SUMMARYPossible hazardous health effects of radiofrequency electromagnetic radiations emitted from mobile phone on the reproductive system have raised public concern in recent years. This systemic review and meta-analysis was prepared following standard procedures of the Cochrane Collaboration and the Preferred Reporting Items for Systematic Reviews and Meta-Analyses statement and checklist. Relevant studies published up to May 2013 were identified from five major international and Chinese literature databases: Medline/PubMed, EMBASE, CNKI, the VIP database and the Cochrane Central Register of Controlled Trials in the Cochrane Library. Eighteen studies with 3947 men and 186 rats were included in the systemic review, of which 12 studies (four human studies, four in vitro studies and four animal studies) with 1533 men and 97 rats were used in the meta-analyses. Systemic review showed that results of most of the human studies and in vitro laboratory studies indicated mobile phone use or radiofrequency exposure had negative effects on the various semen parameters studied. However, meta-analysis indicated that mobile phone use had no adverse effects on semen parameters in human studies. In the in vitro studies, meta-analysis indicated that radiofrequency radiation had detrimental effect on sperm motility and viability in vitro [pooled mean difference (MDs) (95% CI): À4.11 (À8.08, À0.13), À3.82 (À7.00, À0.65) for sperm motility and viability respectively]. As for animal studies, radiofrequency exposure had harmful effects on sperm concentration and motility [pooled MDs (95% CI): À8.75 (À17.37, À0.12), À17.72 (À32.79, À2.65) for sperm concentration and motility respectively]. Evidence from current studies suggests potential harmful effects of mobile phone use on semen parameters. A further multicentred and standardized study is needed to assess the risk of mobile phone use on the reproductive system.

Proteinase activities in human seminal plasma were detected using gelatin-containing sodium dodecyl sulphate-polyacrylamide gel electrophoresis zymography. Three prominent bands of activity of Mr 60,000, 66,000 and 90,000 were observed as well as 9 other bands of less intensity (34,000-158,000). These proteinases were dependent upon calcium for optimal activity, did not hydrolyse casein, and were predominantly in the soluble portion of seminal plasma. Examination of seminal plasma of men with different sperm concentrations, split ejaculates, and prostatic secretions indicated that the prostate gland was a source of most of these activities. Proteinase activities of Mr 34,000, 37,000, 82,000 and 120,000 were expressed more frequently in seminal plasma from normozoospermic men than from seminal plasma of oligo- or azoospermic men, indicating that they may also arise from spermatozoa in the semen sample. The proteinases of Mr 60,000 and 66,000 were found in all seminal plasmas whereas there was variation in the expression of the other molecular forms of enzyme, even in the normozoospermic samples. There are multiple forms of gelatinolytic proteinase activities in human seminal plasma which appear to arise from multiple sources in the reproductive tract including the Cowper's/urethral glands, the prostate gland, seminal vesicle and/or spermatozoa. Their function(s) in semen remains to be established.