TECHNICAL FIELD The present invention relates to an improved, cost effective, in vitro method for the selection of motile sperm from a sperm-containing sample, for example the ejaculate.

BACKGROUND ART It has been estimated that approximately 14-16% of all couples attempting to get pregnant have difficulty in conceiving and are defined by fertility therapists as infertile. 40% of these cases result from male factors. In a substantial proportion of these, treatment is available to ameliorate or relieve the condition which leads to infertility, these treatments usually require the separation of motile sperm from semen.

Methodology for separating motile sperm from semen is well established in the clinical literature [Practical Laboratory Andrology, David Mortimer, pages 267-286]. In general terms, four basic approaches exist (1) simple dilution and washing; (2) sperm migration; (3) selective washing procedures; and (4) adherence methods for the elimination of debris and dead spermatozoa. The success of such methods is often assessed by the yield of motile sperm, the cost of the procedure and the likely contamination with seminal components.

Migration of sperm into human cervical mucus in vitro is a standard test for analysing spermatozoa. The use of cervical mucus, however, suffers from several disadvantages, so substitutes have been described. These should behave in a similar manner to cervical mucus, and also be readily available, easily prepared, cheap and stable. Substitutes which have been used include bovine cervical mucus [Keel & Webster (1988) Fertil. Steril. 49: 138-143], polyacrylamide gel [Lorton et al. (1981) Fertil. Steril. 35: 222-225], but the most common substitute is hyaluronic acid [Aitken et al. (1992) J. Androl. 13: 44-54]. It has been shown that swim-up from semen into a medium containing an aqueous salt of hyaluronic acid gives a significantly higher percentage of motile spermatozoa compared to swim-up from a washed pellet [Wikland et al. (1987) Hum. Reprod. 2: 191].

Whilst hyaluronic acid-based media are effective in separating motile sperm, a significant disadvantage is they are expensive (especially in the form of'Sperm Select') and require storage conditions of between 2-8°C.

DISCLOSURE OF THE INVENTION We have now advantageously found that the process of selecting motile sperm may be improved if the penetration medium contains a cellulose derivative, such as a methylcellulose or a hydroxymethylcellulose.

This present invention provides methods and products for separating motile sperm cells in vitro from a sperm containing sample by means of contacting said sample with a layer of an aqueous artificial penetration medium for sperm cells, said medium having incorporated therein a cellulose derivative, whereby motile sperm are caused to migrate into said layer from said sperm containing sample. Sperm may be recovered from the medium after separation.

Preferably, the penetration medium contains a cellulose derivative, more preferably hydroxymethylcellulose (HMC) and most preferably methylcellulose (MC). Cellulose derivatives, such as MC and HMC are widely used in industry, they are well characterised, stable and extremely inexpensive.

The cellulose derivative preferably has a viscosity of 15-5000cp, more preferably 400-4500cp and most preferably 1500-4000cp. The concentration of cellulose derivative in solution is preferably 1-50mg/ml, more preferably 2-25mg/ml and most preferably 5-10 mg/ml.

The medium is preferably buffered at a physiological pH, for example, between a range of pH 6.5-7.6. Examples of suitable buffers include, but are not limited to, HEPES buffered media and Earle's balanced salt solution (EBSS). Protein (eg. human serum albumin) may be added to the medium, preferably at a concentration of around 1 g/L. The osmolarity of the medium is preferably between 270-300 mOsm.

The medium may be in the form of a gel (e. g. a gel column). A semen sample can be applied to the gel (e. g. at the base of a column). Where sperm are to be removed after separation, the gel may include a portion adapted for removal at a pre-determined distance from the sample.

The method may be performed at any suitable temperature, preferably within the range 17°C to 40°C, and more preferably within the range 30°C to 37°C. It is preferred that an essentially constant temperature is maintained throughout the method.

The improved medium according to the present invention results in a very inexpensive yet dependable and effective separation of motile sperm from a sperm-containing sample. The motile sperm, upon having migrated in the medium, can be removed and used for fertilisation

within an assisted reproductive technique (ART) such as intra-uterine insemination (IUI) or in vitro fertilisation (IVF).

MODES FOR CARRYING OUT THE INVENTION Sperm separation Spermatozoa penetration media were prepared by dissolving either hyaluronic acid (from rooster comb, MW~2x106) or a cellulose derivative, such as methylcellulose (obtained from Sigma-Aldrich) in Earle's balanced salt solution (EBSS-Gibco-BRL) to the test concentrations used. EBSS was used as a control.

Flattened glass capillary tubes (Camlab), with dimensions of 1.2 x 4.8mm and an inner diameter vision path of 0.4mm, were filled with the penetration media, and one end sealed with PlasticineTM. The open end of the tube was placed in a 1.5ml microfuge tube containing 1001l1-200p1 liquefied semen. Sperm were allowed to migrate into the gel for 30min at room temperature, and the tube then wiped and observed under the microscope.

Numbers of spermatozoa per field of view, using x10 or x20 objectives and a x10 eyepiece to give final magnifications of x 100 or x200, were counted on an Olympus BH-2 microscope at lcm distances from the open end of the tube. The area observed using a x10 objective is 0. 785mm2, and using the x20 objective is 0.196mm2.

Penetration of test media by spermatozoa were as follows: Initial number spermatozoa(perfield)of Test Microscope of sperm substanceobjective (%2cm3cm4cm1cm Hyaluronic acid 5mg/ml 69 12 5- lmg/ml 126 (79%) x20 300 2-- g 0.5mg/ml"150'2- 4000cp Methyl Cellulose 1402810mg/ml187 5mg/ml 120 (79%) x20 100 1- 2mg/ml 38 1 I mg/ml 40- 1500cp Methyl Cellulose IQmg/ml 68 5mg/ml 120 (79%) x20 48 2mg/ml 78 1 mg/ml 80 400cp Methyl Cellulose 10-10mg/ml98 5mg/ml 120 (79%) x20 40 5 2mg/ml 15-- 1 mg/ml 110 25cpMethyl Cellulose lOmg/ml 1-- 5mg/ml 120 (79%) x10 4 2mg/ml 2 1 mg/ml 15 15cp Methyl Cellulose 10mg/ml x20 190 5mg/ml 120 (79%) x10 41-- (1)At2.5cm.

Thus cellulose derivatives (e. g. methylcellulose) are suitable for separating motile sperm from a semen sample, performing well in comparison with HA.

Comparison of normal and oligozoospermic samples Semen samples were collected, in accordance with HFEA Code of Practice, from men after 2-4 days sexual abstinence, by masturbation into sterile plastic containers. Samples were left to liquefy at 37°C for 30 minutes. Semen analyses were performed according to WHO criteria [WHO laboratory manual for the examination of human semen and sperm-cervical mucus interaction. World Health Organization, 4th edition 1999]. Sterile disposable 5ml pipettes were used to determine the volume of the semen sample. Sperm motility was assessed on a freshly prepared wet preparation, which was examined at a total magnification at x400 using an Olympus BH2 microscope with phase contrast optics. Motility was classified according to WHO guidelines- (a) rapid progressive, (b) slow progressive, (c) non-progressive, (d) immotile. An improved Neubauer haemocytometer was used to determine the concentration of spermatozoa. Slides were prepared using a thin and well-spread smear, which was air-dried,

fixed and stained using Papanicolaou stain. Examination of the slides was performed using bright field illumination at magnification x1000.

Samples were divided into normal (>20x106 sperm/ml, 50% progressive motility, 5% normal forms) and oligozoospermic (<20x106 sperm/ml), and then split into 4 groups: MorphologyConcentrationMotility* Group Semen n (%normal(106/ml)(%) forms) 71Normal85 (20-520) 58 (31-78) 25 (6-50) A 6(0-17)Oligozoospermic39 21 (0-90) 6 (0-14) 63Normal53 (21-152) 55.5 (31-78) 24 (6-50) B OligozoospermicOligozoospermic29 6 (0-17) 13 (0-60) 8 (0-16) 86Normal42 (20-520) 60.5 (43-75) 24 (8-50) C 8Oligozoospermic13 (1-17) 38 (0-55) 4 (0-14) 63.5Normal52 (21-152) 56 (27-78) 24 (6-50) D 5(0-17)16(0-60)7(0-14)Oligozoospermic29 Figures given as median (and ranges) * classes (a) and (b) as defined by WHO standard The following penetration media were used: -4000cp methylcellulose (MC4000), at 10mg/ml and lmg/ml methylcellulose(MC15),at10mg/mland4mg/ml-15cp The media were introduced into flattened glass capillary tubes (Scm long) as described above, taking care to prevent air bubbles becoming trapped in the column. The open end of the tube was placed in a 1.5ml microfuge tube containing 100µl of the liquefied semen sample. Sperm were allowed to migrate into the gel for 30min at 24+3°C, and the tube then wiped and observed under the microscope using phase contrast optics on an Olympus BH2 microscope at magnification x200. The number of spermatozoa per high-power field (HPF, area=0.785mm2) were recorded at 1,2,3 and 4cm distances from the base of the capillary tube.

Results were as follows: Separation Penetration height Group medium Semen 3cm4cm1cm2cm MC4000 Normal 125. 6 35. 2 13. 1 5. 68 A ICmg/mlOiigozoospermic 2.69 0.74 0.36 0.00 MC4000 Normal 56. 9 7. 08 1. 23 0. 32 B 1 mg/ml Oligozoospermic 0. 59 0. 14 0. 04 0. 00 MC 15 Normal 81.4 7.79 1.76 0.51 C 10mg/ml10mg/mlOligozoospermic 0.46 0.08 0.00 0.00 MC 15 Normal 51. 2 4. 60 0. 77 0. 11 4mg/ml4mg/mlOligozoospermic 0.52 0.07 0.00 0.00

This further demonstrates the suitability of methylcellulose for separating motile sperm from a semen sample, with the number of sperm being inversely related to penetration height for both normal and oligozoospermic samples.

Formal statistical comparison of the penetration media was carried out through paired analyses, based on samples which were tested in more than one medium. Results obtained with the non-parametric two-sided Sign test for penetration numbers at 2cm were: Migration medium MC15 MC4000 MC4000 1 Omg/ml 1 mg/ml 1 Omg/ml MC15, 4mg/ml 0. 791 <0. 001 <0.001 0.508<0.001MC15,10mg/ml <0.001MC4000,1mg/ml

Similar results were obtained at 3cm and Icm with similar significance. The Wilcoxon Signed Rank test gave qualitatively similar results in each case.

Strong associations were observed between the number of penetrating sperm at lcm and the conventional parameters of concentration, motility, concentration of motile cells and morphology. Spearman Rank correlations were as follows: Rho p Concentration (x 106/ml) 0.84 <0. 001 Motility(%) 0. 70 <0.001 Concentration of motile cells (x106/ml) <0.001 Morphology (%) 0.70<0. 001

Similar results were obtained at 2,3 and 4cm penetration depths.

Temperature study To assess the effect of temperature on penetration of MC4000 (lOmg/ml), 34 normal samples and 33 oligozoospermic samples were used: Group n Concentration Motility Morphology (106/ml) (%) (% normal forms) Normal 34 77.5 (20-158) 52.5 (26-87) 19.5 (6-50) Oligozoospermic 33 9 (0-19) 32 (0-67) 3 (0-18) Penetration into the medium was tested at 17°C, 22°C, 30°C and 37°C: Penetration height Temperature Semen- lcm 2cm 3cm 4cm Normal 30. 4 16. 6 8. 88 2. 22 17°C Oligozoospermic 2.71 0.52 0.00 0.00 26.016.96.33Normal65.8 22°C Oligozoospermic 4.58 1.42 0.18 0.03 Normal 118. 7 72. 7 44. 2 10. 0 30°C Oligozoospermic 12.5 5.32 2.35 0.24 68.223.15.13Normal84.5 37°C Oligozoospermic 20.6 9. 76 3. 8 1. 4 The evidence for a temperature effect is clear (p<0.001 for penetration heights 1,2 and 3cm), and compares with the effect seen for MC [Tang et al. (1999) Hum. Reprod. 14: 2812-2817].

For normal samples, the data are consistent with a ten-fold increase in penetration numbers at each distance over the temperature range 17°C to 37°C.

Comparison of MC4000, hyaluronic acid and Sperm Select' 26 semen samples were used to compare these media. Penetration tests were performed as described above at 37°C for 30 minutes. Each sample was split into 3xlOOpl aliquots and the number of sperm penetrating MC4000 (lOmg/ml), rooster comb HA, and'Sperm Select'were recorded.

The median sperm concentration in the 26 samples was 79.5xl06/ml (range 3-159x106) and the median motility was 63% (range 37-85%). At lcm, only small differences in penetration numbers could be seen. As penetration distances increased, however, the numbers achieved in MC4000 were significantly greater (p<0.001) than either of the other two media. At these distances, no significant differences (p>0.05) between HA and'Sperm Select'were identified.

Thus MC4000 at 1 Omg/ml is extremely effective for in vitro separation of motile spermatozoa.

It will be understood that the invention has been described by way of example only and modifications may be made whilst remaining within the scope and spirit of the invention.