FR2844524A1 - Separating epithelial cells from human or animal urine for diagnostic or therapeutic applications, employs membrane microfiltration and recovery using trypsin solution - Google Patents

Abstract

The sample is subjected to membrane microfiltration, through a mean pore diameter of 11 mum, to obtain a retentate which largely contains the epithelial cells. The cells are recovered by treating the membrane with a solution containing trypsin. The sample is subjected to membrane microfiltration, through a mean pore diameter of 11 mum, to obtain a retentate which largely contains the epithelial cells. The cells are recovered by treating the membrane with a solution containing trypsin. In addition, the suspension of cells obtained, is re-filtered and recovered using the same procedure. Microfiltration comprises one or more stages of microfiltration followed by washing of the retentate. After membrane filtration at a mean pore diameter of 11 mum, a buffer solution is used for washing. This dissolves any crystals of urates present in the retentate. Microfiltration is achieved by aspiration of the filtrate.

Description

The present invention relates to the field of purification of

  epithelial cells contained in human or animal urine, in

  particularly for diagnostic or therapeutic purposes.

  From the pelvis to the beginning of the urethra, the wall of the urinary tree is lined with a stratified transition epithelium. In the bladder, this epithelium is made up of about seven layers of cells. The deepest are cylindrical, while the intermediate cells are of various shapes. The surface is covered with transitional cells, also called urothelials. Transitional or urothelial cells are normal elements of the urinary sediment.

  The exfoliation of epithelial cells, specifically urothelial cells, in the urine is a normal physiological process in humans or animals. This exfoliation is the result of renewal

normal epithelium.

  However, there is an increase in exfoliation in certain pathologies, including pathologies causing an increase in cell mortality and pathologies weakening the link between cells and

  the basement membrane of the epithelium.

  Thus, during the various pathological processes affecting humans or animals, cells derived from the epithelial lining of the urinary tree, also called urothelial cells, peel in the urine flow. The examination and characterization of these exfoliated cells, from a urine analysis, makes it possible to diagnose a pathology including

  the individual from which these cells originate may be affected.

  The main techniques commonly used for the characterization and detection of pathological urothelial cells, in particular malignant urothelial cells, are in particular the following: - analysis of the morphological characteristics of the cells, by urinary cytology; - detection of antigens specific to urothelial cells, such as the bladder tumor antigen (BTA), cytokeratins, fibrinogen degradation products or even the NMP-22 antigen; - analysis of massive chromosomal abnormalities, in particular by flow cytometry, in situ fluorescence (FISH) or techniques of the Quanticyt type; - detection of changes in chromosomal regions, for example by amplification and analysis of repeated sequences (microsatellite markers); - analysis of the number of copies or of the expression of genes involved in carcinogenesis constituting marker genes, such as

telomerase or survivin.

  In order to improve the sensitivity and specificity of the various diagnostic tests carried out using epithelial cells present in the urine, there is a need in the prior art for methods for enriching the urine samples with starting in epithelial cells and simultaneously eliminating the numerous other constituents of urine which may affect the specificity or the sensitivity of the diagnosis, or even in certain cases making this diagnosis impossible. In addition to the epithelial cells, the urine may also contain other components, cellular or non-cellular, in particular nucleated cells such as leukocytes and, where appropriate, also sperm, anucleated cells such as red cells, possibly microorganisms. of bacterial or fungal origin, or of non-cellular components, in particular of

  numerous crystals, of various size and chemical nature.

  In particular, it is observed that the presence of numerous leukocytes and red blood cells in the urine causes false positive or false negative results to be obtained during the implementation of the various tests for

diagnosis listed above.

  For diagnostic techniques requiring preparation of the DNA contained in the epithelial cells, for example for the analysis of alleles of different DNA markers contained in the genome of the urinary epithelial cells, it has been proposed to separate the human cells. contaminating organisms, mainly microorganisms contained in the urine, before proceeding to the DNA extraction step. In particular, the team at VAN DER HEL et al. carried out work aimed at determining the optimal conditions of specificity and sensitivity of a PCR amplification from DNA isolated from frozen urine

  (VAN DER HEL et al., 2002, Analytical Biochemistry, vol. 304: 206-211).

  These authors notably tested the possible technical advantage that a step of separation of the epithelial cells from the 5 other constituents of the urine could constitute, prior to the extraction and then to the amplification of the DNA. For this purpose, this team carried out a separation of the epithelial cells from frozen urine samples comprising (i) a first stage of centrifugation of the urine, (ii) resuspension of the solid pellet thus obtained, ( iii) first filtration by centrifugation of the resuspended pellet through a 40 Nylon filter

  µm pore size, (iv) resuspending the sediment contained in the filtrate, (v) centrifuging filtration of the resuspended sediment through a nylon filter having a pore size of 11 µm, then ( vi) direct lysis of the cells retained on the surface of the filter in order to extract the DNA whose analysis by amplification is sought.

  However, these authors observed that, on the one hand, there is no correlation between the degree of bacterial contamination of the starting samples and the success of the amplification by PCR of the DNA extracted from the urinary cells and that, d on the other hand, the bacterial decontamination of the 20 samples by the above filtration process does not make a difference

  significant in the final results of DNA amplification. These same authors conclude that the lengthening of the duration of the process, due to the bacterial decontamination step prior to the extraction of the DNA, is unjustified in view of the absence of apparent correlation between the quality of the DNA amplification and the degree of bacterial contamination.

  Surprisingly, it has been shown according to the invention that a separation of the epithelial cells contained in a human or animal urine sample by microfiltration of this urine sample on a filtering membrane having an average pore diameter of 11 μm, then recovery of the cells retained on the surface of said filtering membrane with a solution containing trypsin made it possible to obtain a cell suspension containing predominantly epithelial cells, said epithelial cells having moreover preserved their physical integrity, which makes them able to be used directly for the implementation of various diagnostic tests of cytological, immunological, type

  flow cytometry or genetic type mentioned previously.

  The subject of the invention is a method of separating epithelial cells from a human or animal urine sample containing these cells, characterized in that it comprises the following steps: a) microfiltration of the urine sample on a filter membrane having an average pore diameter of 11 μm, in order to obtain a microfiltration retentate containing predominantly epithelial cells,

  b) recovery of the cells from the microfiltration retentate, by treatment of the filtering membrane with a solution containing trypsin.

  It has been shown according to the invention that the above process makes it possible to obtain a purified suspension containing mainly epithelial cells, due to the elimination of a large part of the undesirable cellular and non-cellular constituents initially contained in the urine, respectively leukocytes, bacterial or fungal microorganisms, red blood cells, crystals of various natures as well as filaments or mucus, these various undesirable components being found respectively in 75%, 30%, 54%, 21% and 23% of samples from

departure tested.

  In particular, it has been shown according to the invention that the method of

  separation of the epithelial cells from a urine sample above makes it possible to completely eliminate the leukocytes initially present in 80% of the samples tested. In addition, in approximately 90% of the samples tested, less than 10% of the leukocytes initially present in the urine are found.

  The term "average pore diameter" is, at the end of this

  description, synonymous with "mesh size" of a filter membrane, these expressions being used respectively depending on whether the filter membrane comprises a porous surface or whether the filter membrane comprises a screen.

  According to the above method, the animal urine sample is preferably a mammalian urine sample. The urine sample treated according to the method of the invention may be a freshly collected urine sample, if necessary kept cold, for example at 40C, for a period ranging from a few minutes to a few hours, for example.

  example 48 hours. The urine sample can also come from urine collected at any time, then stored in frozen form, for example at a temperature below -200C, then thawed shortly before use. The urine sample capable of being treated according to the above method can also come from a urine sample taken from man or animal by endoscopy or with a catheter.

  In all cases, the urine sample has not undergone any step of separation of its various cellular or non-cellular constituents,

  prior to its treatment by the method of the invention.

  If necessary, the urine sample can be diluted with a medium which makes it possible to maintain good viability of the epithelial cells, or at the very least to maintain an excellent physical integrity of these cells, such as for example a Hank's medium. Any cell culture medium

  mammals can also be used.

  Preferably, step b) of recovery of the cells from the microfiltration retentate obtained in step a) is carried out by treatment of the microfiltration membrane with a solution containing a concentration of trypsin allowing detachment of the cells from the retentate from the surface of the filtering membrane without significantly affecting the viability of the epithelial cells, or at the very least without significantly affecting the physical integrity of these cells, that is to say by keeping the cell membrane intact as much as possible. Appropriate concentrations of trypsin can be readily determined by those skilled in the art, in light of the numerous scientific publications describing the use of trypsin for detaching adherent cells in culture, from a culture support, for example from from the surface of a Petri dish or from the

  microplates for cell culture.

  Advantageously, a buffer solution containing a

  final concentration of 0.25% trypsin.

  Advantageously, trypsin is added to a medium which makes it possible to maintain good viability of the epithelial cells, or at least a good physical integrity of these cells, such as for example a

  Hank's, in the absence of calcium or magnesium.

  The efficiency of the separation of the epithelial cells from the other constituents of urine with the method according to the invention can be further improved by carrying out an additional separation cycle comprising a second stage of microfiltration of the cells of the retentate from the first stage. microfiltration, in order to eliminate more completely the

  various other undesirable constituents.

  Thus, according to a particular embodiment of the method according to the invention, said method comprises the following additional steps: c) microfiltration of a suspension of the cells obtained in step b) of the method on a filtering membrane having an average diameter of 11 μm pore, in order to obtain a microfiltration retentate containing predominantly epithelial cells; d) recovery of the cells from the microfiltration retentate by treatment of

  the filter membrane with a solution containing trypsin.

  The operating conditions of the optional second microfiltration cycle

  defined above are preferably identical to those used for the first stage of microfiltration then recovery of the cells of the retentate.

  It has been shown according to the invention that the use of a solution containing trypsin to recover the epithelial cells present in the microfiltration retentate makes it possible to achieve a high yield of epithelial cells, since, when the process is carried out with a step of recovering the cells in the absence of trypsin, only from 0% to a maximum of 60% of the cells obtained with the method of the invention are recovered. In addition, the method for separating urinary epithelial cells of the invention results in reduced cell loss, which is at most equal to 25% of the epithelial cells initially contained in the samples,

for 60% of the samples tested.

  As previously stated, urine samples contain

  various crystals, which was observed in 21% of the samples tested.

  Some of the crystals present in the urine have a size greater than 11 μm 30 and are therefore retained by the filter membrane during step a) of microfiltration. When present in large quantities, these crystals, mainly urate crystals, can cause an at least partial clogging of the filter membrane, so as to considerably reduce the separation characteristics of the microfiltration step. This technical problem can be overcome by performing step a) of microfiltration in

  presence of a buffer solution capable of dissolving the urate crystals. Preferably, the passage of the urine sample and the passage of said buffer solution will be alternated, during step a) of microfiltration, in order to periodically dissolve the urate crystals 5 capable of being retained at the surface of the filter membrane to prevent clogging.

  Thus, according to a particular embodiment of the above method, said method is characterized in that step a) of microfiltration comprises one or more cycles of microfiltration then washing of the retentate, each cycle 10 comprising the following steps: ai) microfiltration of a urine sample on a filter membrane having an average pore diameter of 11 μm,

  a2) washing the retentate obtained in step ai) by adding a buffer solution capable of dissolving the urate crystals possibly present in the microfiltration retentate.

  Advantageously, to dissolve the urate crystals, a

  TBS buffer solution (25 mM Tris, 140 mM NaCI, 3 mM KCI, pH 7.4).

  The number of cycles of stages ai) of microfiltration then of stage a2) of washing of the retentate is variable. The number of cycles is adjusted according to the quantity and size of the crystals retained on the surface of the

  membrane filtration, as well as the volume of the urine sample to be microfiltrated.

  In general, the number of stage cycles ai) of microfiltration then a2)

  The washing is preferably 1 to 10 and most preferably 1 to 5. The higher the number of cycles, the greater the risk of cell loss.

  Most preferably, the filter membrane is chosen from

  a material that does not promote cell adhesion to its surface. A preferred filter membrane according to the invention is a nylon filter with a pore size of 11 µm, such as nylon filters with a pore size of 11 µm sold by the Millipore Company.

  Preferably, step a) of microfiltration is carried out by suction

of the filtrate.

  The applicant has in fact observed that carrying out step a) of microfiltration by centrifugation leads to poor results. In particular, when the initial urine sample contains urate crystals capable of being retained by the filter membrane, a clogging of the filter is observed, without the possibility of dissolving said urate crystals in

centrifugation course.

  Advantageously, the suction of the filtrate during step a) of microfiltration is carried out using a conventional vacuum pump device.

  However, any other device designed to create a vacuum on the side of the filter membrane where the microfiltration filtrate is collected can also

to be used.

  The combination of the technical characteristics of the separation method according to the invention makes it particularly suitable for being integrated into a method in which it is sought to obtain a sufficient quantity of DNA from epithelial cells, for example in order to carry out genetic tests, while minimizing the simultaneous presence of contaminating bacterial or fungal DNA or even contaminating DNA originating from other nucleated cells initially present in urine samples,

  mainly contaminating DNA from leukocytes.

  Indeed, the high selectivity of the separation method according to the invention for epithelial cells, with elimination of a large part, and in some cases all, of the leukocytes initially present in the samples, makes it possible to prepare suspensions cells containing a reduced proportion of human DNA from other cells

than epithelial cells.

  Likewise, the majority, and most often all, of

  Bacterial or fungal microorganisms possibly present in some urine samples are eliminated in the microfiltration filtrate.

  In addition, many inhibitory factors for the various reactions

  necessary for carrying out genetic tests, such as PCR amplification, are also eliminated in the microfiltration filtrate. These include red cells, if necessary sperm, microorganisms and the various crystals initially present in the urine.

  Finally, maintaining the integrity of the epithelial cells separated according to the method of the invention allows the recovery of a maximum amount of DNA originating from the epithelial cells, and this in a controlled manner, that is to say exclusively after the completion an additional step of lysis of the cells. Such a result could not be achieved with a process in which the cell lysis step is carried out in situ, that is to say directly on the filter membrane, without step b) prior to recovery of the cells to using trypsin, due to the inevitable loss of a large part of the DNA initially contained in the cells retained on the surface of the filter membrane. Consequently, another object of the invention is a high-yield method for extracting the DNA contained in the epithelial cells present in a sample of human or animal urine, comprising the following steps: a) separation of the epithelial cells other constituents of urine by carrying out the separation process as defined above; b) lysis of the cells separated in step a), then extraction of the DNA released at

from lysed cells.

  Step b) of lysis of the separated cells, then extraction of the DNA 15 can be carried out by a person skilled in the art by any conventional technique. In particular, the lysis step is carried out with a conventional lysis buffer solution, then extraction of the DNA, for example in a suitable volume of absolute ethanol. A person skilled in the art may advantageously refer to the article in VAN for the implementation of step b) of the above method.

DER HEL et al. (2002), supra.

  The experimental results presented in the examples show the technical advantages conferred by the implementation of the method for separating urinary epithelial cells of the invention as a step prior to the use of the DNA originating from the epithelial cells in an analysis. patient allelotyping genetics. The selectivity and specificity of the allelotyping step are increased, compared to those of known methods, due to the high quality of starting DNA, which essentially consists of DNA originating from epithelial cells and which contains a reduced proportion of undesirable contaminating DNA, and the fact that the starting DNA contains a reduced amount of inhibitors of the PCR amplification reaction which are initially contained in the urine samples. In addition, in most of the tests, the method of the invention makes it possible to obtain a large quantity of starting DNA, which is also such as to increase the selectivity and the specificity of the allelotyping step. , through

compared to previous processes.

  According to yet another aspect, the invention relates to a process for allelotyping a DNA contained in the epithelial cells present in a sample of human or animal urine, characterized in that it comprises the following steps a) extraction of the DNA contained in said epithelial cells according to the extraction method as defined above;

  b) use of the DNA extracted in step a) in a DNA allelotyping test.

  Step b) can comprise the following steps: bl) Amplification of the DNA extracted in step a) using at least one nucleotide primer hybridizing specifically with a genomic DNA fragment containing a polymorphic genetic marker ; b2) characterization of the allele of said polymorphic genetic marker from the DNA amplified in step b1), for example by sequencing of said

Amplified DNA.

  Step b) of using the DNA in an allelotyping test can be carried out by any technique known to those skilled in the art. To carry out step b) of the method, a person skilled in the art will advantageously refer to

  the article by SCHNEIDER et al. (2000, Cancer Research, vol.60: 4617-4622).

  The invention also relates to a method for analyzing the epithelial cells contained in a human or animal urine sample, characterized in that it comprises the following steps: a) separation of the epithelial cells from the other constituents of

  urine by implementing the method according to one of claims 1 to

  ; b) analysis of the characteristics of the epithelial cells separated at

  step a) by a cytological, immunological, genetic in vitro diagnostic test or by flow cytometry.

  The present invention is further illustrated, without being

  limited, by the following figure and examples.

  Single figure Figure 1 illustrates the allelotyping results of the DNA contained respectively in the blood leukocytes (upper line), the cells present in the microfiltration retentate (middle line) and the cells

  present in the microfiltration filtrate (bottom line).

  The allelotyping analysis was carried out for three distinct markers, respectively the TP53 marker, the D45243 marker and the THO marker.

  described in Schneider et al. (2000, supra).

  DA represents the value of the allelic imbalance compared to a DNA

  control and is calculated as described in the article by SCHNEIDER et al. 10 (2000, supra).

  ò On the upper line are shown, for each marker, the allelotyping results, obtained with DNA

  control extracted from circulating leukocytes from the blood.

  On the midline are shown, for each of the 15 markers, the allelotyping results, obtained with DNA

  extracted from the cells contained in the microfiltration retentate, after implementation of the method of the invention.

  On the lower line are represented, for each of the markers, the allelotyping results obtained with DNA extracted from the cells found in the filtrate of

  microfiltration, after implementation of the process of the invention.

EXAMPLES

  A. Materials and methods used in Examples 25 A.1. Patients and samples The development of the selective screening process for urine, its feasibility and the impact of filtration on the result of a tumor cell detection test were assessed on samples from 48 tumor patients. of the bladder and followed using the allelotyping test (microsatellites). The patients concerned are included in the 1997 Hospital Clinical Research Protocol (Pr P. OUDET and Pr D. JACQMIN. All the patients included in the study signed a consent agreement to participate in it. Urine samples (volume 50 to 150ml) are collected by natural urination and stored at 40C until treatment. The time elapsing between urination and urine processing

varies from 12 to 48 hours.

  A.2. Urine filtration Urine is passed directly through a nylon pore filter

  11 µm in diameter and deposited on a filtration medium. The passage of urine over the filter is facilitated by the use of a vacuum tube. After passing the urine, the filter is rinsed with 5ml of PBS (phosphate buffer).

  The filtrate is collected in tubes which are centrifuged. The resulting pellets 10 are stored at -20 ° C. after microscopic examination of the

  elements present in the filtrates.

  A3. recovery of urothelial cells present on the filter: The filter is collected and washed for 5 min in 20 ml of PBS / Trypsin. The action of trypsin is stopped by the addition of 20 ml of PBS (dilution effect). The filter is then removed from this washing solution and the cells thus detached and suspended in the washing solution are centrifuged. Ten fold of the cell pellet are removed for microscopic observation. The rest of the pellet is frozen at -80 ° C. 20 A.4. Filtration procedure in the presence of urates: When the urine contains urate crystals (visible to the eye

  naked, very large pink crystals; running the risk of clogging the filter), the urine is filtered in small volumes alternating with TBS (Tris Borate buffer which has the property of dissolving urates).

  AT 5. Allelotyping test: The DNA of the cells filtered from the urine is extracted, together with the DNA originating from the blood leukocytes and which constitutes the control sample for the allelotyping test. The allelotyping test is described in the Schneider et al. DNA was extracted from blood and urine samples from the same patient. The urine has previously been filtered. DNA from cells on the filter and cells in the filtrate was extracted. PCRs were carried out using 3 different markers using as DNA the DNA extracted from blood leukocytes (control DNA, top line), cells recovered on the filter (middle line) and cells in the filtrate (bottom line). The 2 peaks corresponding to the 2 alleles are noted by their relative height. The allelic imbalance (DA) is calculated relative to the control DNA (Schneider et ai). B. Results Example 1: Evaluation of the loss of epithelial cells The filtration technique under study is based on the use of a nylon filter 10 whose mesh size is 11 μm. This pore size allows the retention of elements whose size exceeds 11 μm and the elimination of elements of size less than 11 μm. This differential filtration must in this case result in the separation of the epithelial cells, the diameter of which is generally greater than 20 μm, from the leukocytes which have a size of approximately 10 μm. The filter is placed on a filtration support to which is connected a vacuum tube which allows to apply a force entraining the urine and the elements to be eliminated. Due to the application of this force, some cells still pass through the pores or are plated

  strongly on the filter, which generates losses.

  1.1 Optimization of the recovery of cells adhering to the filter after filtration: Preliminary tests had shown that after filtration, a large percentage of cells adhere to the filter. A washing of the filter with trypsin allows a significant restitution of the cells after

  filtration. The action of trypsin was evaluated using the following experiment: after filtration of urine, we cut the filter in half. Half was washed in trypsin, the cells are recovered by centrifugation and then counted. The other half was washed in PBS (cells centrifuged and counted).

  The results of the analysis of the cell recovery rate at the end of the

  process are shown in Table 1 below.

Table 1:

  Evaluation of cell restitution using trvpsin half of the% of cells half of the filter filter returned washed in the absence of trypsin washed with PBS trypsin Patient 1,700,000 cells 5,000 7% Patient 2,600,000 15,000 25% Patient 3,500 0 0% Patient 4 50,000 30,000 60%

  The variability observed in this experiment is linked to the cellular profile of the urine, which is very different from one patient to another: the factors that may contribute to this variability are the initial number of cells, the size of the cells, their condition, the presence of 'aggregates, etc. In all cases, washing the filter with trypsin allows a significant return of the material.

  1.2 Evaluation of cell loss by passage through the filter

  After filtration, we assess the proportion of cells lost in the filtrate. The results are presented in Table 2 below.

2844524 15

Table 2:

  Evaluation of cell loss Losses Number of samples observed concerned% or less 29/48 (60%)

% 9/48 (19%)

% 9/48 (19%)

  more than 50% 1/48 (2%) Example 2: Influence of filtration on the DNA extraction yield DNA losses can be observed following the cell losses described above. After filtration, the cells are centrifuged and the DNA is extracted and quantified. Of the 48 urine samples, 3 were dropped from the study for insufficient DNA. The insufficient amount of DNA was due in these 3 cases to a very low initial number of cells and not to a loss of cells

during filtration.

  Example 3 Influence of Filtration on the Quality of DNA Filtration of the urine allows the elimination of a medium of very variable chemical composition from one patient to another and loaded with potential PCR inhibitors. Elements such as erythrocytes, germs, crystals of different kinds, sperm are also eliminated very effectively. These molecules or elements other than the epithelial cells,

  if not eliminated can also alter the quantification of DNA.

  Example 4: Elimination of leukocytes and impact on the result of the PCR amplification carried out during an allelotyping analysis on DNA of epithelial cells originating from samples of human urine. The presence of a significant number of leukocytes in the urine, following inflammatory situations or infections leads to a decrease

  sensitivity of the allelotyping test. Indeed, their DNA not being altered, they are likely to negate the signal of allelic loss (since the observed allelic imbalance is the result of a population of cells).

  The mesh of Il pm allows the selective elimination of leukocytes, given their size less than 11 pm. Among the 48 samples tested, 35 ie 73% contained leukocytes. The final ratio after filtration between epithelial cells and leukocytes was evaluated:

  - for 28/35 (80%) of the cell pellets, all the leukocytes were eliminated.

  - for 3/35 (8.5%) of the cell pellets, a leukocyte / cell ratio

  25% epithelial was observed.

  - for 4/35 (11.5%) of the cell pellets, a leukocyte / cell ratio

  50% epithelial was observed.

  If we consider the elimination of leukocytes compared to the number of leukocytes initially present in the urine, the following efficiencies are observed: - number of urines in which less than 10% of the initial leukocytes remain: 33/37 ( 89%) - number of urines in which 25% of the initial leukocytes remain: 2/37

(5.5%)

  - number of urines in which 50% of the initial leukocytes remain: 2/37

(5.5%).

  In the case where the urine is sufficiently rich in epithelial cells, it is possible to carry out a second filtration if the elimination of the leukocytes is not sufficient. In this case, the second

  filtration allows the total elimination of leukocytes.

  The elimination of leukocytes from the urine is essential for the sensitivity of the test (see protocol). Filtration therefore eliminates 30 cells whose genome has not been altered and which would have the effect of "diluting"

  the signal of allelic imbalance given by tumor cells. To assess this point, we extracted the DNA from the cells retained by the filter (potentially tumor-like) as well as the DNA from the cells eliminated by filtration. The rearrangements were evaluated by PCR. The results are shown in Figure 1, as well as in Table 3 below.

  TABLE 3: Analysis of the allelic imbalances detected in the cells retained on the filter in comparison with the cells present in the filtrate% DA cells% DA cells Composition of the filtrate retained retained on the filter in filtrate (proportion C. epithelial cells / leukocytes) Patient A D4S243 ** 15 12.2> 90% leukocytes

ACTBP2 * 50 18

MJD52 ** 7 4.5

  Patient B MJD52 ** 3 16 30% THO leukocytes * 26 6.9 70% epithelial

D17S654 * 23 13.5

  Patient C THO * 26 9.7> 90% leukocytes

ACTBP2 ** 11.4 13.5

TP53 * 27 9

  Patient D D9S171 ** 10.7 7.5> 90% leukocytes

IFNA ** 18 8

D17S654 * 26 30

TP53 ** 14.2 5.7

  Patient E D4S243 * 27 4> 90% leukocytes

ACTBP2 * 29 5

D9S162 * 34 8

D9S747 * 29 12.5

MJD52 * 21 1.4

  PatientF D9S171 ** 13 2> 90% of leukocytes

TP53 * 20 9

D17S654 ** 3.8 3

  * DA values considered significant.

  ** undisturbed markers, used as controls (goal: to show that filtration induces no false positive signal in the filtrate).

  As shown in the results of Figure 1, the filtration made it possible to specifically retain on the filter epithelial cells, whose genome is reworked (allelic imbalance of 27, 29, and 34% for markers 1, 2 and 3, respectively ). In the filtrate, the imbalances are clearly reduced (4, 5 and 8%). We evaluated this filtration selectivity in 15 samples from 6 patients whose urine contained leukocytes (Table 3). In most cases, the allelic imbalances are significantly lower in the filtrate compared to the filter, indicating that the procedure developed leads to a marked increase in the sensitivity of the test. Exceptions are found in the case of patient B (markers MJDJ2), and of patient D (marker D17S564). For patient B, the 5 imbalances found in the filtrate are explained by the fact that a significant number of epithelial cells were lost during filtration (see composition of the filtrate, Table 3). This phenomenon is only visible for certain markers, certainly indicating the selective loss of certain epithelial cells, a loss which is nonetheless counterbalanced by the number of 10 markers tested. The study on undisturbed markers (**) shows that it

  there is no false positive signal induction related to filtration.

  Example 5: Adaptability of the technique to urines of very diverse composition and adjustment of the procedure in the presence of 15 urate crystals.

  The composition of urine is extremely variable from one patient to another.

  The major components in addition to epithelial cells and leukocytes are: - red blood cells, present in 26/48 (54%) of the samples, 20 germs, present in 14/48 (30%) of the samples, - filaments or mucus in 11/48 (23%) of the samples,

  - crystals of different natures, in 10/48 (21%) of the samples.

  Filtration effectively removes red blood cells, germs and most crystals. When urate crystals are present which, therefore, clog the filter and prevent the passage of urine, a

  particular procedure is implemented due to their high abundance and.

  In this case, the urine is filtered in the presence of TBS buffer, which has for

  property of dissolving urate crystals.

  In the presence of a significant amount of mucus or filaments, filtration is a little slower and can be the cause of poor

  leukocyte removal efficiency (observed in 2 out of 48 samples).

Claims (8)

  1. A method of separating epithelial cells from a human or animal urine sample containing these cells, characterized in that it comprises the following steps: a) microfiltration of the urine sample on a filtering membrane having an average pore diameter of 11 μm, in order to obtain a microfiltration retentate mainly containing epithelial cells,   b) recovery of the cells from the microfiltration retentate, by treatment of the filtering membrane with a solution containing trypsin.   2. Method according to claim 1, characterized in that it comprises the following additional steps: c) microfiltration of a suspension of the cells obtained in step b) of the process on a filter membrane having an average pore diameter of 11 µm, in order to obtain a microfiltration retentate mainly containing epithelial cells; d) recovery of the cells from the microfiltration retentate by treatment of   the filter membrane with a solution containing trypsin.   3. Method according to one of claims 1 or 2, characterized in that step a) of microfiltration comprises one or more cycles of microfiltration   then washing the retentate, each cycle comprising the following steps: ai) microfiltration of a urine sample on a filter membrane 25 having an average pore diameter of 11 μm, a2) washing the retentate obtained in step ai) by addition of a buffer solution capable of dissolving the urate crystals if necessary   present in the microfiltration retentate.   4. Method according to one of claims 1 to 3, characterized in that the   filter membrane is a nylon filter.   5. Method according to one of claims 1 to 4, characterized in that   step a) of microfiltration is carried out by suction of the filtrate. 35 6. A high-yield process for extracting DNA contained in the epithelial cells present in a human or animal urine sample, comprising the following steps: a) separation of the epithelial cells from the other constituents of the urine   by implementing the method according to one of claims 1 to 5,   b) lysis of the cells separated in step a), then extraction of the DNA released at from lysed cells.   7. A method of allelotyping DNA contained in epithelial cells present in a human or animal urine sample, characterized in that it comprises the following steps a) extraction of the DNA contained in said epithelial cells according to the method of claim 6,   b) use of the DNA extracted in step a) in a DNA allelotyping test.   8. Method for analyzing the epithelial cells contained in a sample of human or animal urine, characterized in that it comprises the following steps: a) separation of the epithelial cells from the other constituents of   urine by implementing the method according to one of claims 1 to 5;   b) analysis of the characteristics of the epithelial cells separated in step a) by a cytological, immunological in vitro diagnostic test,   genetic or by flow cytometry.