Vol. 51, No.1, January 1989
FERTILITY AND STERILITY
Printed in U.S.A.
Copyright <> 1989 The American Fertility Society
Effect of human semen on herpes-simplex virus-2
Jerome K. Sherman, Ph.D.* Paul N. Morgan, Ph.D.t University of Arkansas for Medical Sciences, and John L. McClellan Memorial Veterans Hospital, Little Rock, Arkansas
There is little information on the interaction between semen and viruses in man. Such knowledge could help clarify the nature of sexual transmission of viral diseases from man to potential mother and child during both normal and artificial insemination. It was reported in 19861 that herpes simplex virus-type 2 (HSV -2) survived in human semen without any loss in infectivity during cryobanking. This study also suggested that although the cryoprotective agent glycerol, freezing and thawing, as well as frozen storage, were without effect, seminal fluid appeared to reduce the infectivity of HSV -2. Inhibition of the virus in semen was proposed as a possible explanation for the failure to detect HSV in the semen of 447 men in other studies. 2 •3 Viral infectivity in the 1986 study 1 was measured in terms of plaque-forming units, a method that could show enough normal variability to question the data that indicated the inhibition by seminal fluid. The unresolved suggestion of an inhibitory effect of semen on HSV -2, therefore, prompted a more definitive study, one that employed alternate quantitative methods for verification and clarification of possible relationships between HSV -2 and components of human semen. The description and results of this study are presented here.
Received May 12, 1988; revised and accepted August 23, 1988. *Reprint requests: J. K. Sherman, Ph.D., Department of Anatomy, University of Arkansas for Medical Sciences, 4301 West Markham Street, Little Rock, Arkansas 72205-7199. 't Department of Microbiology and Immunology, John L. McClellan Memorial Veterans Hospital.
Sherman and Morgan
MATERIALS AND METHODS TCID 5 o Method
Specimens were assayed for viral infectivity by determining the highest dilution of the fluid medium that produced cytopathic effects (CPE) in 50% of A549 cells in cultures inoculated with the HSV -2 strain MS virus. The end point was the 50% tissue culture infectious dose (TCID50 ). 4 A microtiter culture method permitted the use of small volumes of cells and virus in a greater number of iden· tical culture replicates per treatment for more accurate quantitative analysis. Treatments
Three ejaculates were collected from each of five donors in the semen procurement program at the university cryobank. Each ejaculate was used in a separate experiment in which, once the semen was divided, one part was the whole semen with spermatozoa and the other part its seminal fluid. This permitted comparison between the behavior of the virus in semen with spermatozoa, and the virus in seminal fluid without spermatozoa, from the same ejaculate of the same individual. Seminal fluid was prepared from the semen by removal of the cells after microcentrifugation of the liquefied ejaculate for 10 minutes at 9000 rpm or 7000 gs. In each experiment whole semen, its seminal fluid, and the tissue culture inedium (control) were compared as to viral infectivity by the TCID 50 method, in identical treatment, after inoculation with the same quantity of virus, and incubation at 36oC for 0, 5, and 24 hours. Procedures
Stock virus HSV -2 (0.1 ml), which showed a TCID50 of 4.8 (range, 4.5 to 5.0) per 25 Jtl, was
Fertility and Sterility
thawed from frozen-storage and added to 0.9 ml EMEM medium for an initial 1:10 dilution; 0.1 ml of this dilution was then inoculated into 0.9 ml of the control medium (EMEM with 10% calfs serum), into 0.9 ml of whole semen, and into 0.9 ml of seminal fluid, for an infective dose of 100 to 1000 TCID 50 in each sample treatment to be tested. Samples were incubated at 36oC for up to 24 hours. Aliquots were removed for quantitation of virus at 0, 5, and 24 hours of incubation. Dilutions of 10-3 to 10-6 were then made for each sample. Twentyfive microliters from each dilution of each sample was then added to each of a series of six-microtiter wells, each containing a monolayer of A549 cells in 0.2 ml ofEMEM-3% calfs serum. The concentration of antibiotics in the culture medium for CPE measurements in all experiments was 50 JLg gentamycin and amphotericin B, and 100 JLg of vancomycin, per ml of viral dilution. The preparations were incubated at 36oC in a 5% C0 2 chamber for 5 to 7 days, at which time CPE observations (and ratings) were made and TCID 50 values calculated for each treatment. An additional experiment was performed with whole semen and with seminal fluid, with A549 cells but without virus, using the identical techniques in processing and testing of the same dilutions. This was done to observe if semen or seminal fluid itself caused CPE without the presence of virus as a possible confounding factor in evaluation. Electron Microscopy
Two-milliliter aliquots of three semen samples from two donors were inoculated with HSV-2 to make a 10- 2 dilution, which was incubated for 5 and 24 hours to parallel procedures used for TCID50 measurements. Samples were then fixed in 4% glutaraldehyde and centrifuged to form pellets, which were postfixed in 1% osmium tetroxide. Pellets were embedded in epon-araldite, sectioned, and processed for observations and photographic recording in an JEOL lOOCX II electron microscope (Tokyo, Japan). Hundreds of sections from cubes of the six semen-virus preparations for each incubation time were studied. Observations were made on thousands of sectioned spermatozoa and their milieu in evaluating ultrastructural evidence for viral attachment to, penetration of, and replication within spermatozoa. T.o increase chances for detection of virions, additional "overload" experiments were run. One
Vol. 51, No.1, January 1989
Table 1 Infectivity of HSV -2 During Incubation in Human Semen• Control
EMEM Hours of incubation 0 5 24
TCID 5o SEM 3.17 2.80 1.09
0.07 0.07 0.13
3.13 2.66 1.90
0.07 0.10 0.12
2.98 2.35 1.17
0.10 0.12 0.13
a Values based on CPE measurements in 15 experimental runs.
thousand times the original concentration of virus was added to aliquots of three semen samples from the same two donors used initially, following the same procedures described above for electron microscopy. Semen Analysis
A complete multifactorial semen analysis was performed for each semen sample at the time of collection, as well as for aliquots of three of the ejaculates after 5 and 24 hours incubation, in each experiment. The initial analysis was performed to ensure ,the use of semen with characteristics in a favorable profile. The purpose of the analysis after incubations was to detect possible deleterious effects of viral infection reflected in a cytolytic loss of spermatozoa. RESULTS Statistical Analysis
Data on CPE for each series of six culture wells were employed in computing the mean TCID 50 values for each treatment and time period in the three experiments performed on each of the five semen samples of both whole semen and seminal fluid, and the controls. The data were first analyzed using a three-factor repeated measures analysis of variance design. Interaction between treatment and time was found to be significant. Each time period was then analyzed separately by utilizing twofactor repeated measures techniques. No significant difference in treatment was found for the 0 hour time period, but both the 5 and 24 hour time periods revealed significant treatment differences. Duncan's multiple range test was used to further examine these differences. Table 1 summarizes the mean values of the
Sherman and Morgan Communications-in-brief
HSV -2 infectivity titer (TCID 50 ) study, which revealed the following upon statistical analysis: (1) at time 0 hour there was no significant difference between the three treatments (control, whole semen, and seminal fluid); (2) after 5 hours of incubation control and whole semen were not significantly different, but infectivity values for seminal fluid were significantly lower than both other treatments; and (3) after 24 hours there was no significant difference between control and seminal fluid, but viral infectivity of whole semen was significantly higher than both. There was no CPE noted in the experiment with semen and seminal fluid, but without virus. This demonstrated that CPE in measurements of viral infectivity was due to populations of the virus, and that the CPE values reflected the viral infectivity in valid TCID 50 measurements. Electron Microscopy
Ultrastructural observations revealed no virions attached to or within spermatozoa. In some specimens some virions were noted free in the extracellular compartment. These probably were carried down during centrifugation in the cell-laden fraction, within viscous elements of seminal fluid (Fig. 1). There was no evidence for a close ultrastructural relationship between virions and spermatozoa after incubation for both 5 and 24 hours, even with the overload concentration of the viral population.
Figure 1 Representative electron micrograph showing the extracellular location or'herpes-simplex-2 virion (arrow) relative to spermatozoon, during incubation of human semen inoculated with a population of virus. No intimate ultrastructural connection with spermatozoon is noted, in this case with its head, after 5 hours of incubation (original magnification, X55,000).
There was no reduction in number (millions/ml) of spermatozoa in those semen samples observed initially at the time of collection and subsequently after 5 and 24 hours of incubation with virus. DISCUSSION
There was a natural expected loss of viral infectivity with time in the control culture medium (EMEM) during incubation. During the first 5 hours, however, the loss was significantly greater in the seminal fluid, supporting the suggestion of inhibition by this component of semen. 1 Whole semen, however, behaved like the control medium during this period of incubation, questioning the suggestion of viral inhibition by whole semen. 1 Seminal fluid apparently ran its course of inhibition after 5 hours. It then behaved like the control
Sherman and Morgan
medium during a statistically similar loss in viral infectivity to 24 hours, whereas whole semen clearly enhanced maintenance with less loss than the control medium during this period. Such maintenance in whole semen suggests a relationship between spermatozoa and HSV -2. The expected relationship in a culture of cells with HSV -2 is viral penetration and its intracellular replication, usually during a few hours of incubation. Measurement of viral infectivity depends upon the freedom of the virus to enter test cells in culture for the induction of CPE. Entrance of virus into spermatozoa, therefore, may actually decrease the TCID 50 of inoculated whole semen because fewer virions would be available (free) for induction of CPE within test tissue culture cells. This could not have been the basis for the viral inhibition by seminal fluid during the first 5 hours of incubation, how-
Fertility and Sterility
ever, because no sperm cells were present. Therefore, it appears that a constituent of seminal fluid, such as spermine phosphate/ could have exerted a type of chemical inhibition. The more favorable nature of whole semen, as a maintenance medium during the 24 hours of incubation with HSV -2, however, introduced the presence of spermatozoa as the difference between both control medium and seminal fluid, which were inferior to whole semen in maintenance of viral infectivity and whole semen. An intimate favorable association or relationship between spermatozoa and HSV-2 was thus demonstrated. Extensive electron microscopic observations, however, revealed no evidence for a close ultrastructural spermatozoon-virion relationship. This is surprising because a most unique feature of HSV-2 is its ability to penetrate and replicate in almost any type of cell from many different animals. This nonspecificity is probably the basis for its responsibility for a wide spectrum of disease. It is possible, that, in spite of the extensive number of sections and observations, virions within and attached to the spermatozoa were not detected by electron microscopy. The relatively low numbers of virions compared with much greater numbers of spermatozoa even in the overload experiments and the size difference between the two in representative thin sectioning, make this possible. If we assume that intracellular virions and attached virions were missed in the electron microscopic preparations and that, indeed, a number of virions was actually within and attached to spermatozoa, then we would expect a reduced or inhibited level of viral infectivity with few free virions, rather than the observed enhanced maintenance of infectivity, in whole semen. It is more likely, however, that results with electron microscopy were correct and that HSV-2 does not enter a spermatozoon, a highly specialized germ cell with haploid not the somatic diploid number of chromosomes. The superiority whole semen in maintenance of HSV -2 infectivity could be viral stabilization by the available extracellular proteins. 5 However, semen and seminal fluid contain the same proteins. Results also reveall:ld no evidence for loss of spermatozoa in cytolysis to provide such a source of protein. It appears that the basis for inhibition of HSV -2 infectivity resides in a component of the seminal flui,d, at least during the first 5 hours of incubation
Vol. 51, No.1, January 1989
after inoculation of the viral titer. Creation of conditions that are favorable for improved maintenance of HSV -2 infectivity, however, is a function of spermatozoa in whole semen, but probably not because of attachment or intracellular replication. It is suggested, therefore, that both inhibition and enhancement could be based upon chemical, biochemical, or immunologic characteristics of the milieu in seminal fluid and of whole semen. Explanation for enhancement of maintenance of viral infectivity in whole semen, involving spermatozoa, however, remains difficult even in theory. SUMMARY
This study was designed to evaluate the suggestion of inhibition of infectivity of HSV -2 by seminal fluid 1 by examining relationships between HSV -2 and components of human semen. Viral infectivity was quantitated by TCID50 measurements of CPE in comparing viral inoculated aliquots of whole semen, seminal fluid, and control (EMEM) medium during incubation at 0, 5, and 24 hours. Statistical analysis of data from 15 experiments revealed inhibition by seminal fluid during 5 hours of incubation. After 24 hours viral infectivity was ·maintained significantly higher with whole semen than with other treatments, suggesting favorable spermatozoal interaction. Electron microscopy revealed no intimate ultrastructural relationship between the virion and the spermatozoon. Acknowledgements. The authors gratefully acknowledge the fine technical assistance of Absalom Tilley in viral studies, Joy Thompson in electron microscopy, and Carolyn Thompson in statistical analysis. REFERENCES 1. Sherman JK, Menna JH: Cryosurvival of Herpes simplex
virus-2 during cryopreservation of human spermatozoa. Cryobiology 23:383, 1986 McGowan MP, Hayes K, Kovacs GT, Leydon JA: Prevalence of cytomegalovirus and Herpes simplex virus in human semen. Int J Androl6:331, 1983 Tjiam KH, van Heijst BYM, Polak-Vogelzang A, Rothbarth PH, van Jooost T, StolzE, Michel MF: Sexually communicable microorganisms in human semen samples to be used for artificial insemination by donor. Genitourin Med 63:116, 1987 Landry ML, Hsiung GD: Primary isolation of viruses. In Clinical Virology Manual, Edited by S Specter, GJ Lancz. New York, Elsevier Press, 1986, p 31 Rapp F: Personal communication
Sherman and Morgan Communications-in-brief