Standardization and Quality Control for Determination of Fructose in Seminal Plasma

In this age of evidence-based medicine, there is nothing more important than the quality of lab tests (Westgard and Darcy, 2004). However, reliability of semen analysis has been suspected for lack of standardization, and significant variation exists within the same laboratory and among different laboratories (Chong et al, 1983; Keel et al, 2000; Keel et al, 2002b; Brazil et al, 2004; Keel, 2004). In recent years, andrologists have placed greater emphasis on standardization of semen analysis (Keel, 2002a; Huang and Li, 2005). At present, quality controls for determinations of sperm concentration, motility, morphology, and antisperm antibodies have been emphasized widely (Cooper et al, 2002; Toft et al, 2005). However, few reports exist on standardization and quality control for biochemical analysis of seminal plasma (Lu et al, 2007). Therefore, we attempted to establish standardization and quality control for the determination of fructose concentration in seminal plasma. Determination of seminal fructose concentration has been used in examination of obstructive azoospermia and inflammation of male accessory glands (Carpino et al, 1997; Manivannan et al, 2005). Inflammation may lead to atrophy of the seminal vesicle and low seminal fructose concentration. When ejaculatory ducts are blocked, fructose concentration in seminal plasma usually decreases, and may become undetectable (Coppens, 1997). Additionally, determination of fructose concentration in seminal plasma is useful to auxiliary diagnosis of obstructive and nonobstructive azoospermia. Seminal fructose concentration in nonobstructive azoospermia is usually higher than or equal to that in males of normal fertility (Buckett and Lewis-Jones, 2002). However, fructose concentration in seminal plasma of patients with obstructive azoospermia is usually absent or significantly lower than that in men of normal fertility (Manivannan et al, 2005). Absence of seminal fructose is also found in patients with congenital vas deferens-seminal vesicle developmental defect (Kise et al, 2000; Kumar et al, 2005). Determination of fructose concentration has also been used for auxiliary diagnosis of retrograde ejaculation, where fructose concentration in urine from the bladder after ejaculation was detected. Thus, the accuracy of fructose concentration determination in seminal plasma is very important. To guarantee accuracy of fructose determination, assays of fructose in seminal plasma must be standardized, and quality control should be involved. Here, we report effects of standard fructose solution, centrifugation velocity, standing time of semen and seminal plasma, freezing-thawing, and chymotrypsin exposure on detection of fructose concentrations in seminal plasma. Lastly, we compare fructose concentrations determined by 2 technicians. Semen analysis was performed with a computer-assisted sperm analysis (CASA) system (WLJY-9000; Beijing Weili New Century Science & Tech Dev Co, Ltd, Beijing, China). Spectrophotometer (721 type) was provided by Shanghai Analysis Instrument Factory (Shanghai, China). KDC-1044 Low Speed Tabletop Centrifuge was provided by USTC Chuangxin Corporation Limited ZONKIA Branch (Hefei, China). Both low ([18 ± 2.5] × 106/mL) and high ([35 ± 5] × 106/mL) precalibrated standard latex bead solutions (Hamilton Thorne Biosciences, Beverly, Mass) were used as quality control solutions for sperm concentration. Chymotrypsin for injection (5 mg/4000 U, Lot 0506042), dissolved with 0.5 mL of normal saline to be 8000 U/mL of stock solution, was provided by the Department of Pharmacy, Nanjing Jinling Hospital. D-fructose (analytical reagent), prepared to be 2.78 mmol/L of stock solution with distilled water, was the product of the Chemical Reagent Limited Company, Shanghai Guoyao Bloc (Shanghai, China). Semen samples, collected by masturbation, were from outpatients of the Department of Andrology, Nanjing Jinling Hospital. All subjects signed informed consents to participate in this study, which was approved by the Ethics Committee of Nanjing Jinling Hospital. After liquefaction, routine analysis of semen samples was performed, including assessment of semen volume, pH, sperm concentration, motility, and grade a and b motility, and the remaining samples were applied to our investigation. Standard fructose solution (0.28 mmol/L) was prepared with stock solution by adding distilled water and stored at 4°C, then stability was continuously monitored for 35 days by the resorcinol method. One mL of standard fructose solution, 1 mL of resorcinol solution (8.47 mmol/L), and 3 mL of HCl (10 mol/L) were mixed and maintained at 90°C for 10 minutes. Standard fructose solution was replaced with distilled water as blanks. Lastly, the optical density (OD) values were read at 490 nm against blanks. Twenty samples of fresh liquefied semen were randomly chosen and centrifuged at 3000 × g for 15 minutes. Then fructose concentrations in seminal plasma, after standing at room temperature (20–25°C) for 0, 2, 4, or 6 hours, were detected as previously described (Huang and Xu, 1999). Fructose in seminal plasma reacts with resorcinol in concentrated HCl solution to form a red compound under heating. In detail, 0.1 mL of fresh seminal plasma sample was mixed with 2.9 mL of distilled water. Then 0.5 mL of Ba(OH)2 solution (0.15 mol/L) and 0.5 mL of ZnSO4 solution (0.175 mol/L) were added, mixed, and allowed to stand for 5 minutes to remove seminal proteins. After centrifugation at 3000 × g for 15 minutes, 1 mL of the supernatant was collected for determination of fructose level. The supernatant was replaced with standard fructose solution (0.28 mmol/L) to serve as standard and replaced with distilled water as blanks. Subsequently, 1 mL of resorcinol solution (8.47 mmol/L) and 3 mL of HCl (10 mol/L) were added into tubes and maintained at 90°C for 10 minutes. Lastly, absorbance values were read at 490 nm against blanks. Fructose concentration in seminal plasma was expressed as mmol/L: absorbance value of test tube/absorbance value of standard tube × 11.12. Forty-eight samples of fresh liquefied semen were randomly chosen and divided into triplicate. One was centrifuged immediately (0 hours) at 3000 × g for 15 minutes, and the other 2 were centrifuged in a similar manner after standing at room temperature (20°–25°C) for 2 hours or 4 hours. Fructose concentration in seminal plasma was detected by the described method. Ten fresh samples of semen were randomly chosen and washed twice with normal saline. Sperm suspensions of various concentrations were prepared for detection of fructose concentration. Each of 68 samples of fresh semen was divided into 2 aliquots, one centrifuged at 1000 × g for 10 minutes and the other at 3000 × g for 15 minutes. Fructose and sperm concentrations in seminal plasma were detected by the resorcinol method and CASA system with Makler chamber, respectively. Precalibrated standard latex bead solutions were used as quality control solutions to provide accuracy of sperm count. In detail, a drop of 5-μL standard bead solution or seminal plasma obtained at 1000 × g for 10 minutes and 3000 × g for 15 minutes was put on chambers. To ensure the accuracy of bead concentration or sperm counting, more than 10 fields or a minimum of 200 beads/sperm were counted for each chamber. Each of 72 samples of fresh liquefied semen was divided into 2 aliquots, one without chymotrypsin and the other with 1% (v/v) of chymotrypsin (8000 U/mL). Samples were incubated at 37°C for 30 minutes, centrifuged at 3000 × g for 15 minutes, and analyzed for seminal plasma fructose concentrations. Thirty-nine samples of fresh nonliquefied semen, obtained 1 hour after collection, were incubated with 1% (v/v) of chymotrypsin (8000 U/mL) at 37°C for 30 minutes and centrifuged at 3000 × g for 15 minutes. Fructose concentration in seminal plasma was detected. An additional 2 samples of seminal plasma, obtained by centrifugation at 3000 × g for 15 minutes without sperm residue, as confirmed by CASA, were frozen. Every other day, both samples were thawed to determine the fructose concentration. The remains were then refrozen for subsequent tests. The process of thawing and refreezing was repeated 10 times within 20 days. Ten samples of fresh semen were centrifuged at 3000 × g for 15 minutes, and resultant seminal plasma samples were divided into 2 aliquots to be blindly determined by 2 technicians. Data were expressed as mean ± SD and analyzed with SPSS 11.0 software (SPSS Inc, Chicago, Ill). Comparisons of fructose concentrations in seminal plasma with different standing durations, obtained from centrifuged semen, with or without chymotrypsin were analyzed by paired t test. Comparisons between fructose concentrations and between sperm concentrations in seminal plasma obtained by centrifugation at different velocities were also analyzed by paired t test. Comparison of fructose concentrations detected by 2 technicians was shown with Bland and Altman plot. Correlation between decrease of fructose concentration in seminal plasma obtained from centrifuged semen after different standing times and motile sperm concentration was analyzed by Pearson's correlation. Comparison of fructose concentration between liquefied and nonliquefied semen samples was analyzed by independent-sample t test. Level of significance was set at P less than .05. The stability of standard fructose solution was monitored for 35 days successively, as indicated in Figure 1. The variation of optical density (OD) was wide within the first 2 weeks after preparation of standard fructose solution. The OD value decreased slightly and remained stable during the following 2 weeks. About 1 month later, the OD value decreased quickly. . OD value change of standard fructose solution with time. Fructose concentrations in seminal plasma standing for 0, 2, 4, or 6 hours after centrifugation of semen samples are shown in Table 1, indicating that there was no significant difference of fructose concentration in seminal plasma with different standing times (P > .05). Fructose concentrations in seminal plasma obtained from centrifuged semen after standing for 0, 2, or 4 hours are shown in Table 2. Fructose concentration decreased with length of standing time. Fructose concentration after standing for 2 hours was significantly lower than that after 0 hours, and much more significantly lower after 4 hours than after 0 or 2 hours. It was also found that the decrease of fructose level after standing semen was correlated with sperm concentration and motility. Thus, the slope rate (K) was calculated, representing the decrease of each seminal fructose using Excel to plot the value of fructose level at 0, 2, and 4 hours. Meanwhile, each of the motile sperm concentrations was calculated by motility × sperm concentration. Subsequently, Pearson's correlation analysis was performed between K and motile sperm concentration, showing that the decrease of seminal fructose concentration had significantly positive correlation with motile sperm concentration (r = .374, P = .009; Figure 2). . Correlation between the decrease of fructose concentrations in semen (K) and motile sperm concentration (r = .374). Fructose concentrations were all 0 for 10 samples of sperm suspension, indicating that fructose in sperm suspension was undetectable. Fructose concentrations in seminal plasma obtained by centrifugation at 1000 × g for 10 minutes or 3000 × g for 15 minutes were (15.96 ± 7.62) mmol/L and (16.34 ± 8.06) mmol/L, respectively, and not significantly different (t = −1.369, P = .176). After centrifugation at 1000 × g for 10 minutes, there were 92.64% (63/68) of seminal plasma containing sperm with an average concentration of 2.81 × 106/mL (range from 0 to 19.40 × 106/mL). Conversely, centrifugation at 3000 × g for 15 minutes resulted in 44.12% (30/68) of seminal plasma containing sperm with average concentration of 0.41 × 106/mL (range from 0 to 5.67 × 106/mL), which was a significant difference compared to centrifugation at 1000 × g for 10 minutes (t = 7.546, P < .001; Figure 3). . Comparison of sperm concentrations in seminal plasma obtained by centrifugation at 1000 × gfor 10 minutes or 3000 × gfor 15 minutes. Fructose concentrations in seminal plasma with or without chymotrypsin are shown in Table 3, which indicates that chymotrypsin had no apparent effect on seminal fructose determination. After 1% (v/v) of chymotrypsin (8000 U/mL) was added to 39 samples of semen, liquefaction was improved; seminal fructose concentration is shown in Table 4. While chymotrypsin may improve liquefaction of nonliquefied semen, there was no significant difference in fructose concentration between liquefied and nonliquefied samples. Two samples of semen were stored at −20°C (not frost-free) for 20 days, and the seminal fructose concentration was detected every other day. Results are shown in Figure 4. Fructose concentration in sample 1 was (25.96 ± 1.33) mmol/L with 5.05% coefficient of variation (CV), and (15.68 ± 1.06) mmol/L in sample 2 with 6.61% CV, suggesting that freezing-thawing had no apparent effect on seminal fructose concentration. . Fructose concentrations of samples 1 and 2 during 20 days. Ten samples of seminal fructose concentrations detected by 2 technicians are shown in Figure 5. Seminal fructose concentrations detected by 2 technicians were (16.12 ± 5.78) mmol/L and (16.01 ± 5.78) mmol/L, respectively, and there was no significant difference between their results (P = .643). . Bland and Altman plot of seminal fructose concentrations determined by 2 technicians. SD = 0.31. The range of 95% CI is from −0.62 to 0.62. The normal function of seminal vesicle is essential for sustaining fertility. When the function of seminal vesicle decreases, semen coagulation, sperm motility, stability of sperm chromatin, and semen immunoprotection are affected (Montagnon et al, 1990). One of the most important markers for the seminal vesicular function is the concentration of fructose in seminal plasma. Fructose is a main carbohydrate source in seminal plasma and necessary for sperm motion (Buckett and Lewis-Jones, 2002; Santiani et al, 2005). The measurement of seminal fructose has been used in most laboratories. Therefore, the World Health Organization manual recommends measurement of seminal fructose as a marker of seminal vesicular function (World Health Organization, 1999). Methods for determination of seminal fructose mainly include gas chromatography, indole coloration, and resorcinol coloration. In particular, the resorcinol method has been used widely in clinical andrology laboratories for its simplicity of operation, high specificity, and no need for special instrument. In the present study, 2 technicians were responsible for the determination of 10 samples of seminal fructose, showing no significant difference between their results (P = .643). This supports the notion that the resorcinol method used in determination of seminal fructose is now feasible for general technicians working in andrology laboratories. In routine assays of semen, we found that standard fructose solution tended to be stable within 15 days after preparation. Thus, we monitored the OD value change of standard fructose solution. Results showed that there was a large variance of OD values within the first 2 weeks and that OD values remained relatively stable within the subsequent 2 weeks, falling quickly from 0.30 to 0.19 within the last 10 days. Fructose, a polyhydroxy ketose, has dissymmetry carbon atoms with optical activity, which is mutarotation in water (Tan et al, 2005). About 2 weeks later, 5 possible isomers, including types α (levo- and dextro-rotation), β (levo- and dextrorotation), and straight chain, achieved a balance, in accordance with specific rotation achieving a balance with relative stability. Therefore, OD values of standard fructose solution were relatively stable, with a slight variance 2 weeks after preparation. About 4 weeks later, with the weak nucleophilic reagent of water, ketose was transformed into aldose and also achieved a balance within a period of time. As resorcinol and concentrated hydrochloric acid react with ketose to form more red compound than with aldose, OD value would decrease rapidly within the last 10 days. For these reasons, in order to gain accurate and reliable seminal fructose concentration, standard fructose solution should be stored for 2 weeks before use and discarded when the decrease of OD value appears. Our data showed that there was no obvious difference of fructose concentrations in seminal plasma standing for 0, 2, 4, or 6 hours. Next the change of seminal fructose within 4 hours of standing of semen was observed, showing that seminal fructose concentration decreased with length of standing duration. The concentration of fructose at 2 hours was significantly lower than that measured immediately, and at 4 hours significantly lower than at 2 hours. Furthermore, the decrease of fructose concentration was significantly positive correlated with motile sperm concentration (r = .374, P = .009), verifying the demonstrations that motile sperm in vitro can unceasingly consume fructose, and evaluation of seminal vesicular function should be performed within 3 hours after ejaculation (Rui et al, 1986). We suggest that standing time of semen should be standardized and that semen should be centrifuged immediately after liquefaction and separated into seminal plasma and sperm, to ensure accuracy of seminal fructose determination. Fructose concentrations in seminal plasma obtained by centrifugation at 1000 × g for 10 minutes or 3000 × g for 15 minutes were also observed, showing that there was no significant difference of fructose concentrations (t = −1.369, P = .176) but there was significant difference of sperm concentrations (t = 7.546, P = .000) in seminal plasma obtained by centrifugation at different velocity. When centrifugation velocity increased, seminal fructose concentration tended to increase, possibly due to apparently decreased sperm concentration in seminal plasma. Our investigation of sperm suspension showed that sperm did not contain fructose. So rudimentary sperm in seminal plasma obtained at low-velocity centrifugation would occupy some volume, leading to the decrease of actual volume of seminal plasma and then a decrease of fructose concentration. This indicates that though centrifugation velocity had no statistically significant effect on the determination of seminal fructose concentration, centrifugation velocity should not be below 3000 × g for 15 minutes to gain "pure" seminal plasma and comparable results of seminal fructose detection among different laboratories. Nonliquefied semen samples are often found in andrology laboratories. To improve the liquefaction of semen samples, chymotrypsin, extranase, or trypase are often added to semen samples (World Health Organization, 1999). Among them, chymotrypsin is inexpensive and used most widely. Therefore, we added chymotrypsin to liquefied semen samples and compared fructose concentrations in seminal plasma with or without chymotrypsin, showing that chymotrypsin had no significant influence on the determination of seminal fructose (P = .389). In addition, liquefaction of 39 nonliquefied semen samples was improved apparently after adding chymotrypsin, without any viscin thread when sampling. Meanwhile, there was no significant difference of seminal fructose concentration between liquefied and nonliquefied semen samples (P = .304). Findings indicated that nonliquefied semen samples with addition of 1% (v/v) chymotrypsin (8000 U/mL) in advance and incubated at 37°C for 30 minutes could be involved in detection of fructose concentration. Refrigeration of seminal plasma is necessary for establishing quality control of seminal fructose determination. Therefore, we investigated fructose concentrations in seminal plasma stored at −20°C for 20 days successively, with 5.05% and 6.61% of CVs for 2 samples of seminal plasma respectively, indicating that freezing-thawing had no influence on seminal fructose concentration, and that freezing-thawing seminal plasma could serve as a quality control product for determination of seminal fructose concentrations. Internal quality control (IQC) and external quality control (EQC) has been routine in laboratories performing clinical chemistry or endocrine analysis (Neuwinger et al, 1990). In recent years, many andrologists have been trying to introduce EQC into andrology laboratories (Auger et al, 2000). In order to establish IQC and EQC programs successfully in andrology laboratories, the prerequisite is standardization of technical procedures, which requires objective methods of semen parameters determination. Now such methods are being established and evaluated. With the help of strict IQC and EQC programs, results from different laboratories will be more comparable in the future. In conclusion, the resorcinol method to determine seminal fructose concentration can now be accepted, and determination of seminal fructose concentration should be standardized. First, standard fructose solution should be kept for 2 weeks at 4°C before use, and the valid period for use is about 2 weeks. Next, semen samples should be centrifuged and separated into sperm and seminal plasma immediately after liquefaction; otherwise, motile sperm will consume fructose. Thus, the standing time of semen should also be standardized. For nonliquefied semen samples, 1% (v/v) of chymotrypsin (8000 U/mL) can be added in advance and incubated at 37°C for 30 minutes to aid in liquefaction, and then for the detection of fructose concentration. Although centrifugation velocity had little influence on the determination of seminal fructose concentration, remnant sperm or other noncell components in seminal plasma may mildly affect the fructose level. Lastly, freezing-thawing had no apparent effect on seminal fructose concentration, indicating that frozen seminal plasma could serve as a quality control product for the determination of seminal fructose concentration among different laboratories.