- Email: [email protected]
Cyclic changes in the concentration of glucose and fructose in human cervical mucus
FERTILITY AND STERILITY
Vol. 57, No.3, March 1992
Copyright e 1992 The American Fertility Society
Printed on ocid-free paper in U.S.A.
Cyclic changes in the concentration of glucose and fructose in human cervical mucus
Paul J. Q. van der Linden, M.D.*t Mienke Kets, M.Sc.:!: Joke A. Gimpel, Ph.D.:!: Maarten A. H. M. Wiegerinck, M.D., Ph.D.* Sint Joseph Hospital, Veldhoven, and Academic Hospital, State University Utrecht, Utrecht, The Netherlands
Objective: To determine a possible cyclic change in the concentration of glucose and fructose in the aqueous phase of human cervical mucus (CM). Design: Concentrations of glucose and fructose were longitudinally determined in the aqueous phase of CM of normal cycling women using enzymatic techniques, modified for small quantities. Setting: Patients visiting a fertility clinic were selected. Patients: Nine healthy women with regular menstrual cycles of 28 ± 3 days that appeared to be ovulatory, demonstrated by sonographic follicle immaging and serum progesterone (P) measurements. Interventions: Cervical mucus samples were longitudinally collected preovulatory, postovulatory, and premenstrual in ovulatory cycles, monitored by ultrasound and blood estradiol and P measurements. Main Outcome Measures: The study was designed to measure glucose and fructose longitudinally on three different points during one cycle. Results: The preovulatory glucose concentrations in CM were lower than postovulatory and premenstrual. The preovulatory fructose concentrations were lower than premenstrual. The glucose concentration correlated with the blood P level. Conclusion: There is a consistent change in the glucose concentration measured in human CM in three phases of the menstrual cycle. The preovulatory and premenstrual fructose concentrations differ significantly. Knowledge of the carbohydrate metabolism in human cervical mucus may contribute in illuminating the possible role of the carbohydrate metabolism in sperm migration at midcycle and implantation in the luteal phase. Fertil Steril1992;57:573-7 Key Words: Cervical mucus, glucose, fructose, cyclical changes
The role of the carbohydrate composition in the female genital tract in the processes of sperm transport, early embryonic development, and implantation is not well known. Human cervical mucus (eM) is a hydrogel composed of an aqueous phase with Received April 5, 1991; revised and accepted November 13, 1991. * Department of Obstetrics and Gynecology, Sint Joseph Hospital. t Reprint requests: Paul J .Q. van der Linden, M.D., Department of Obstetrics and Gynecology, Sint Joseph Hospital, Post Office Box 7777, 5500 MB Veldhoven, The Netherlands. Department of Clinical Chemistry, Academic Hospital, State University Utrecht.
low viscosity and a gel phase with high viscosity. The gel phase is composed of glycoproteins that form a tricot-like network (1). Both phases play an important role in the function of the eM. During the menstrual cycle, the percentage ofthe aqueous phase varies from 90% to 98%. There are no cyclic changes in the carbohydrate composition of the glycoproteins (2). There are some contradictory reports on the concentrations in mucus of sialic acid and fucose (3, 4). The aqueous phase or cervical plasma of eM is a mixture of soluble components that not only originate from the eM but are also derived from the peritoneum, ovaries, tubes, vagina, and by serum transudation (5, 6). It contains lipids, fatty acids,
Vol. 57, No.3, March 1992
van der Linden et aI.
*
Carbohydrates in human CM
573
immunoglobulines, prostaglandines, trace metals, serumproteins, enzymes, hormones, glucose, maltose, mannose, and lactate (1, 5, 7). The chemical composition changes during the menstrual cycle. Generally, the concentration of soluble components decreases in the middle of the menstrual cycle, probably because of a decrease in secretion and/or a dilution caused by an increase of the aqueous phase. However, some components, such as NaCI, do increase. In animal experiments (dogs and cows), a relationship is suggested between the menstrual cycle and the composition of carbohydrates, measured in CM and endometrium. In CM of cows, fructose and glucose contents are low before ovulation and high after ovulation. At the end of the cycle, large amounts of glucuronic acid and sorbitol can be found and less fructose and glucose (8). Suga (9) also demonstrated the presence of sorbitol in cow's endometrium. A deviation in this carbohydrate composition was reported to disturb or prevent fertilization in cows (10). Glucose and fructose are present in the uterine fluid of several species of animals, i.e., rats, rabbits, cows, sheep, and pigs (11). There are indications that disturbances in the carbohydrate composition playa possible role in human infertility (7, 12-14). The goal of our study was to measure cyclic changes of some carbohydrates in human CM by means of a specific and gentle assay procedure. The latest study looking into the glucose content of CM dates back 20 years (12). In the present study, the use of enzymatic techniques allowed detection of carbohydrates in small amounts of mucus. Monitoring of the menstrual cycle made longitudinal sample collection possible in relation to the time of ovulation and menstruation.
During one menstrual cycle, CM was collected at three different moments in nine patients who visited our fertility clinic. Their ages ranged from 26 to 32 years. Seven patients suffered from infertility because of a male factor. In two cases the cause of infertility was unexplained. All patients had regular menstrual cycles of 28 ± 3 days that appeared to be ovulatory, as demonstrated by sonographic follicle immaging and serum progesterone (P), measured in the luteal phase. They had no uterine anomalies and were healthy. During the collection cycle, they did not use any drugs, and they did not have sexual intercourse 24 hours before sample collection. The first CM sample was taken preovulatory, no more than
2 days before ovulation, according to sonographic criteria. The second sample was taken within 2 days after ovulation. The time of ovulation was verified by serial sonographic follicle monitoring. The third sample was taken before the expected menstruation, no more than 3 days before the start of menstruation. The time of this third sample was planned, according to expected cycle length, based on former cycles. In all cases, three samples were collected except in one patient in whom menstruation had already begun at the moment of the third collection. On every day of sample collection, a serum sample was taken for P and estradiol (E 2 ) measurements. Mucus was aspirated in a dry sterile 1.0-mL tuberculine syringe after cleaning the ectocervix by gentle sponging. The cone of the syringe was carefully inserted into the cervical canal. Aspiration was performed with gentle traction to avoid damage to the endocervical epithelium. When collecting samples in the luteal phase, it often was necessary to cut the thick mucus at the external os with a little scissor after aspiration into the syringe to avoid retraction of the mucus into the cervical canal. The samples were kept at a temperature of -70°C until further processing took place. The glycoprotein fraction of the mucus samples was separated from the aqueous phase by acid precipitation. Seventy percent HCI0 4 solution was added to the sample to a final concentration of 20% and mixed thoroughly. The samples were put on ice for 30 minutes and centrifuged for 3 minutes in a Beckman microfuge (Beckman, Annaheim, CA). An aliquot of the supernatant was neutralized with KHC0 3 (1 moljL), TRIS 10 mmoljL to a final pH between 6 and 8. The samples were frozen to precipitate the KCI0 4 • After thawing, the samples were centrifuged, and the supernatant was collected for the analysis of glucose and fructose. Glucose was assayed fluorometrically according to Lachenicht and Bernt (15), adapted to a sensitive fluorimeter (Perkin-Elmer 650-10 M, Perkin-Elmer, Pomona, CA). When the glucose assay was finished, an internal standard of glucose was added to the samples. Thereafter, the sensitivity ofthe apparatus was increased, and fructose was measured in the same cuvette according to Bernt and Bergmeyer (16) by the addition of Phosphoglucose-isomerase. After the measurement was completed, an internal standard of fructose was added. The sensitivity for glucose and fructose was 0.12 JIm (concentrations in the mucus samples). The variation coefficients were 4.6% for glucose and 5.1 % for fructose. The repeated measures ANOVA and
Carbohydrates in human CM
Fertility and Sterility
MATERIALS AND METHODS
574
van der Linden et al.
Fructose concentration mmol/L
Glucose concentration mmol/L mmol/L
mmol/L
0.8
14
0,7
12
0,6
10
0,5 80,4 8 0.3 4
0,2
2 0
0,1 0
preovulatory postovulatory
premenstrual
preovulatory postovulatory
premenstrual
Figure 1 Concentration of glucose in CM in mmoljL. The lines connect the three values of the measurements of each patient, followed longitudinally during one ovulatory cycle.
Figure 2 Concentration of fructose in CM in mmol/L. The lines connect the three values of the measurements of each patient, followed longitudinally during one ovulatory cycle.
Pearson's correlation coefficient were employed to evaluate data.
statistically significant correlations could be demonstrated (Table 1).
RESULTS
DISCUSSION
The amount of mucus obtained ranged from 10 to 460 JLL. The preovulatory glucose concentration ranged from 0.03 to 0.46 mmol/L; mean ± SD: 0.15 ± 0.17 mmoljL. The postovulatory glucose concentration ranged from 0.04 to 7.41 mmoljL; mean: 2.02 ± 2.32 mmoljL. The premenstrual glucose concentration ranged from 1.16 to 12.39 mmoljL; mean: 4.49 ± 3.37 mmol/L. (Fig. 1.) The preovulatory glucose concentration in one patient always was lower than the postovulatory concentration (P < 0.05). The preovulatory glucose concentration was also lower than the premenstrual values (P < 0.02). The preovulatory fructose concentration ranged from 0.01 to 0.10 mmol/L; mean ± SD: 0.038 ± 0.031 mmoljL. The postovulatory fructose concentration ranged from 0.02 to 0.33 mmoljL; mean: 0.12 ± 0.11 mmoljL. The premenstrual fructose concentration ranged from 0.02 to 0.38 mmoljL; mean: 0.18 ± 0.12 mmoljL. (Fig. 2.) Only the preovulatory fructose concentrations were significantly lower than the premenstrual concentrations (P < 0.05). Seven of eight subjects showed continuous increases in glucose from pre ovulation to postovulation to premenses; in contrast, the fructose patterns were quite heterogeneous. There was a positive correlation between the concentration of glucose in the cervical mucus and the serum P (r = 0.58; P < 0.002) (Fig. 3). The concentration of fructose correlated with serum E2 (r = -0.71; P < 0.001) (Fig. 4). No other
Our study shows that there is a consistent change in the glucose concentration measured in human CM in three phases of the menstrual cycle. The fructose concentration preovulatory and premenstrual differ significantly. Weed and Carrera (12) studied cyclic changes of glucose in CM. They concluded that the amount of glucose rose to a peak at ovulation. Reviewing the study of Weed and Carrera, one could however find a more general incremental progression in the values
Figure 3 The correlation between the concentration of glucose in CM in mmol/L and serum P in nmol/L. (r = 0.58; P < 0.002).
Vol. 57, No.3, March 1992
van der Linden et al.
Correlation between glucoae and progeaterone Glucose mmol/L
14 12 10 8 8 4 2 0 0
50
100 Progesterone nmol/L
150
Carbohydrates in human CM
200
575
Correlation between estradiol and fructose
Douglas et al. (14) studied glucose and fructose concentrations in the endometrium by introducing a semipermeable diffusion chamber in the uterus. Glucose concentration ranged from 10 to 200 mg/ 100 mL. The fructose range was 5 to 55 mg/100 mL. Within one patient, the level of glucose was found to be constant. Higher fructose concentrations were found after the 11th day of the cycle. In vaginal fluid, fructose followed the same pattern. Glucose levels were much lower than those in the endometrium (14). In this study, the midcyde group also contained preovulatory and postovulatory samples. Furthermore, the introduction of a diffusion chamber can result in a foreign body reaction, possibly leading to an increased endometrial permeability. In our study, however, fructose levels were also higher during the second half of the cycle. Hughes et al. (13) found in a transversally conducted study the glucose concentration in uterine fluid to be the highest during the 14th to the 16th day of the cycle. In a heterogenic group of patients with impaired fertility, they found lower values of glucose. Concomitantly, they reported the glycogen content of the endometrium to be lower during the proliferate phase and to be the highest from day 17 to day 24 of the cycle. They concluded that the endometrium metabolizes carbohydrates, which influence the implantation of the fertilized ovum (13). Assuming that the reported changes of the glucose concentrations i.n the endometrium are reflected in the CM composition, it is plausible that the highest concentrations of glucose in the endometrium are found postovulatory. The high levels of glucose that we found in CM are compatible with an active secretion of carbohydrates postovulatory in the female genital tract. A good correlation between serum P and the concentration of glucose in the CM was
Fructose mmol/l
0,4
0,3
0,2
-----------------
0,1
,
o
-----, -~------'----r----o
-----,--------
40
20
---,--'-----~
80
60
100
Estradiol nmolll
Figure 4 The correlation between the concentration of fructose in CM in mmol/L and serum E2 in nmol/L. (r = -0.71; P < 0.001).
they presented. In their study, the time of ovulation was assessed using basal body temperature measurements. They also reported lower glucose values in patients with unexplained infertility (12). Our study, using ultrasound to determine ovulation, demonstrated high glucose levels postovulatory but not preovulatory. CasslEm and Nilsson (17) measured glucose and fructose concentrations in uterine fluid. The concentrations were not much different from those found in serum. An insignificant increase in glucose concentration was observed at midcycle (17). It should be mentioned that the group at midcycle consisted of preovulatory and postovulatory samples. Another difference with our study is the fact that they did not sample and measure longitudinally, but measurements were performed in pooled samples.
Table 1
The Concentration of Glucose and Fructose in the Aqueous Phase of CM on Three Different Phases During One Cycle Preovulatory
Patient
Glucose
1 2 3 4 5 6 7 8 9
0.04 0.05 0.03 0.03 0.43 0.11 0.46 0.03 0.13
Fructose
mmol/L
Postovulatory
E2
P
Glucose
7.7 3.3 4.3 2.0 2.2 3.2 4.6 1.0 1.4
7.41 2.25 0.04 0.17 0.48 0.47 2.16 1.87 3.37
nmol/L
0.02 0.02 0.02 0.01 0.10 0.02 0.07 0.02 0.06
0.52 0.54 0.91 0.89 0.27 0.54 0.60 0.65 0.53
Fructose
Premenstrual P
Glucose
160 24 7.0 26 47 22 25 28 13
1.49 5.77 ND 1.4 12.39 1.16 3.77 6.12 3.78
E2
mmol/L
nmol/L
NDa 0.33 0.02 0.04 0.11 0.06 0.02 0.17 0.19
0.50 0.21 0.45 0.29 0.39 0.22 0.35 0.32 0.29
Fructose
mmol/L
P
E2 nmol/L
0.13 0.25 ND 0.28 0.11 0.08 0.02 0.20 0.38
0.23 0.33 ND 0.13 0.60 0.22 0.14 0.40 0.53
15 15 ND 4.4 58 22 4.2 22 32
a ND, not done.
576
van der Linden et al.
Carbohydrates in human CM
Fertility and Sterility
found, although the two hexose/steroid correlations show a rather large scatter that diminishes the significance of these correlations (Figs. 3 and 4.). Whether the elevation of glucose in the CM is initiated by P has to be established. Clinical significance of knowledge of the metabolism of these substances lies in their possible role in sperm migration at midcycle on one hand and ovum implantation in the luteal phase on the other hand. The advantage of using CM instead of endometrium is that aspiration of CM does not interfere with the endometrium. Using this technique, it was possible to measure carbohydrates in very small amounts of CM without the need for aggressive chemical modifications. Further research has to show if there are differences in the cyclic concentration changes of carbohydrates between fertile and infertile women. REFERENCES 1. Schumacher GFB. Biochemistry of cervical mucus. Fertil Steril1970;21:697-705. 2. Van Kooij RJ, Roelofs HJM, Kathmann GAM, Kramer MF. Human cervical mucus and its mucous glycoprotein during the menstrual cycle. Fertil Steril 1980;34:226-33. 3. WolfDP, Sokolski JE, Litt M. Composition and function of human cervical mucus. Biochim Biophys Acta 1980;630:545558. 4. Iacobelli S, Garcea N, Angeloni C. Biochemistry of cervical mucus: a comparative analysis of the secretion from preovulatory, postovulatory and pregnancy periods. Fertil Steril 1971;22:727-34. 5. Daunter B. Biochemical and functional-structural aspects of human cervical mucus. Scanning Electron Microsc 1984;1: 343-58. 6. Mettler L. Cervical factors in infertility. Curr Opin Obstet Gynecol 1990;2:193-9.
Vol. 57, No.3, March 1992
7. Gladdines MM, Ackermans MT, Everaerts FM, van der Linden PJQ, Wiegerinck MAHM. Analysis of the aqueous phase of human cervical mucus by reversed-phase high performance liquid chromatography and capillary isotachophoresis. J Chromatogr 1988;431:317-27. 8. Van der Horst CJG. Carbohydrate metabolism and the temporal occurrence of a serotonin-like indole in the male and female reproductive tract in relationship with PGF and infertility: a review. Cytobios 1986;45:85-95. 9. Suga T. Pathways of carbohydrate in the bovine endometrial chorionic unit as an embryonic nutrient. JARQ 1975;4:22530. 10. Zaayer D, van der Horst CJG. Cyclical changes in hormones, carbohydrates and indole metabolism in cervical mucus of normal, fertilizing cows and the relationship with non-fertility. Cytobios 1983;37:113-27. 11. Zavy MT, Clark WR, Sharp DC, Robberts RM, Bazer FW. Comparison of glucose, fructose, ascorbic acid and glucose phosphate isomerase enzymatic activity in uterine flushings from non-pregnant and pregnant gilts and pony mares. Bioi Reprod 1982;27:1147-58. 12. Weed JC, Carrera AE. Glucose content of cervical mucus. Fertil Steril 1970;21:866-72. 13. Hughes EC, Jacobs RD, Rubulis A, Husney RM. Carbohydrate pathways of the endometrium. Effects on ovular growth. Am J Obstet Gynecol 1963;85:594-609. 14. Douglas CP, Garrow JS, Pugh EW. Investigation into the sugar content of endometrial secretions. J Obstet Gynaecol Br Commonw 1970;77:891-4. 15. Lachenicht R, Bernt E. Fluorimetrische Bestimmung van Dglucose im Blut mit Analyse-Automaten. In: Bergmeyer HU, editor. Methoden der enzymatischen Analyse. Weinheim: Verlag Chemie, 1974:1241-6. 16. Bernt E, Bergmeyer HU. Die Fructose. In: Bergmeyer HU, editor. Methoden der enzymatischen Analyse. Weinheim: Verlag Chemie, 1974:1349-52. 17. Casslen B, Nilsson B. Human uterine fluid, examined in undiluted samples for osmolarity and the concentrations of inorganic ions, albumin, glucose and urea. Am J Obstet Gynecol 1984;150:877-81.
van der Linden et al.
Carbohydrates in human CM
577
Vol. 57, No.3, March 1992
Copyright e 1992 The American Fertility Society
Printed on ocid-free paper in U.S.A.
Cyclic changes in the concentration of glucose and fructose in human cervical mucus
Paul J. Q. van der Linden, M.D.*t Mienke Kets, M.Sc.:!: Joke A. Gimpel, Ph.D.:!: Maarten A. H. M. Wiegerinck, M.D., Ph.D.* Sint Joseph Hospital, Veldhoven, and Academic Hospital, State University Utrecht, Utrecht, The Netherlands
Objective: To determine a possible cyclic change in the concentration of glucose and fructose in the aqueous phase of human cervical mucus (CM). Design: Concentrations of glucose and fructose were longitudinally determined in the aqueous phase of CM of normal cycling women using enzymatic techniques, modified for small quantities. Setting: Patients visiting a fertility clinic were selected. Patients: Nine healthy women with regular menstrual cycles of 28 ± 3 days that appeared to be ovulatory, demonstrated by sonographic follicle immaging and serum progesterone (P) measurements. Interventions: Cervical mucus samples were longitudinally collected preovulatory, postovulatory, and premenstrual in ovulatory cycles, monitored by ultrasound and blood estradiol and P measurements. Main Outcome Measures: The study was designed to measure glucose and fructose longitudinally on three different points during one cycle. Results: The preovulatory glucose concentrations in CM were lower than postovulatory and premenstrual. The preovulatory fructose concentrations were lower than premenstrual. The glucose concentration correlated with the blood P level. Conclusion: There is a consistent change in the glucose concentration measured in human CM in three phases of the menstrual cycle. The preovulatory and premenstrual fructose concentrations differ significantly. Knowledge of the carbohydrate metabolism in human cervical mucus may contribute in illuminating the possible role of the carbohydrate metabolism in sperm migration at midcycle and implantation in the luteal phase. Fertil Steril1992;57:573-7 Key Words: Cervical mucus, glucose, fructose, cyclical changes
The role of the carbohydrate composition in the female genital tract in the processes of sperm transport, early embryonic development, and implantation is not well known. Human cervical mucus (eM) is a hydrogel composed of an aqueous phase with Received April 5, 1991; revised and accepted November 13, 1991. * Department of Obstetrics and Gynecology, Sint Joseph Hospital. t Reprint requests: Paul J .Q. van der Linden, M.D., Department of Obstetrics and Gynecology, Sint Joseph Hospital, Post Office Box 7777, 5500 MB Veldhoven, The Netherlands. Department of Clinical Chemistry, Academic Hospital, State University Utrecht.
low viscosity and a gel phase with high viscosity. The gel phase is composed of glycoproteins that form a tricot-like network (1). Both phases play an important role in the function of the eM. During the menstrual cycle, the percentage ofthe aqueous phase varies from 90% to 98%. There are no cyclic changes in the carbohydrate composition of the glycoproteins (2). There are some contradictory reports on the concentrations in mucus of sialic acid and fucose (3, 4). The aqueous phase or cervical plasma of eM is a mixture of soluble components that not only originate from the eM but are also derived from the peritoneum, ovaries, tubes, vagina, and by serum transudation (5, 6). It contains lipids, fatty acids,
Vol. 57, No.3, March 1992
van der Linden et aI.
*
Carbohydrates in human CM
573
immunoglobulines, prostaglandines, trace metals, serumproteins, enzymes, hormones, glucose, maltose, mannose, and lactate (1, 5, 7). The chemical composition changes during the menstrual cycle. Generally, the concentration of soluble components decreases in the middle of the menstrual cycle, probably because of a decrease in secretion and/or a dilution caused by an increase of the aqueous phase. However, some components, such as NaCI, do increase. In animal experiments (dogs and cows), a relationship is suggested between the menstrual cycle and the composition of carbohydrates, measured in CM and endometrium. In CM of cows, fructose and glucose contents are low before ovulation and high after ovulation. At the end of the cycle, large amounts of glucuronic acid and sorbitol can be found and less fructose and glucose (8). Suga (9) also demonstrated the presence of sorbitol in cow's endometrium. A deviation in this carbohydrate composition was reported to disturb or prevent fertilization in cows (10). Glucose and fructose are present in the uterine fluid of several species of animals, i.e., rats, rabbits, cows, sheep, and pigs (11). There are indications that disturbances in the carbohydrate composition playa possible role in human infertility (7, 12-14). The goal of our study was to measure cyclic changes of some carbohydrates in human CM by means of a specific and gentle assay procedure. The latest study looking into the glucose content of CM dates back 20 years (12). In the present study, the use of enzymatic techniques allowed detection of carbohydrates in small amounts of mucus. Monitoring of the menstrual cycle made longitudinal sample collection possible in relation to the time of ovulation and menstruation.
During one menstrual cycle, CM was collected at three different moments in nine patients who visited our fertility clinic. Their ages ranged from 26 to 32 years. Seven patients suffered from infertility because of a male factor. In two cases the cause of infertility was unexplained. All patients had regular menstrual cycles of 28 ± 3 days that appeared to be ovulatory, as demonstrated by sonographic follicle immaging and serum progesterone (P), measured in the luteal phase. They had no uterine anomalies and were healthy. During the collection cycle, they did not use any drugs, and they did not have sexual intercourse 24 hours before sample collection. The first CM sample was taken preovulatory, no more than
2 days before ovulation, according to sonographic criteria. The second sample was taken within 2 days after ovulation. The time of ovulation was verified by serial sonographic follicle monitoring. The third sample was taken before the expected menstruation, no more than 3 days before the start of menstruation. The time of this third sample was planned, according to expected cycle length, based on former cycles. In all cases, three samples were collected except in one patient in whom menstruation had already begun at the moment of the third collection. On every day of sample collection, a serum sample was taken for P and estradiol (E 2 ) measurements. Mucus was aspirated in a dry sterile 1.0-mL tuberculine syringe after cleaning the ectocervix by gentle sponging. The cone of the syringe was carefully inserted into the cervical canal. Aspiration was performed with gentle traction to avoid damage to the endocervical epithelium. When collecting samples in the luteal phase, it often was necessary to cut the thick mucus at the external os with a little scissor after aspiration into the syringe to avoid retraction of the mucus into the cervical canal. The samples were kept at a temperature of -70°C until further processing took place. The glycoprotein fraction of the mucus samples was separated from the aqueous phase by acid precipitation. Seventy percent HCI0 4 solution was added to the sample to a final concentration of 20% and mixed thoroughly. The samples were put on ice for 30 minutes and centrifuged for 3 minutes in a Beckman microfuge (Beckman, Annaheim, CA). An aliquot of the supernatant was neutralized with KHC0 3 (1 moljL), TRIS 10 mmoljL to a final pH between 6 and 8. The samples were frozen to precipitate the KCI0 4 • After thawing, the samples were centrifuged, and the supernatant was collected for the analysis of glucose and fructose. Glucose was assayed fluorometrically according to Lachenicht and Bernt (15), adapted to a sensitive fluorimeter (Perkin-Elmer 650-10 M, Perkin-Elmer, Pomona, CA). When the glucose assay was finished, an internal standard of glucose was added to the samples. Thereafter, the sensitivity ofthe apparatus was increased, and fructose was measured in the same cuvette according to Bernt and Bergmeyer (16) by the addition of Phosphoglucose-isomerase. After the measurement was completed, an internal standard of fructose was added. The sensitivity for glucose and fructose was 0.12 JIm (concentrations in the mucus samples). The variation coefficients were 4.6% for glucose and 5.1 % for fructose. The repeated measures ANOVA and
Carbohydrates in human CM
Fertility and Sterility
MATERIALS AND METHODS
574
van der Linden et al.
Fructose concentration mmol/L
Glucose concentration mmol/L mmol/L
mmol/L
0.8
14
0,7
12
0,6
10
0,5 80,4 8 0.3 4
0,2
2 0
0,1 0
preovulatory postovulatory
premenstrual
preovulatory postovulatory
premenstrual
Figure 1 Concentration of glucose in CM in mmoljL. The lines connect the three values of the measurements of each patient, followed longitudinally during one ovulatory cycle.
Figure 2 Concentration of fructose in CM in mmol/L. The lines connect the three values of the measurements of each patient, followed longitudinally during one ovulatory cycle.
Pearson's correlation coefficient were employed to evaluate data.
statistically significant correlations could be demonstrated (Table 1).
RESULTS
DISCUSSION
The amount of mucus obtained ranged from 10 to 460 JLL. The preovulatory glucose concentration ranged from 0.03 to 0.46 mmol/L; mean ± SD: 0.15 ± 0.17 mmoljL. The postovulatory glucose concentration ranged from 0.04 to 7.41 mmoljL; mean: 2.02 ± 2.32 mmoljL. The premenstrual glucose concentration ranged from 1.16 to 12.39 mmoljL; mean: 4.49 ± 3.37 mmol/L. (Fig. 1.) The preovulatory glucose concentration in one patient always was lower than the postovulatory concentration (P < 0.05). The preovulatory glucose concentration was also lower than the premenstrual values (P < 0.02). The preovulatory fructose concentration ranged from 0.01 to 0.10 mmol/L; mean ± SD: 0.038 ± 0.031 mmoljL. The postovulatory fructose concentration ranged from 0.02 to 0.33 mmoljL; mean: 0.12 ± 0.11 mmoljL. The premenstrual fructose concentration ranged from 0.02 to 0.38 mmoljL; mean: 0.18 ± 0.12 mmoljL. (Fig. 2.) Only the preovulatory fructose concentrations were significantly lower than the premenstrual concentrations (P < 0.05). Seven of eight subjects showed continuous increases in glucose from pre ovulation to postovulation to premenses; in contrast, the fructose patterns were quite heterogeneous. There was a positive correlation between the concentration of glucose in the cervical mucus and the serum P (r = 0.58; P < 0.002) (Fig. 3). The concentration of fructose correlated with serum E2 (r = -0.71; P < 0.001) (Fig. 4). No other
Our study shows that there is a consistent change in the glucose concentration measured in human CM in three phases of the menstrual cycle. The fructose concentration preovulatory and premenstrual differ significantly. Weed and Carrera (12) studied cyclic changes of glucose in CM. They concluded that the amount of glucose rose to a peak at ovulation. Reviewing the study of Weed and Carrera, one could however find a more general incremental progression in the values
Figure 3 The correlation between the concentration of glucose in CM in mmol/L and serum P in nmol/L. (r = 0.58; P < 0.002).
Vol. 57, No.3, March 1992
van der Linden et al.
Correlation between glucoae and progeaterone Glucose mmol/L
14 12 10 8 8 4 2 0 0
50
100 Progesterone nmol/L
150
Carbohydrates in human CM
200
575
Correlation between estradiol and fructose
Douglas et al. (14) studied glucose and fructose concentrations in the endometrium by introducing a semipermeable diffusion chamber in the uterus. Glucose concentration ranged from 10 to 200 mg/ 100 mL. The fructose range was 5 to 55 mg/100 mL. Within one patient, the level of glucose was found to be constant. Higher fructose concentrations were found after the 11th day of the cycle. In vaginal fluid, fructose followed the same pattern. Glucose levels were much lower than those in the endometrium (14). In this study, the midcyde group also contained preovulatory and postovulatory samples. Furthermore, the introduction of a diffusion chamber can result in a foreign body reaction, possibly leading to an increased endometrial permeability. In our study, however, fructose levels were also higher during the second half of the cycle. Hughes et al. (13) found in a transversally conducted study the glucose concentration in uterine fluid to be the highest during the 14th to the 16th day of the cycle. In a heterogenic group of patients with impaired fertility, they found lower values of glucose. Concomitantly, they reported the glycogen content of the endometrium to be lower during the proliferate phase and to be the highest from day 17 to day 24 of the cycle. They concluded that the endometrium metabolizes carbohydrates, which influence the implantation of the fertilized ovum (13). Assuming that the reported changes of the glucose concentrations i.n the endometrium are reflected in the CM composition, it is plausible that the highest concentrations of glucose in the endometrium are found postovulatory. The high levels of glucose that we found in CM are compatible with an active secretion of carbohydrates postovulatory in the female genital tract. A good correlation between serum P and the concentration of glucose in the CM was
Fructose mmol/l
0,4
0,3
0,2
-----------------
0,1
,
o
-----, -~------'----r----o
-----,--------
40
20
---,--'-----~
80
60
100
Estradiol nmolll
Figure 4 The correlation between the concentration of fructose in CM in mmol/L and serum E2 in nmol/L. (r = -0.71; P < 0.001).
they presented. In their study, the time of ovulation was assessed using basal body temperature measurements. They also reported lower glucose values in patients with unexplained infertility (12). Our study, using ultrasound to determine ovulation, demonstrated high glucose levels postovulatory but not preovulatory. CasslEm and Nilsson (17) measured glucose and fructose concentrations in uterine fluid. The concentrations were not much different from those found in serum. An insignificant increase in glucose concentration was observed at midcycle (17). It should be mentioned that the group at midcycle consisted of preovulatory and postovulatory samples. Another difference with our study is the fact that they did not sample and measure longitudinally, but measurements were performed in pooled samples.
Table 1
The Concentration of Glucose and Fructose in the Aqueous Phase of CM on Three Different Phases During One Cycle Preovulatory
Patient
Glucose
1 2 3 4 5 6 7 8 9
0.04 0.05 0.03 0.03 0.43 0.11 0.46 0.03 0.13
Fructose
mmol/L
Postovulatory
E2
P
Glucose
7.7 3.3 4.3 2.0 2.2 3.2 4.6 1.0 1.4
7.41 2.25 0.04 0.17 0.48 0.47 2.16 1.87 3.37
nmol/L
0.02 0.02 0.02 0.01 0.10 0.02 0.07 0.02 0.06
0.52 0.54 0.91 0.89 0.27 0.54 0.60 0.65 0.53
Fructose
Premenstrual P
Glucose
160 24 7.0 26 47 22 25 28 13
1.49 5.77 ND 1.4 12.39 1.16 3.77 6.12 3.78
E2
mmol/L
nmol/L
NDa 0.33 0.02 0.04 0.11 0.06 0.02 0.17 0.19
0.50 0.21 0.45 0.29 0.39 0.22 0.35 0.32 0.29
Fructose
mmol/L
P
E2 nmol/L
0.13 0.25 ND 0.28 0.11 0.08 0.02 0.20 0.38
0.23 0.33 ND 0.13 0.60 0.22 0.14 0.40 0.53
15 15 ND 4.4 58 22 4.2 22 32
a ND, not done.
576
van der Linden et al.
Carbohydrates in human CM
Fertility and Sterility
found, although the two hexose/steroid correlations show a rather large scatter that diminishes the significance of these correlations (Figs. 3 and 4.). Whether the elevation of glucose in the CM is initiated by P has to be established. Clinical significance of knowledge of the metabolism of these substances lies in their possible role in sperm migration at midcycle on one hand and ovum implantation in the luteal phase on the other hand. The advantage of using CM instead of endometrium is that aspiration of CM does not interfere with the endometrium. Using this technique, it was possible to measure carbohydrates in very small amounts of CM without the need for aggressive chemical modifications. Further research has to show if there are differences in the cyclic concentration changes of carbohydrates between fertile and infertile women. REFERENCES 1. Schumacher GFB. Biochemistry of cervical mucus. Fertil Steril1970;21:697-705. 2. Van Kooij RJ, Roelofs HJM, Kathmann GAM, Kramer MF. Human cervical mucus and its mucous glycoprotein during the menstrual cycle. Fertil Steril 1980;34:226-33. 3. WolfDP, Sokolski JE, Litt M. Composition and function of human cervical mucus. Biochim Biophys Acta 1980;630:545558. 4. Iacobelli S, Garcea N, Angeloni C. Biochemistry of cervical mucus: a comparative analysis of the secretion from preovulatory, postovulatory and pregnancy periods. Fertil Steril 1971;22:727-34. 5. Daunter B. Biochemical and functional-structural aspects of human cervical mucus. Scanning Electron Microsc 1984;1: 343-58. 6. Mettler L. Cervical factors in infertility. Curr Opin Obstet Gynecol 1990;2:193-9.
Vol. 57, No.3, March 1992
7. Gladdines MM, Ackermans MT, Everaerts FM, van der Linden PJQ, Wiegerinck MAHM. Analysis of the aqueous phase of human cervical mucus by reversed-phase high performance liquid chromatography and capillary isotachophoresis. J Chromatogr 1988;431:317-27. 8. Van der Horst CJG. Carbohydrate metabolism and the temporal occurrence of a serotonin-like indole in the male and female reproductive tract in relationship with PGF and infertility: a review. Cytobios 1986;45:85-95. 9. Suga T. Pathways of carbohydrate in the bovine endometrial chorionic unit as an embryonic nutrient. JARQ 1975;4:22530. 10. Zaayer D, van der Horst CJG. Cyclical changes in hormones, carbohydrates and indole metabolism in cervical mucus of normal, fertilizing cows and the relationship with non-fertility. Cytobios 1983;37:113-27. 11. Zavy MT, Clark WR, Sharp DC, Robberts RM, Bazer FW. Comparison of glucose, fructose, ascorbic acid and glucose phosphate isomerase enzymatic activity in uterine flushings from non-pregnant and pregnant gilts and pony mares. Bioi Reprod 1982;27:1147-58. 12. Weed JC, Carrera AE. Glucose content of cervical mucus. Fertil Steril 1970;21:866-72. 13. Hughes EC, Jacobs RD, Rubulis A, Husney RM. Carbohydrate pathways of the endometrium. Effects on ovular growth. Am J Obstet Gynecol 1963;85:594-609. 14. Douglas CP, Garrow JS, Pugh EW. Investigation into the sugar content of endometrial secretions. J Obstet Gynaecol Br Commonw 1970;77:891-4. 15. Lachenicht R, Bernt E. Fluorimetrische Bestimmung van Dglucose im Blut mit Analyse-Automaten. In: Bergmeyer HU, editor. Methoden der enzymatischen Analyse. Weinheim: Verlag Chemie, 1974:1241-6. 16. Bernt E, Bergmeyer HU. Die Fructose. In: Bergmeyer HU, editor. Methoden der enzymatischen Analyse. Weinheim: Verlag Chemie, 1974:1349-52. 17. Casslen B, Nilsson B. Human uterine fluid, examined in undiluted samples for osmolarity and the concentrations of inorganic ions, albumin, glucose and urea. Am J Obstet Gynecol 1984;150:877-81.
van der Linden et al.
Carbohydrates in human CM
577