes. These media can therefore be suspensions or liquids. It may therefore be, as indicated above, cell culture media or other biological fluids, such as blood and food media, very preferably milk.

 The milk can come from any female dairy cow, goat, sheep, donkey, buffalo, singly or in mixture. The process of the invention is especially suitable for the treatment of raw whole milk.

 In another aspect, the invention also relates to products obtainable by the aforementioned method.

 Thus, the invention more particularly relates to a milk containing less than 10,000 somatic cells per ml. It also relates to a milk containing less than 10,000 somatic cells per ml while retaining at least 85% of its original content in fat globules.

The invention also relates to such a milk in which the initial total flora is practically not modified.

The invention will be further illustrated, without being in any way limited by the description which follows, made with reference to the accompanying drawings in which
FIG. 1 is a schematic diagram of microfiltration
FIG. 2 is a schematic view partially broken away of a multichannel tangential filtration element
FIG. 3 is a detail view in section of a channel of the element illustrated in FIG. 2.

 Fig. 1 is a block diagram illustrating the general microfiltration technique. The medium 1 to be treated is supplied by a pump, not shown, on a microfiltration unit or module 2. It is separated into a permeate or microfiltrate 3 and a retentate 4. Usually, the retentate 4 is recycled as shown in dotted line 5 to be mixed with the medium 1 to be recirculated in the unit 2 .

 FIG. 2 illustrates a multichannel element of tangential filtration which has proved advantageous for the implementation of the method of the invention. The element 10 shown comprises a plurality of channels 12 which are held in a porous support 16. The product or medium to be treated is fed by a pump not shown to circulate through the channels as represented by the arrows 11. The permeate or microfiltrate is collected at 13 after passing through the element 10. The retentate exits the element 10, as represented by the arrows 14.

 The mounting of a membrane 17 within a channel is shown in more detail and in large scale in Figure 3. We see the membrane 17 and the porous support 16 which allow the selective passage of the permeate.

The tests reported hereinafter were carried out on whole cow's milk in a microfiltration unit type MFS-38 from Alfa.
Laval, with Membralox membranes in alpha alumina having a cut-off of about 12 "m.

 In what follows, the first section contains information on the choice of pumping, volumetric or centrifugal types, as well as recirculation rates and residence times. In the second part, the results on permeation rates, possible concentration factors, cell retention, and the biochemical composition of milks and retentates were collected. The third part describes the results obtained by optimizing the technological parameters and also, in this same part, has been described the influence of the treatment temperature on microfiltration.

 The description also sets forth a number of practical cheese making trials from microfiltered milk.

Part N0 1: Food - Conditioning milk
The mode of reheating milk (from 20C to 500C, or 550C) has an influence on somatic cell survival (or counting).

 - A batch reheat, via a positive displacement pump at a rate of 1000 I / h did not destroy cells. Under these same conditions, a centrifugal pump resulted in progressive destruction over time (5% in 15 minutes, 16% in 30 minutes).

 - Reheating in double-walled tanks did not cause any destruction.

These results led to warming the milk in a double-walled tank for pump testing. Subsequently, continuous feeding via centrifugal pumping was used. Such a choice was justified by
the small destruction obtained in 15 minutes in a discontinuous manner (5%);
- the ability to work at continuous feed rates
less than 1m3 / h.

 It has been determined, in continuous feed, that the centrifugal feeding pump of the installation did not cause destruction.

 In addition, keeping the milk cold did not change the amount of cells counted in the milk at 24 and 48 hours. The milk tested was thus stored cold before treatment.

Choice of the type of pumping
All the tests conducted to determine the choice of the recirculating pumping mode made it possible to retain the volumetric lobe pump.

- The best survival rate of the cells is obtained with the volumetric pump by recirculating the retentate at 5m3 / h
* After 10 minutes of recirculation, 90% of the cells are
conserved;
* after 20 minutes, the survival rate is only
80%.

 - The rolling at the rate of 7.9 m3 / h with the positive displacement pump causes the strongest destruction of the cells (73% in 10 minutes).

 - Centrifugal centrifuge retentate pump tests show rapid destruction in the shortest time (<15 minutes). It is actually the turbine in the body of the pump that is causing the destruction. The highest recirculation rate (9.1 m3 / h), correspondingly lowering the residence time in the pump casing, results in the lowest destruction rate (in 15 minutes 34% at 9.1 m3 / h, 45% at 7.9 m3 / h, 47% at 5 m3 / h ,.

Part N02: Choice of permeation rate and volume concentration factor (VCF)
This study was conducted during tests N09, 10, 11, 12, 13 and 14.

From these tests, it appears that the optimal recirculation flow rate was 4 m31h (scanning speed 4.6 m / s), the permeation flow of 400 11h 2000 I / h / m2 and 25 FCV. thus obtained
a good passage of the fat (Test N013: 95.8%),
- less passage of cells in the permeate (Test N013
3%, Test N014: 9%),
- a slight change in transmembrane pressure (PT) (Test
N015: 25 mbar / h).

Part N03: Optimization of microfiltration
Two trials (E1 and E2) were carried out at one month intervals, with for each test a quantity of whole milk treated of 2000 1. The treatment parameters and the results obtained are collected at
Table I.

Table I
E1 E2
Volumetric recirculation (m / h) 4 4
Duration of the test 4h20 5h
Temperature (C) 53 55
Average permeate flow (l / h / m) 2000 1950
Continuous extraction FCV 25 25
Evolution of the PT (mbar / h) 21 40
Fat transmission (%) 88 95.5
Living cells (%) 96 80
Cells (permeate / milk) (%) 7 1
Part N 04: Influence of the microfiltration temperature
a) Action of the recirculation regime with the volumetric Domoe
The destruction of the cells at 5 m3 / h was less important at
1 50C at 500 C. Two concentration tests at this recirculation regime were carried out.

b) Microfiltration at 150 C. FCV 20. 500 I / hlm2
The membrane clogged in less than one hour under these conditions.

Conclusion
From the overall results, it appears that
- 10% of the milk cells are found in the permeate, but variations in the rate of destruction of the cells are observed as a function of the residence time. With the installation used, it took between 25 and 30 minutes to reach the FCV 25. This residence time can be decreased 2 times, or even 5, by increasing the surface / volume ratio of the loop.

With the installation used, this ratio is 0.026. The increase of this ratio makes it possible to reduce the destruction of the cells and to perform
FCV 30 to 35 to decrease the amount of retentate to be reprocessed.

 - The rate of passage of fat varies from 85 to 95%; it depends on the initial composition of the milk, more precisely on the size of the fat globules and their agglomerates.

 - Optimum microfiltration is carried out at a concentration of 25, at a rate of 2000 I / h / m2, at a scanning speed of 4.6 m / s with a lobe volumetric pump.

 - One can microfiltrate raw whole milk between 50 and 550C: this parameter decreases the total flora of the milk by half, favors microfiltration flow rates by decreasing the viscosity of milk and concentrates.

 - The implementation of a whole milk containing nearly 200,000 cells per ml leads to a microfiltrate containing 1000 to 5000 cells per ml. The total flora is practically unaffected (44000 CFU / ml in the starting milk, 76000 CFU / ml in the microfiltrate). The fat content is slightly lowered from 44.7 g / kg in the milk used to 41.5 g / kg in the microfiltrate. These results were obtained at 400C for in-feed / outgoing retentate flow ratios of 20. The permeate flow was maintained at 1000 l.h1 .m2.

REPORT ABSTRACT TESTS
Test 1: Influence of the pump type
Test of the feed pump (PA)
- ExR 200 I / h (15 minutes duration)
- 500C
As a result, the centrifugal feed pump generates virtually no cell destruction for a period of 15 minutes.

Volumetric recirculation pump test {PCVRJ
- Change of flow every 15 minutes
- Variation from 1 to 9 m3 / h
Result: little destruction (<10%) up to 4 m3 / h.

Centrifugal Pump Test (PCCR)
- Change of flow every 1 5 minutes
- Variation from 1 to 7.9 m3 / h
Result: the centrifugal pump, from the first quarter of an hour, destroys 57% of the cells.

Trial 2: Confirmation of Trial 1
- Maintain a constant flow rate of 7.9 m3 / h for 7 minutes
Result more linear destruction with the volumetric pump. Destruction identical to test 1 with the centrifugal pump.

With a time of less than 7 minutes, the centrifugal pump destroys few cells at 7.9 m3 / h.

Test 3: Influence of temperature and heat exchangers
- Bain-marie (30 minutes)
- Exchanger with continuous centrifugal pump and in
batch
- Exchanger with a positive displacement pump and in
batch
Results: The temperature of 500C has no influence on the cells. Exchanger pumps cause destruction of the cells. A water bath or a double-walled tank is more efficient.

Test 4: Volumetric pump test (PCVR)
- Flow rate: 7.9 m3 / h
- Recirculation for one hour
Centrifugal pump test (PCCR)
- Same conditions
Result: At this rate, the destruction of cells is much greater with the positive displacement pump.

Test 5: Confirmation of test 4
Result: we observe the same variations as in the previous test.

Test 6: Centrifugal Pump Test (PCCR)
- Flow rate retentate: 5 m3 / h
- Duration: 2 hours
As a result, the destruction is greater over time compared with the results obtained for a flow rate of 7.9 m3 / h.

Test 7: Volumetric Pump Test (PCVR)
- Retentate flow: 5 m3 / h (2 hours)
As a result, these experimental conditions provide the best cell conservation over time.

Test 8: Centrifugal Pump Test (PCCR)
- Retentate flow rate: 9.1 m3 / h (1 hour)
Results: they are equivalent to those obtained with a flow rate of 7.9 m3 / h. The destruction of cells is important.

Test 9: Concentration with the volumetric pump, study of the passage of the fat
- Recirculation flow retentate: 5 m3 / h
- Permeate extraction rate: 200 I / h
- Concentration up to FCV 23
Results: this test shows a slight change in transmembrane pressure (0.05 b in 55 min). 87% of the fat passes into the permeate. Destruction of 35% of FCV cells 23.

Test 10: Concentration with the volumetric pump, study of the passage of the fat
- Recirculation flow retentate: 5 m3 / h
- Permeate extraction rate: 200 I / h
- Concentration up to FCV 40
Results: there is an evolution of 0.06 b in 90 minutes of the transmembrane pressure. There is a rate of passage of fat of 92% compared to the initial milk in the middle permeate. AT
FCV 25: 50% destruction of the cells. At FCV 40: 60%.

Test 11: Concentration with the volumetric pump, study of the passage of the fat
- Recirculation flow retentate: 5 m3 / h
- Permeate extraction rate: 400 I / h
- Concentration up to FCV 40
Results: we find a very good passage of the fat in the permeate (> 92%). The number of living cells remains constant at around 70% up to FCV 40. The transmembrane pressure increased by 0.08 bar in 45 minutes.

Test 12: Concentration with the volumetric pump, study of the passage of the fat
- Recirculation flow retentate: 4m3 / h
- Permeate extraction rate: 200 I / h
- Concentration up to FCV 40
Results: the cell numbers are less well preserved during the concentration, since only 40% of FCV 40 remains. An increase of 0.03 bar in transmembrane pressure was noted in 100 minutes. The passage of fat is 93%.

Test 13: Concentration with the volumetric pump, study of the passage of the fat
- Recirculation flow retentate: 4 m3 / h
- Permeate extraction rate: 400 I / h
- Concentration up to FCV 40
Results: increase in transmembrane pressure of 0.15 bar in 50 minutes. 93% of the passage of the fat.

Test 14: Concentration with the volumetric pump
- Recirculation flow retentate: 4 m3 / h
- Permeate extraction rate: 400 I / h
- Concentration up to FCV 40
Result: increase of the transmembrane pressure of 0.15 bar in 45 minutes (as in the test N013).

Test 15: * Concentration and extraction at FCV 25
- Temperature: 550C
- Recirculation flow retentate: 4 m3 / h
- Permeate extraction flow rate: 400 I / h
- Concentration up to FCV 25 - Retention extraction rate 16.6 I / h at FCV 25
Results: slight variation of transmembrane pressure 25 mbar / h for 3 hours. 87% fat passage is obtained.

* Concentration at FCV 250 (discontinuous)
- Recirculation flow: 4 m3 / h
- Permeate extraction rate: 200 I / h
Result: The majority of fat retained passes through the membrane but with many cells.

Test 16: Optimizing microfiltration at FCV 25
- Temperature: 550C
- Recirculation flow retentate: 4 m3 / h (volumetric pump)
- Permeate extraction flow rate: 400 I / h
- Concentration up to FCV 25
- Extraction rate retentate 16.6 I / h at FCV 25
Results 10% overall cell destruction. 21 mbar / h evolution of the transmembrane pressure in 4h20. 88% fat passage. Decrease of 50% of the total flora in the permeate.

Test 17: Concentration at FCV 250 on MRT (discontinuous)
- Recirculation flow retentate: 4 m3 / h
- Permeate extraction rate: 200 I / h
Results 82% of passage of the fat.

Increase of 1.15 bar of transmembrane pressure in 25 minutes.

8.5% of passage of the cells.

Test 18: Concentration at FCV 34
- Recirculation flow retentate: 4 m3 / h
- Permeate extraction flow rate: 400 I / h
- Retentat extraction rate 10.2 I / h
Results: increase in transmembrane pressure of 40 mbar / h in 5 hours. 95% fat passage. 9% of cell passage. 20% of the destroyed cells.

Test 19: Optimization at FCV 25
Recirculation flow retentate: 4 m3 / h
Permeate extraction: 400 I / h
Retentate extraction: 16.6 I / h at FCV 25
Results: evolution of the transmembrane pressure of 40 mbar / h in 5 hours. 20% destruction of the cells.

Test 20: Concentration at FCV 25
Cheese aptitude of microfilmed milk
Result: There is no difference in the cheese ability between the microfiltered milk and the initial milk.

 The usual experiments carried out to assess the aptitude of a milk for cheese processing, in particular using the "formagraph" apparatus known to those skilled in the art, made it possible to determine the parameters such as the setting time. , the firming time and the firmness of the cheeses respectively obtained with the initial milk and the microfiltered milk.

 In order to demonstrate the decisive role of the selection of microfiltration membranes, comparative tests have also been carried out using the same plant as in the previous tests but by equipping it with membranes whose cut-off point is less than 10 μm.

 Comparative tests (a): microfiltration on 3 m membrane.

In this test, raw whole milk (200 I) was first skimmed at 400 C. The raw skimmed milk was then subjected to tangential microfiltration on the MFS1 plant with 3 gm membranes. , with a surface area of 0.2 m2 and a temperature of 400 C. The other microfiltration parameters were as follows:
Permeation rate: approximately 100 I / h or 500 I / h / m2.

 - FCV 5, 10, 20 to 100 I / h.

 The results of microbiological and somatic cell counts are summarized in Table 2.

Table 2
Products Coliform cells Total flora
(10 / ml) (/ ml) (/ ml)
Whole milk 221 61 223.10
Skimmed milk 80 70 147.10
Cream (52.5) 20 82.10
Microfiltrate FCV 20 0 98
Retentate FCV 20 1146 953 165.10
The results of the biochemical analyzes are shown in Table 3.

Table 3
EST products (g / kg) MAT (g / kg) mg (g / kg)
Whole milk 129.25 32.37 45.5
Skimmed milk 89.48 33.69 0.5
Cream 615.27
Microfiltrate 89.37 33.87
Retentate 96.3 35 17
Comparative tests (a) above indicate a high retention of fat and also coliforms (microorganisms of small size, about a few microns are also retained at this cutoff) in the retentate.

 Comparative tests (b): microfiltration on 1.4 "m membrane.

The experimental conditions are, in a nutshell, as follows: Raw materials: whole milk of large mixture - Centrifugal creaming temperature 370C - Microfiltration MFS1, membrane 1,4pm, temperature 370C, flow = 500
I / h / m2, FCV 20
Results of determination of leukocytes and cells
are summarized in Table 4.

Table 4
Products Cells x 10 / ml Califorms Total flora
Whole raw milk 279 3.7.10 2.104
Laitecr 154 3.2.102 1.9.104
Microfiltered bed 2 limit detection) Abs. 3.6.101
Restoring microfiltration 1503 8.8.104 1.5.108
Skim milk-cream mixture 144 2.9.10 2.104
Microfiltered milk-cream mixture (1) (90-10) 11 1.8.10 1.3.10
Microfilter-cream milk mix (2) (90-10) 9
Indigestible cream / g 3.3.10 1.1.104
The results of the biochemical analyzes are shown in Table 5.

Table 5
MAT products (g / kg) EST (g / kg) mg (g / kg)
Whole raw milk 34,27 128,78 43
Raw skimmed milk 35,87 90,96 0,5
Microbial re-release (FCV 20) 39.45 103.73 11.5
Microfiltrate 35.72 89.07 0.5
pF + cream 34.41 119.70 35
Skim milk + cream 34.56 118.85 34
Comparative tests (b) with cutoff thresholds of 1.4 microns indicate losses of the order of 50% somatic cells.

On the other hand, this cut-off threshold does not make it possible to treat whole milk by separating the fat cells from the somatic cells by this method.

 In the foregoing, the method of the invention has been more particularly illustrated in its application to the treatment of milk, especially raw milk, as it may be intended for cheese processing. In addition to this preferred application, the selective separation of somatic cells by tangential microfiltration is of interest for the rapid determination of microorganisms linked to them. For example, the enrichment phase could be suppressed because of the very high volume concentration factors (> 200) that can be achieved in microfiltration operations. The selective concentration of the cells of size close to those present in the milk can also be carried out by the same method on non-dairy liquids: other food liquids and cell culture media both for technological purposes (clarification for example) and for objectives nutritional or medical.

 On another level, the selective concentration of large fat globules is of interest in dairy technology: cream, butter and cheese.

Claims (12)

 1. A method for eliminating somatic cells present in food or biological media, comprising subjecting the medium to be treated to at least one tangential microfiltration on a membrane having a cut-off threshold greater than 10 "m, preferably between 10 and 50 pm and recover, as microfiltrate (or permeate), the desired treated product.  2. Method according to claim 1 characterized in that the cutoff threshold is between 10 and 15 IJm, and is for example about 12 "m.  3. Method according to one of claims 1 or 2 characterized in that one uses mineral or ceramic membranes, in particular alpha alumina.  4. Method according to any one of claims 1 to 3 characterized in that it is implemented continuously or discontinuously (that is to say in batches).  5. Method according to any one of claims 1 to 4 characterized in that the medium to be treated is pumpable, so as to be brought into tangential contact with the membrane.  6. Method according to claim 5 characterized in that the medium to be treated is a suspension or a liquid selected from food or biological media, for example cell culture media or other biological fluids, such as blood.  7. Method according to any one of claims 1 to 6 characterized in that the medium to be treated is milk, in particular whole milk.  8. Products obtainable by the process according to any one of claims 1 to 7.  9. Milk containing less than 10,000 somatic cells per ml.  10. Milk containing less than 10,000 somatic cells per ml while retaining at least 85% of its original content in fat globules.  11. Milk containing less than 10,000 somatic cells per ml and in which the initial total flora is practically unaffected.  12. Products, in particular cheese, obtainable after processing of the milk according to any one of claims 9 to 11.