Recent advances in the treatment of HIV infections have prompted more and more HIV-infected people to have children. If only the man is HIV-infected, the risk of contaminating the sexual partner and the child can be avoided by adoption, or by insemination with semen from a healthy HIV-seronegative donor . The other alternatives are unprotected sex and assisted medical procreation. Unprotected sex leads to an estimated 0.15% HIV transmission by vaginal receptive intercourse  and is thus considered to be too dangerous, even if the sexual intercourse occurs only during ovulation. Thus, recommending natural conception may lead to increased unprotected intercourse and to infection of the women . Assisted medical procreation requires the processing of semen to remove HIV, as insemination with semen containing HIV-1 leads to the contamination of women.
Despite the finding of HIV-1 virus particles in spermatozoa by electronic microscopy in an experimental context [4,5], recent studies indicate that they are not infected . Spermatozoa do not bear significant levels of CD4, CCR5 and have low levels of CXCR4 , all of which are required for HIV infection. HIV-1 particles are found in the seminal plasma and HIV-1 provirus has been detected in seminal non-spermatozoa cells (NSC) . Because spermatozoa are not infected with HIV-1, the selective isolation of motile spermatozoa should minimize transmission during artificial insemination . Medically assisted conception using spermatozoa tested for HIV-1 [8,9] should then be a much safer alternative to condom-free intercourse during ovulation .
As infection with both HIV and hepatitis C virus (HCV) is frequent and HCV can be transmitted during sexual intercourse or assisted medical procreation [10,11], it is necessary to know also what fraction of semen harbours HCV infectivity. The sexual transmission of HCV seems to be much less likely than that of HIV  as only a few cases have been reported . The presence of HCV in semen is still controversial. Few studies have been carried out on the subject and it is technically difficult to detect HCV RNA due to PCR inhibitors and the low numbers of virus particles present. Studies on HCV RNA in the semen of infected individuals have yielded conflicting results. Some have found no HCV RNA [14–17], whereas others have repeatedly found HCV RNA [10,11,18,19] or HCV-specific antigens . However, all of the studies tested only a small number of samples, the genome amplification techniques used were not standardized, and the semen fractions tested were heterogeneous. The semen fraction that could contain HCV infectivity is unknown, but isolation of motile spermatozoa may reduce the risk of HCV transmission .
This study was performed to assess the removal of HIV-1 and HCV virions from spermatozoa during preparation by ‘washing and swim-up'; HIV-1 and HCV genomes were sought in semen samples at several steps of preparation and in the corresponding blood samples. In addition, HIV-1 RNA concentrations in seminal plasma and blood plasma of treated patients from whom multiple samples were taken were measured.
Materials and methods
A group of 32 HIV-1 infected men donated semen and blood samples. They were all outpatients attending the Male Sterility Center and CECOS Midi-Pyrénées (Hôpital La Grave, Toulouse) (Table 1). Their mean age was 37 years and the mean duration of HIV-1 infection was 12 years. They were all clinically asymptomatic. Fifteen of them were former injecting drug users, three were contaminated by blood transfusion and 14 by homosexual or heterosexual intercourse. Twenty-seven patients were on antiretroviral therapy: six were receiving two nucleoside inhibitors and 21 were receiving three or more drugs. The semen and blood samples were collected during the same visit. Patients were tested for anti-human T cell lymphotrophic virus types 1 and 2, HCV, cytomegalovirus and syphilis antibodies.
Fifty-one semen samples were obtained by masturbation, after a recommended 3 day abstinence, and collected in sterile containers. The samples were processed in the laboratory within 2 h of ejaculation. Seminal plasma and semen cells were isolated from an aliquot by centrifugation at 11 000 g. Spermatozoa were prepared from whole semen cell pellets by differential density gradient (PureSperm, J.C.D. S.A., Lyon, France) centrifugation over 50, 70 and 90% of PureSperm. The pellet (90% PureSperm) was washed twice with BM1 medium (Elios Biomedia, Paris, France). The 50% fraction should contain mostly low motile, dead, or abnormal spermatozoa and seminal NSC. The motile fraction was separated further by the ‘swim-up’ method from the 90% fraction. Briefly, the spermatozoa pellet was overlaid with 1.1 ml medium and incubated at 37°C, 5% CO2 for 60 min in a tube inclined at 45 degrees to allow the motile spermatozoa to swim up. The ‘swim-up’ fraction should contain only viable motile spermatozoa. The whole semen pellet, 50% fraction cells, and the spermatozoa obtained after swim-up were collected and analysed for virus. All samples were stored as dry pellets at −80°C.
Nucleic acid detection in blood
Plasma HIV-1 RNA was quantified with the Amplicor HIV-1 Monitor v1.5 assay (Roche Diagnostic Systems, Meylan, France) using the ultrasensitive protocol (detection limit 20 copies/ml).
The COBAS Amplicor HCV assay (Roche Diagnostic Systems, Meylan, France) was used to detect HCV RNA in blood serum with a detection limit of 100 copies/ml. The HCV RNA in blood serum was quantified using the HCV Monitor assay (Roche Diagnostic Systems, Meylan, France) with a detection limit of 1000 copies/ml.
Nucleic acid detection in seminal plasma
The Amplicor HIV-1 Monitor v1.5 assay was used to quantify HIV-1 RNA in seminal plasma. The assay was modified as described below. The internal quality standard (IQS) was added to 100 μl seminal plasma before HIV-1 RNA extraction, as for blood plasma samples. RNA was then extracted from seminal plasma using Nuclisens NASBA Diagnostics extraction kit (Organon Teknika S.A., Fresnes, France) as recommended by the manufacturer. All extracted HIV-1 RNA was processed using the ultrasensitive Monitor assay protocol. Controls were included in each run. The detection limit of the assay in seminal plasma was 100 copies/ml. Specificity of the assay was 100% on an HIV-1-negative seminal plasma panel of 20 samples.
A modified Amplicor assay was used to detect HCV RNA in the seminal plasma and cells of patients in whose blood HCV RNA (COBAS Amplicor HCV assay) was detected. The internal control (IC) was added to 100 μl seminal plasma before HIV-1 RNA extraction, as for serum samples. RNA was extracted from seminal plasma using the Nuclisens NASBA Diagnostics extraction kit as recommended by the manufacturer. All extracted HCV RNA was processed using the standard Monitor assay protocol (detection limit of 100 copies/ml). Controls were included in each run. Assay specificity was 100% on an HCV-negative seminal plasma panel of 20 samples. The detection limit of the assay in seminal plasma was 100 copies/ml.
Nucleic acid detection in seminal cells
A modified HIV-1 Monitor v1.5 assay (Roche Diagnostic Systems, Meylan, France) was used to detect HIV-1 RNA or both HIV RNA and DNA. Frozen pellets of 2 × 106 cells were suspended in 600 μl Monitor lysis buffer containing IQS (Amplicor HIV-1 Monitor v1.5 assay). Cells were lysed by thermal shock (three rounds of 15 sec in liquid nitrogen and 30 sec at 60°C) and a final 2 h incubation at room temperature. Samples for RNA assay were incubated with DNase (Appligene, Illkirch, France) at 37°C for 60 min. Both DNA and RNA were detected in untreated samples. Extraction was performed as recommended by the manufacturer for plasma RNA. Fifty μl of the 100 μl suspended RNA were amplified and reverse transcription–PCR products were detected using the Monitor detection kit. Assay specificity was 100% on an HIV-1-negative seminal cell panel of 60 samples. The assay detection limit in seminal cells was 20 HIV-1 genome copies/106 cells.
A modified HCV Amplicor assay was used to detect HCV RNA seminal cells. As for serum, the IC was added to a 2 × 106 cell pellet before RNA extraction. Cells were lysed by thermal shock (three rounds of 15 sec in liquid nitrogen and 30 sec at 60°C) and incubation for 1 h at 60°C. Proteinase K was inactivated by incubation at 95°C for 10 min and RNA was extracted with phenol and precipitated with ethanol. Fifty μl extracted RNA and IQS were amplified and reverse transcription–PCR products were detected using COBAS Amplicor HCV system. Assay specificity was 100% on an HIV-1-negative seminal cells panel of 60 samples. The assay detection limit in seminal cells was 100 copies/106 cells.
The χ2 test or Fisher's exact test was used to compare distribution ratios. The Mann–Whitney U test was used to compare RNA levels in semen and blood plasma to treatment, CD4 cell count, age and HIV–HCV coinfection. Results below the assay detection limits for viral RNA were assigned the value of the detection limit. The correlation between variables was examined using Spearman's rank correlation test.
HIV-1 RNA in seminal plasma
The mean HIV-1 RNA in seminal plasma level was 553 copies/ml with a median of 88 copies/ml (Table 1). HIV-1 RNA was detected in 14 samples, not detected in 32 samples, not tested in three samples (owing to insufficient volume) and inhibitors were present in two samples.
The mean HIV-1 RNA in blood plasma was 6458 copies/ml, with a median of 414 copies/ml. HIV-1 RNA was undetectable in 10 samples using the Monitor ultrasensitive assay (detection limit 20 copies/ml), 12 samples had detectable RNA (< 200 copies/ml), and 12 samples had RNA concentrations > 3000 copies/ml.
The seminal plasma of almost all patients whose blood plasma contained no detectable HIV-1 RNA (n = 8) contained no detectable RNA (seven samples), with detectable RNA in one sample. The seminal plasma of patients with detectable HIV-1 RNA in their blood plasma (n = 38) contained no detectable RNA (25 samples) or had detectable RNA (13 samples). The HIV-1 RNA concentrations in seminal and blood plasma (r = 0.258, P = 0.37), as well as qualitative results (P = 0.29), were not significantly correlated. Patient 15 had a higher virus load in his seminal plasma than in his blood plasma.
Antiretrovirus treament did not influence significantly the detection of HIV-1 RNA in seminal plasma (P = 0.17) (Table 2). The five samples from patients receiving no antiretroviral treatment all had detectable virus in the blood plasma and two of them had HIV-1 in their seminal plasma (Table 1). The 21 patients receiving three- or four-drug antiretroviral therapy included 18 patients with at least one blood plasma sample with detectable virus; HIV-1 was detected in the seminal plasma of six of them at least once.
The 11 patients with a CD4 cell count < 350 × 106/1 included 11 with virus in their blood plasma, four of whom had HIV-1 in their seminal plasma. The CD4 cell count and seminal plasma HIV-1 RNA concentration were not significantly correlated (P = 0.25) (Table 2).
Two or three samples were obtained from 16 patients. All of the concomitant blood and semen samples from 14 patients were tested. The 11 patients who had persistent detectable HIV-1 RNA in their blood plasma included two in whom HIV-1 RNA was detected repeatedly in the seminal plasma, six in whom HIV-1 RNA was repeatedly undetectable in the seminal plasma, and three whose results for seminal plasma varied. All of these three patients (3/14, 21%) had detectable virus in their blood plasma. Two patients had RNA concentrations that varied across the 20 copies cut off, so that virus was repeatedly undetectable in seminal plasma. Only one patient had consistently (two samples) undetectable HIV-1 RNA in his blood and seminal plasma.
HIV-1 RNA/DNA in the three seminal cell fractions
HIV-1 RNA and/or provirus DNA were detected in 16 out of 51 samples: nine out of 50 (24%) of the whole semen cell pellets, seven out of 45 (15.5%) of the 50% cell fraction and none of the 40 of the spermatozoa fractions.
At least one cell fraction from 12 cases had detectable HIV-1 genome and both the whole semen cell pellet and the 50% cell fraction were positive for HIV-1 genome in four cases. HIV-1 genome was detected only in the whole semen cell pellet in five cases, HIV-1 genome was detected only in the 50% cell fraction in three cases. Four samples of seminal plasma were negative for HIV-1 RNA, and at least one cell fraction was positive. A significant correlation was found between HIV-1 RNA detection in the seminal plasma or in blood plasma and HIV-1 RNA detection in the cellular fractions (P > 0.05;Table 2).
No HIV-1 genome (DNA or RNA) was detected in the motile spermatozoa, regardless of the RNA found in the blood or seminal plasma, or genome detected in upstream fractions.
HCV RNA in sperm fractions
Twenty (62.5%) patients had anti-HCV antibodies in their sera and 16 of them had detectable HCV RNA (detection limit 100 copies HCV RNA) in their sera at the time of sample collection. The mean HCV virus load in serum was 315 000 copies/ml, with a median of 220 000 copies/ml. Four out of 20 (20%) seminal plasma samples were positive for HCV RNA using the Amplicor assay (Table 3). HCV RNA-positive samples were systematically retested for confirmation and were found to be positive. Six individuals gave semen more than once, one was not tested. HCV genome was detected in only one of two samples taken from two individuals. Semen cell fractions were tested for HCV RNA but this was not detected in any of the cell fractions. Detection of HCV RNA in seminal fluid was not correlated with immunodepression (CD4 cell count < 350 × 106/l;P = 0.5) or detection of HIV RNA in seminal plasma (P = 0.26) or blood (P = 0.47).
We have determined the distribution of HIV-1 and HCV in the fractions of semen obtained during the isolation of motile spermatozoa. The HIV-1 genome was detected using an adapted standard assay. A modified HIV-1 Monitor assay was used to amplify virus RNA and DNA in cell lysates. Seminal plasma is known to contain PCR inhibitors of unknown nature that can lead to false-negative results . A silica-based RNA extraction method washes out inhibitors. This assay uses an internal standard that is extracted and amplified along with virus RNA and DNA to monitor extraction and amplification, and so check for possible remaining inhibitors. Proviral DNA was detected only in cell fractions containing NSC. HIV-1 is present in the seminal plasma and NSC [7,21–23]. Recent reports indicated that polymorphonuclear cells may be implicated [23–25]. The motile spermatozoa in the 90% fraction plus ‘swim-up’ were always free of HIV-1 RNA or DNA, in agreement with results obtained for 11 patients using a similar isolation process . We believe that ‘swim-up’ is a security factor because its absence may explain, at least partially, most of the HIV-positive spermatozoa samples reported [6,21,22]. Tachet used a two-layer gradient without swim-up and detected one positive sample (1/16) for HIV-1 DNA and six out of 41 for HIV-1 RNA . These detections of HIV-1 genome could be due to contamination of spermatozoa fraction by NSC, and the fact that the six patients had a seminal plasma virus load higher than or similar to that of the blood plasma. HIV-1 DNA was detected in the spermatozoa fraction prepared by density gradient and swim-up purification in six out of 107 (5.6%) samples by Marina . As suggested by the authors, these results could have been false-positives . These virological results are concordant with the clinical experience of insemination with processed semen from HIV-infected men [9,26]. No case of HIV-1 contamination has been reported.
Only one sample from the three patients from whom at least two semen samples were tested for HIV-1 RNA was positive. The sensitivity of the assay may be involved as well as the intermittent presence of HIV-1 in ejaculates [27,28]. Sexually transmitted diseases or cytomegalovirus infection [27,29] are one of the factors associated with HIV shedding.
The absence of HIV-1 RNA from the blood is not systematically correlated with its absence from seminal plasma (one case out of eight samples in our study), in agreement with Zhang . Similarly, an absence of HIV-1 from seminal plasma is not always correlated with no virus in seminal cells. Most men were on effective antiretroviral treatment and had CD4 cell counts > 200 × 106/l. This may explain the lack of a significant correlation between HIV-1 RNA detection in blood and seminal plasma. The number of spermatozoa left for insemination must be taken into account when testing the cell fraction(s) during spermatozoa purification. The insemination fraction was always negative for HIV-1 in our study, and is the smallest fraction. The 50% cell fraction is discarded during washing and is enriched in NSC and abnormal spermatozoa. This enrichment may explain some positive results in the 50% fraction, whereas the whole seminal cell pellet was negative. The whole seminal cell pellet is the fraction containing spermatozoa to be purified; testing it decreases the yield of spermatozoa available for insemination.
HCV genome was found in the seminal plasma of some samples from patients infected with HIV and HCV. This is the first report of HCV detection with a coamplified IC. The prevalence of HCV RNA in the seminal plasma (4/20) was in the range of those reported by others: one out of 11 , four our of 17 , one of one  and two out of 39 . Previous negative results may have been due to a lack of sensitivity, PCR inhibitors, a low HCV count or the small number of samples tested [14–17]. The HIV–HCV co-infection and the HCV viraemia may also increase the HCV in the semen . The absence of HCV genome from the cell fraction of semen indicates that the HCV particles are extracellular. As the spermatozoa were always free of HCV RNA, isolating motile spermatozoa could also reduce the risk of HCV contamination during artificial insemination with semen from an HCV-infected man. The few patients having HCV RNA in their semen points to a low concentration of HCV and may explain the low risk of sexual transmission. HCV shedding into the semen may also be intermittent, as HCV RNA was not always detected in sequential samples from the same patient.
Thus, the use of density gradient plus ‘swim-up’ to purify motile spermatozoa reduces the HIV-1 and HCV genomes in the spermatozoa of doubly-infected men to undetectable concentrations. This method associated with standardized virus assay could be useful for serodiscordant couples wishing to have children.
The authors thank F. Pontonnier and R. Mieusset for their unvaluable support, F. Cendres, A. Lamartre, D. Reboux, M. F. Durin and C. Fourtané for technical assistance, and O. Parkes for linguistic advice.
1. Anderson D. Assisted reproduction for couples infected with human immunodeficiency virus type 1.
Fertil Steril 1999, 72: 592 –594.
2. Downs AM, De Vincenzi I. Probability of heterosexual transmission of HIV: relationship to the number of unprotected sexual contacts.
:European Study Group in Heterosexual Transmission of HIV.
J Acquir Immune Defic Syndr Hum Retrovirol 1996, 11: 388 –395.
3. Mandelbrot L, Heard I, Henrion GE, Henrion R. Natural conception in HIV-negative women with HIV-infected partners.
Lancet 1997, 349: 850 –851.
4. Dussaix E, Guetard D, Dauguet C. et al. Spermatozoa as potential carriers of HIV.
Res Virol 1993, 144: 487 –495.
5. Baccetti B, Benedetto A, Burrini AG. et al. HIV particles detected in spermatozoa of patients with AIDS.
J Submicrosc Cytol Pathol 1991, 23: 339 –345.
6. Quayle AJ, Xu C, Tucker L, Anderson DJ. The case against an association between HIV-1 and sperm: molecular evidence.
J Reprod Immunol 1998, 41: 127 –136.
7. Kim LU, Johnson MR, Barton S. et al. Evaluation of sperm washing as a potential method of reducing HIV transmission in HIV-discordant couples wishing to have children.
AIDS 1999, 13: 645 –651.
8. Semprini AE, Levi SP, Bozzo M. et al. Insemination of HIV-negative women with processed semen of HIV-positive partners.
Lancet 1992, 340: 1317 –1319.
9. Marina S, Marina F, Alcolea R. et al. Human immunodeficiency virus type 1-serodiscordant couples can bear healthy children after undergoing intrauterine insemination.
Fertil Steril 1998, 70: 35 –39.
10. McKee TA, Avery S, Majid A, Brinsden PR. Risks for transmission of hepatitis C virus during artificial insemination.
Fertil Steril 1996, 66: 161 –163.
11. Levy R, Tardy JC, Bourlet T. et al. Transmission risk of hepatitis C virus in assisted reproductive techniques.
Hum Reprod 2000, 15: 810 –816.
12. Wyld R, Robertson JR, Brettle RP, Mellor J, Prescott L, Simmonds P. Absence of hepatitis C virus transmission but frequent transmission of HIV-1 from sexual contact with doubly-infected individuals.
J Infect 1997, 35: 163 –166.
13. Capelli C, Prati D, Bosoni P. et al. Sexual transmission of hepatitis C virus to a repeat blood donor.
Transfusion 1997, 37: 436 –440.
14. Semprini AE, Persico T, Thiers V. et al. Absence of hepatitis C virus and detection of hepatitis G virus/GB virus C RNA sequences in the semen of infected men.
J Infect Dis 1998, 177: 848 –854.
15. Hsu HH, Wright TL, Luba D. et al. Failure to detect hepatitis C virus genome in human secretions with the polymerase chain reaction.
Hepatology 1991, 14: 763 –767.
16. Fried MW, Shindo M, Fong TL, Fox PC, Hoofnagle JH, Di Bisceglie AM. Absence of hepatitis C viral RNA from saliva and semen of patients with chronic hepatitis C.
Gastroenterology 1992, 102 (4Pt1): 1306 –1308.
17. Terada S, Kawanishi K, Katayama K. Minimal hepatitis C infectivity in semen [letter].
Ann Intern Med 1992, 117: 171 –172.
18. Fiore RJ, Potenza D, Monno L. et al. Detection of HCV RNA in serum and seminal fluid from HIV-1 co-infected intravenous drug addicts.
J Med Virol 1995, 46: 364 –367.
19. Liou TC, Chang TT, Young KC, Lin XZ, Lin CY, Wu HL. Detection of HCV RNA in saliva, urine, seminal fluid, and ascites.
J Med Virol 1992, 37: 197 –202.
20. Kotwal GJ, Rutsgi VK, Baroudy BM. Detection of hepatitis C virus specific antigens in semen fromnon-A non-B hepatitis patients.
Digest Dis Science 1992, 37: 641 –644.
21. Tachet A, Dulioust E, Salmon D. et al. Detection and quantification of HIV-1 in semen: identification of a subpopulation of men at high potential risk of viral sexual transmission.
AIDS 1999, 13: 823 –831.
22. Dulioust E, Tachet A, De Almeida M. et al. Detection of HIV-1 in seminal plasma and seminal cells of HIV-1 seropositive men.
J Reprod Immunol 1998, 41: 27 –40.
23. Quayle AJ, Xu C, Mayer KH, Anderson DJ. T lymphocytes and macrophages, but not motile spermatozoa, are a significant source of human immunodeficiency virus in semen.
J Infect Dis 1997, 176: 960 –968.
24. Speck CE, Coombs RW, Koutsky LA. et al. Risk factors for HIV-1 shedding in semen.
Am J Epidemiol 1999, 150: 622 –631.
25. Ball JK, Curran R, Irving WL, Dearden AA. HIV-1 in semen: determination of proviral and viral titres compared to blood, and quantification of semen leukocyte populations.
J Med Virol 1999, 59: 356 –363.
26. Semprini AE, Fiore S, Oneta M, et al. Assisted reproduction in HIV-discordant couples. Human Reprod
27. Krieger JN, Coombs RW, Collier AC. et al. Intermittent shedding of human immunodeficiency virus in semen: implications for sexual transmission.
J Urol 1995, 154: 1035 –1040.
28. Bréchard N, Galéa P, Silvy F, Amram M, Chermann JC. Study of HIV localization in sperm.
Contracept Fertil Sex 1997, 25: 389 –391.
29. Fiscus SA, Vernazza PL, Gilliam B, Dyer J, Eron JJ, Cohen MS. Factors associated with changes in HIV shedding in semen.
Aids Res Hum Retroviruses 1998, Suppl. 1: S27 –S31.
30. Zhang H, Dornadula G, Beumont M. et al. Human immunodeficiency virus type 1 in the semen of men receiving highly active antiretroviral therapy.
N Engl J Med 1998, 339: 1803 –1809.