If there is a scientific analog of having one’s cake and eating it too, it would have to be iPS cells.
Shinya Yamanaka first described murine inducible pluripotent stem (iPS) cells in 2006, and since then, the stem cell community has been abuzz. iPS cells are differentiated adult cells (such as fibroblasts) that are reprogrammed into a more embryonic state via the introduction of three or four transcription factor genes. It’s the cellular equivalent of rewinding a car’s odometer, except that the resulting cells really do get “younger.”
For stem cell researchers, or at least those interested in embryonic stem cells (ESC), iPS cells were a godsend: a population of cells seemingly offering all the benefits of ESCs, with none of the messy political and ethical entanglements. Researchers could use them, among other things, to study how genetic mutations derail normal cellular differentiation processes, as well as how those errors can be blocked with drugs.
Yet iPS cells have a number of significant drawbacks, not the least of which is that they are, well, inducible. To make an iPS cell, researchers must somehow get the cells to take up and express exogenous genes, a fact that has cast a shadow over this research since day 1. After all, it’s difficult to imagine human therapeutic strategies that require the introduction of potential oncogenes.
As it turns out, there is another source of pluripotent stem cells. Unipotent spermatogonial stem cells (SSC) in culture can spontaneously revert to a more ESC-like state, just as iPS cells do, only without the gene delivery issues.
Takashi Shinohara, professor of molecular genetics at Kyoto University, first described the process in neonatal mouse testis-derived SSCs in 2004. 
In that study, Shinohara took testicular cells from neonatal (0–2-day-old) mice and kept them in culture for four to seven weeks. By that point, some of the cells had changed morphology, from the grape-like clusters of germline stem cells, to the more tightly packed colonies typical of ESC cultures.
Those “ES-like” cells express a blend of both germline and embryonic stem cell markers, including both beta-1 and alpha-6 integrins (germline) and SSEA-1 (embryonic). Yet they can differentiate in culture to form representatives of all three germ layers, produce teratomas in vivo, and when injected into blastocysts, produce live chimeric animals. 
“What you have is what iPS people have been dreaming of, to take a somatic cell and create iPS cells without adding exogenous factors,” says Hans Schöler, director of the department of cell and developmental biology at the Max Planck Institute for Molecular Biomedicine.
Schöler calls these cells “germline pluripotent stem cells,” or gPS cells -- the germline equivalent of iPS cells. He studies murine gPS cells to understand their biology – how factors like Oct4, for instance, drive their decision to either self-renew or differentiate. “By culturing germline stem cells, you can unravel their potential,” he says.
Among the SSC properties that interest Schöler is the fact that of all the cells in the body, only they bear the responsibility for passing genetic material to the next generation. Their DNA should be relatively pristine, he says, and thus represent the best source material for potential therapeutics. “Sperm and eggs have to have the best DNA,” Schöler explains. “The body is using a lot of energy to ensure that the genetic information that is passed on to the next generation is of high quality.”
Schöler’s lab demonstrated that gPS cells can be obtained from adult mouse testes, and like Shinohara before him, the resulting cells were pluripotent – for the most part.  The one pluripotency test those gPS cells didn’t pass, Schöler notes, was a tetraploid aggregation assay, in which the cells must reconstitute an entire embryo. That test failed, Schöler says, likely because gPS cells, coming as they do from male gonadal tissue, carry male epigenetic imprinting. “They have nothing female because all that memory has been erased.” The resulting embryos contain no maternal imprinting at all, a state that results in fetal death.
Still, says Schöler, “There’s no better system than working with germline stem cells if you are interested in reprogramming, because you have a test system at both ends.”
“Stem cell” is really a functional definition, Schöler explains, and the only true test of “stem-ness” is a cell’s ability to differentiate into something else. In this case, the starting cells (SSCs) can be injected into testes to produce sperm, while the gPS cells can be injected into blastocysts to reconstitute chimeric animals.
But just what are gPS cells? They don’t seem to be members of the in vivo population, as cells taken from one testis and transferred directly to another animal are not pluripotent, says Jon Oatley, an assistant professor of molecular biosciences at Washington State University College of Veterinary Medicine who studies transcriptional regulation of SSC self-renewal and differentiation.
It isn’t even clear whether it is the SSC cells themselves or some other testicular cell population that leads to the pluripotent cell, Oatley says. That’s because SSCs cannot be purified, only enriched. “The greatest purity that’s been achieved is only 15% of a population being SSCs,” he says. And because the resulting cells must be passaged in vitro, the ES-like cells could be “an in vitro phenomenon. It really could be an artifact,” he says.
Robert Braun, associate director of The Jackson Laboratory, studies another aspect of SSC biology. Given that SSCs produce the progenitor cells that ultimately produce sperm, it is likely, he says, that some instances of male infertility arise from defects in SSC self-renewal.
In 2004, Braun’s lab identified one such gene linked to the mouse luxoid phenotype, called Plzf, a zinc finger transcription factor that is mutated in some individuals with acute promyelocytic leukemia.  Loss of Plzf, Braun’s team demonstrated, prevents SSC self-renewal, such that these cells were no longer capable of repopulating stem cell-denuded testes.
More recently, Braun has been studying the possibility of targeting SSCs or their descendants to produce a male contraceptive. Using funding from the contraceptive development group at NICHD, Braun and his team are investigating the genetic regulation of the transition from spermatogonia (for instance, A(aligned-16) cells) to more differentiated cells such as A(1) cells.
At the moment, Braun is not yet testing specific compounds. “We’re really at the stage of still doing the basic biology,” he says. But the basic idea is to replicate the process of starting and stopping sperm production as occurs in “seasonal breeders” such as hamsters, groundhogs, minks, and lemurs.
The flip side of this coin uses SSCs is to restore fertility, for instance in young men who through pediatric chemotherapy treatment have lost the ability to produce sperm. (Adult males who undergo chemotherapy have the option to bank sperm ahead of time, but juveniles who have not yet entered puberty, do not have that option, says Ans van Pelt of the Center for Reproductive Medicine at the Academic Medical Center, University of Amsterdam.
van Pelt has published a pair of reports in the Journal of the American Medical Association suggesting that she can maintain human SSCs in culture from both adult  and juvenile males  for several months (28 and 29 weeks, respectively). These cells could be cryopreserved and restored in culture, expressed spermatogonial markers, and were capable of repopulating the testes of xenotransplanted mice.
According to the authors, the adult SSC report “outlines to first, to our knowledge, successful long-term culture and propagation of human spermatogonial stem cells.” 
As with gPS cells, the culture step is key in this process, though for a different reason: Because adult testes are larger than juvenile ones, more cells are needed to restore them, says van Pelt. (She also observed, and removed, gPS colonies in her cultures.)
But Oatley, for one, is skeptical. “The hallmark of stem cells is the ability to transplant the cells and show they can re-derive a cell lineage,” he says. Ralph Brinster of the University of Pennsylvania pioneered that approach for mouse SSCs (both Oatley and Shinohara are former Brinster postdocs). But van Pelt and colleagues could not do that experiment in humans, because the Center for Reproductive Medicine does not yet have the regulatory approval required to restore the cells to their human donors; they only had approval to take the biopsies in the first place.
Now, says van Pelt, she must ensure these cells are genetically stable – a common concern with cultured stem cells.
She should have plenty of time to work that out, she says. The two boys in her most recent study were aged 6.5 and 8 years at the time of biopsy, and so are probably 8.5 and 10 now. They likely will not return for reimplantation until they are 20 years old at least, she says. “So we have still time to improve our methods.”
 M. Kanatsu-Shinohara et al., “Generation of pluripotent stem cells from neonatal mouse testis,” Cell, 119:1001–12, 2004.
 K. Ko et al., “Induction of pluripotency in adult unipotent germline stem cells,” Cell Stem Cell, 5:87–96, 2009.
 F.W. Buaas et al., “Plzf is required in adult male germ cells for stem cell self-renewal,” Nat Genet, 36:647–52, 2004.
 H. Sadri-Ardekani et al., “Propagation of human spermatogonial stem cells in vitro,” JAMA, 302:2127–34, 2009.
 H. Sadri-Ardekani et al., “In vitro propagation of human prepubertal spermatogonial stem cells,” JAMA, 305:2416–8, 2011.