Abstract
Dear Editor:
According to the study of Otasevic et al. (7), incubation of sperm with M40403 is able to rescue sperm from the negative effects of a 3-h in vitro incubation in the Tyrode's medium. The parameters evaluated included sperm motility, mitochondrial functionality (both of which intimately related to sperm function), as well as, and more crucially, the mRNA levels of various mitochondrial proteins and antioxidant enzymes.
One of the characteristics one has to take into account when studying sperm is that sperm is a terminally differentiated cell that cannot replicate/be maintained in a cell culture. For this reason, most of sperm studies are performed a few hours after sample collection, as incubating sperm in the culture medium for long periods will be detrimental. Interestingly though, we have recently shown that good quality sperm may be maintained alive and motile for various days (1), provided that optimal conditions are used. The more abrupt reduction in sperm motility observed in the Otasevic et al. (7) study can be explained by the fact that they did not include serum albumin in their culture medium (as they wanted to use noncapacitating conditions), and thus sperm may have started to aggregate to each other during the 3-h period. Regardless, cellular modifications might also have occurred, including the production of intracellular ROS, which might have affected sperm quality during this period of time. M40403, by scavenging superoxide anion, was able to rescue sperm motility and mitochondrial functionality, most probably by avoiding lipid peroxidation, similarly to what happens when sperm is incubated with SOD (5). We have no major qualms with these data.
However, the authors of the article reasoned that M40403 would stimulate an increase in the transcript levels of different mitochondrial proteins (both nuclear- and mitochondrial-encoded) and antioxidant enzymes. This aspect deserves further attention. Such an explanation would be correct if sperm were similar to somatic cells. However, sperm are transcriptionally silent cells (at least for nuclear-encoded proteins), due to the high condensation architecture of their chromatin, a fact that anyone knowledgeable in the field is aware of. In fact, during spermiogenesis (the last and haploid phase of spermatogenesis), histones are gradually replaced by small, highly basic proteins called protamines, resulting in nuclear compaction [for a review see (6)]. Although ejaculated sperm do possess RNAs with possible clinical relevance [for a recent review see (3)], the general consensus in the field is that ejaculated sperm have no nuclear transcription activity, and that these RNAs are leftovers from the process of spermatogenesis, often residing in cytoplasmic remnants that are not immediately visible by conventional techniques. Thus, the altered transcript levels observed by Otasevic and coworkers (7) require reinterpretation. Although the authors correctly control for possible somatic cell contamination, the decreased nuclear-encoded transcript levels observed in untreated sperm maintained 3 h in the Tyrode's medium could have resulted from RNA degradation. Such degradation seems to be prevented when the SOD mimic is incorporated in the culture medium. As for the increased levels of transcripts observed in the SOD mimic situation when compared to the control, the only plausible explanation would rely on the fact that sperm samples, notably human clinical samples, are very heterogeneous. Thus, the RNA levels of the sperm cells coexisting in a same sample may be quite distinct, and one cannot obviously use exactly the same cells to compare different conditions. Unfortunately, the authors have not specified the exact method of quantification by real-time PCR, and did not mention the efficiency or the absence of primer–dimers in each reaction, factors known to affect quantitative PCR analysis (4), so it is hard to know how exact the quantification was. Furthermore, in this type of experiment using just one control, RNA for normalization would be less than adequate for other cell types, and is especially inadequate for sperm, given the variability in the system and considering the residual (and not active) nature of the RNA pool in these cells. To fully prove their points, the authors should have complemented the experiments by proving active transcription, monitoring RNAs unrelated to redox status/oxidative stress that, according to the model, would not be expected to change after treatments (as negative controls), and shown concomitant changes in protein levels. To be clear, these results would run contrary to all that is known regarding sperm transcription, and the stringency of proof is therefore high in this case.
That the authors completely ignore these issues, thus sidestepping the controversial nature of their work, and proceed to merely discuss the data as if considering just another type of specialized (somatic) cell is, in our opinion, disingenuous.
In conclusion, experimental setups always must consider the particular nature of each cell type. Sperm are distinct from their somatic cell counterparts: they are terminally differentiated haploid cells, devoid of most of the cytoplasm, whose specialized function is to achieve fertilization, and whose chromatin is transcriptionally silent. These characteristics must be taken into account when studying sperm, and not all experimental approaches usually performed in other cell types can be directly applied with the same rationale to the male gamete.
References
Abbreviations Used
reactive oxygen species
superoxide dismutase
Author Response
Bato Korac
Address correspondence to:
Prof. Bato Korac
University of Belgrade
Institute for Biological Research
Bulevar despota Stefana 142
Belgrade 11060
Serbia
E-mail:
Date of first submission to ARS Central, May 23, 2012
date of final revised submission, June 15, 2012
date of acceptance, June 15, 2012
In their Letter to the Editor, the authors further discussed the effect of M40403 on sperm viability and mitochondrial functionality and speculated that it is based on a reduction in oxidative stress and the prevention of lipid peroxide formation. It is correct that M40403 can be useful as an antioxidant supplement in conditions of increased superoxide production. However, in the study by Otasevic et al., this was not the case. The sperm samples used in the experiment were normospermic, free of leukocytes, and incubated in standard, strictly controlled conditions specified by WHO (identical to the conditions used for sperm preparation during in vitro fertilization [IVF]), where the possibility of an increase in ROS was minimal. These are physiological conditions, and there is no place for oxidative stress. Superoxide dismutase (SOD) Mn(II) penta-azamacrocyclic mimics, a novel class of complexes, achieved marked protective effects in inflammation, stroke, atherosclerosis, and hypertension, not only through the removal of superoxide but also through additional redox mechanisms (41, 50). An incubation time of 3 h is commonly used in the preparation of sperm for IVF (8 –11).
Considering that study by Otasevic et al. deals with redox signaling, noncapacitating conditions were used with the purpose of avoiding a marked increase in reactive species production, which is known to occur during sperm capacitation. Numerous studies have examined redox regulation in spermatozoa using noncapacitation conditions (8 –11, 31, 48).
In the second part of the Letter to the Editor, the authors focused on the interpretation of results by Otasevic et al. concerning changes in the transcript levels of different mitochondrial proteins, both nuclear- and mitochondrial-encoded, and antioxidant enzymes. First, the authors speculate in their Letter that the observed decreases in various mRNA levels after 3-h incubation in the Tyrode's medium could be a consequence of RNA degradation, and that such degradation could be prevented when an SOD mimic is added to the medium. This interpretation of the results from study is, in our opinion, simply incorrect and does not stand up to scrutiny. If true, that would be a general phenomenon. However, transcript levels of cytochrome b, ATP synthase, CuZn-SOD, NOX, eNOS, and catalase were not decreased after 3 h of incubation in the Tyrode's medium, whereas transcript levels of GSH-Px, CuZn-SOD, NOX, cytochrome b, and ATP synthase were not increased after incubation in the Tyrode's medium supplemented with M40403. Furthermore, some of the genes examined in the study were not affected by the treatments applied (cytochrome c, ATP synthase, CuZn-SOD, and NOX) and furthermore, could serve as negative controls. The explanation of altered gene expression due to heterogeneity of human sperm samples does not stand up either. Such an interpretation can easily lead to questions regarding all previous studies with human sperm samples, including studies by the authors of this Letter.
As a major criticism of the increase in the mRNA content observed in work, the authors claimed that mRNAs in mature ejaculated sperm cells are leftovers from the process of spermatogenesis, residing in cytoplasmic remnants, and that due to the high condensation of chromatin, caused by replacement of histones by protamines, sperm cells are transcriptionally silent cells (at least for nuclear genes). I disagree with both of these assertions. First, the dogma that sperm RNAs are located only in cytoplasmic remnants was rejected in the late 1980s by Pessot et al. (35), when the presence of RNA was described in rat and human sperm nuclei, and over the past 20 years, an increasing number of published data have reported the accumulation of various RNA populations, mRNA, antisense, and micro-RNA in the sperm nucleus (4, 6, 21, 25, 32, 34, 43, 53) and mRNA in mitochondria (16, 19, 29). In addition, the persistence of low, but detectable, levels of transcription and translation in mature sperm cells clearly shows that the potential for active production of transcripts in mature sperm exists, and that the variety of detected mRNA molecules does not represent a nonfunctional leftover of spermatogenesis (16, 28, 30). Furthermore, the latest research clearly shows that at least some of the mRNA transcripts in highly purified mature sperm are neither remnants strictly related to sperm development and maturation nor contaminants from nearby somatic cells contained in accessory glands (12). These data strongly suggest that the sperm transcript pool is not a result of passive transcript packaging (12). In this regard, it is well known that in mature human spermatozoa, gene transcription is active in mitochondria, and that mammalian sperm is able to synthesize both mitochondrial-encoded RNAs (2, 17, 23, 40) and proteins (1, 39, 49). It has also been shown that mammalian spermatozoa contain nuclear-encoded mRNAs (25, 32, 52). The dogma that sperm are translationally silent cells for nuclear-encoded proteins was dismissed in 2006 by the elegant work of Gur and Breitbart (16), who showed that translation of nuclear-encoded proteins in mitochondrial-type ribosomes was localized either inside or outside the mitochondria. The story of inactive nuclear transcription due to high chromatin condensation has never been that simple and has become more complex due to the recognition that not all of the histones are replaced by protamines, and that a mature sperm nucleus retains 15% chromatin domains with histones that are assembled with DNA in a typical nucleosomal organization (3, 7, 26, 36, 54), whereas protamines are associated with 85% of sperm nuclear DNA (7, 14, 26, 47). Thus, transition of the histone-to-protamine exchange process is incomplete in human spermatids, and dormant sperm chromatin is selectively organized. Gene density more closely corresponds with histone rather than protamine profiles (21), revealing persistence of somatic-like chromatin structure in the mature sperm nucleus. DNA regions linked to histones are potentiated for expression and may represent sites of active transcription in functionally normal sperm cells, due to the nonrandom presence of both transcription factors and RNA polymerase (7, 13, 17, 28, 37, 47). Recent research suggests that this residual nucleosomal compartment, a generally overlooked feature of the male gamete, is far more significant and important than previously thought (27). Apart from RNA polymerase, mature spermatozoa contain nuclease (24), topoisomerase (44), DNA polymerase, and reverse transcriptase (RT) (13, 18, 51). It was recently shown that when spermatozoa are exposed to exogenous RNA, their endogenous RT can retrotranscribe cDNA copies that can be transferred into eggs during IVF (15, 22, 46), making the story of the mature sperm nuclear transcription capacity even more complex. Furthermore, new evidence has appeared that indicates that an RT-dependent process is also triggered when spermatozoa are exposed to exogenous DNA (38). Using a DNA construct, the authors have found that RT-spliced DNA sequences are generated in sperm cells and transmitted to embryos in IVF assays. Therefore, it has been proven that efficient transcriptional machinery is present in mature spermatozoa, which can transcribe the DNA into complementary RNA using an RNA polymerase, correctly remove the intronic sequence, splice the primary RNA transcript, and finally reverse-transcribe it into stable cDNA copies (38). The capacity of sperm cells to integrate exogenous RNA or DNA molecules indicates that such machinery, normally dormant, may be activated under certain conditions (17). In summary, and contrary to the authors' (of the Letter) end point that the function of sperm cells is only to achieve fertilization, it is currently accepted that mammalian sperm do not only deliver the male genome to the egg but also deliver a centrosome and a soluble factor that activate the egg, pointing to the possibility that other components found in sperm may play a yet unrecognized role (20, 42, 45). There is mounting evidence that spermatozoal RNA is delivered into the oocyte and remains intact after fertilization, pointing to functional roles for spermatozoal mRNAs in the oocyte (5, 33). It has become evident that the spermatozoon carries epigenetic factors that include male-specific genomic imprinting related to DNA methylation, correct packaging of the chromatin with protamines, modifications of histones, and a large population of mRNAs and miRNAs (7, 27). Recent evidence of the localization of spermatozoal RNA at the periphery of the nucleus, in close association with spermatozoal histones and potentiated genes, reinforces the hypothesis that sperm RNA may be involved in chromatin packaging in late spermatids and the preservation of paternal genomic imprinting. The mature sperm nucleus, at least in humans and mice, retains a somatic-like chromatin structure, as well as an endogenous RT that can be activated under certain conditions, ensuring the transmission of new genetic information. In general, the contribution of spermatozoa to early embryogenesis is now gaining more attention, and these new data must be taken into account when studying sperm.
References
Abbreviations Used
in vitro fertilization
nitric oxide
reactive oxygen species
superoxide dismutase
Can Spermatozoa Respond to Changes in Their Redox Status with the Selective Activation of Gene Transcription?
Robert John Aitken
Address correspondence to:
Prof. Robert John Aitken
Priority Research Centre in Reproductive Science
Discipline of Biological Sciences
University of Newcastle
Callaghan
NSW 2308
Australia
E-mail:
Date of first submission to ARS Central, May 23, 2012
date of final revised submission, June 15, 2012
date of acceptance, June 15, 2012
Much more contentious is the assertion by Otasevic et al. (7) that the treatment of spermatozoa with SOD mimics elicits an adaptive response from these cells involving changes in gene transcription from both the mitochondrial and nuclear genomes. Spermatozoa contain a complex variety of RNA species, and there is even some evidence to support the de novo translation of pre-existing mRNAs on mitochondrial polysomes (4). However, a central dogma of sperm cell biology is that de novo gene transcription cannot occur in the mature gamete for the reason illustrated in Figure 1. In contrast to the open chromatin structure represented by interphase somatic cell nuclei (Fig. 1A), the packaging of DNA into the sperm nucleus approaches the physical limits of molecular compaction, generating an internal structure that is almost crystalline (Fig. 1B). Within such a densely packed nucleus, it is difficult to imagine how the selective activation of gene transcription could occur, let alone the subsequent editing and transport of the newly synthesized mRNA to another compartment of the cell, the midpiece, for translation by the sperm mitochondria. It is possible, as Amaral and Ramalho-Santos suggest (3), that the suppression of oxidative stress with SOD mimics might have prevented oxidative damage to existing mRNA species, rather than induce de novo gene transcription. However, such an explanation does not readily account for the differential preservation of some mRNA species and not others. Further studies are clearly required to determine the biological significance of the mRNA changes described by Otasevic et al. (7), with primary emphasis on the production of new protein. The claims are clearly controversial, but in science, progress is often made by challenging the status quo.
Differing chromatin structure in
References
Abbreviations Used
reactive oxygen species
superoxide dismutase