A new translation and reader's guide to the first detailed description of the first wave of spermatogenesis in the mouse

Introduction

The foundation for male fertility is provided by spermatogonia, which form in the postnatal mammalian testis. The developmental origin of spermatogonia was unclear until the mid-twentieth century – before then, there was a debate among scientists studying different vertebrate model organisms (e.g. mice, birds, amphibians) whether spermatogonia originated from prospermatogonia (also termed gonocytes) or from somatic Sertoli cells. This debate was settled, at least in rodents, in a landmark study in 1957 by Yves Clermont and Bernard Perey (Clermont & Perey, 1957). They performed quantitative histological analyses of the germ and somatic cell populations in testes from fetal and postnatal rats and concluded that prospermatogonia divide to form spermatogonia, while the Sertoli cell population expands until postnatal day 15 before terminally differentiating to support ongoing spermatogenesis.

We became interested in this 1889 manuscript by Dr. Friedrich Hermann entitled Postfetal Histiogenesis of the Mouse Testicle up to Puberty based on a reference in the aforementioned Clermont-Perey study (Clermont & Perey, 1957). They stated, “Hermann gave a brief histological account of the development of the [seminiferous] tubule in support of his opinion [that prospermatogonia produce spermatogonia]”. If true, then Hermann’s manuscript describes one of the earliest studies providing evidence for the continuity of the mammalian male germline. We recently completed two other projects to translate ‘forgotten manuscripts from the nineteenth century (Geyer, 2018; Jones, Harris, & Geyer, 2019; Sertoli, 2018), and decided to translate this one from German to English and provide a modern interpretation of the findings. Our goal is to uncover groundbreaking discoveries from that era so they can be a historical resource for scientists of the modern era.

We were surprised to find Dr. Hermann’s study did not address the prospermatogonia-to-spermatogonia transition, but rather provided the first known description of the first round, or wave of spermatogenesis. The first wave begins in the postnatal testis with the formation of spermatogonia and culminates with the formation of the first testicular spermatozoa. It is during this developmental period that fundamental germ and somatic cell populations form as the seminiferous epithelium matures in preparation for supporting steady-state spermatogenesis and thus fertility during the lengthy male reproductive lifespan.

While this study was done using mice, the conclusions have broad relevance, as a first wave of spermatogenesis must occur in all mammals. The first spermatogonia typically appear shortly after birth; in males of many species (e.g. mouse, rat, rabbit, cat, goat, donkey, bovine), spermatogenesis then progresses without interruption to form meiotic spermatocytes, postmeiotic spermatids, and ultimately testicular sperm. In humans and other primates, there is a lengthy temporal interruption between the appearance of spermatogonia and their completion of spermatogenesis, which occurs at puberty. In seasonal breeding mammals, a first wave of spermatogenesis must occur at the beginning of each breeding season, as a full complement of germ cells within the seminiferous epithelium must be formed from spermatogonia.

In this manuscript, Dr. Hermann made several important observations. First, he separated male germ cell development into two phases, which he called “the process of extensive regeneration” (from spermatogonia through their development into haploid round spermatids, which we now call ‘spermatogenesis’) and “the actual spermatogenesis” (spermatid morphogenesis, from haploid round spermatids to testicular sperm which we now call ‘spermiogenesis’). He correctly reasoned that the former produced cells for the latter, but at the time researchers in the field seemed to have trouble following the stepwise developmental progression of male germ cells. Dr. Hermann made the prescient observation that male germ cell development in the neonatal and juvenile testis closely resembled adult spermatogenesis, and therefore completed a temporal histological analysis of the neonatal and juvenile testis in which only spermatogenesis (but not spermiogenesis) is occurring.

It is unclear whether this manuscript was reviewed by peer scientists. However, suggestions to improve the study might have included: 1) clarify what was meant by “newborn” – today, that would be postnatal day (P)0, but the presence of mitotic germ cells in Fig. 1 makes that unlikely; 2) perform and analysis and provide images of testes from additional ages, at regular intervals after birth – this would have enabled a clearer determination of cellular relationships during germ cell development; 3) the manuscript ended rather abruptly, so provide a clear summary of the study. We will unfortunately never know why Dr. Hermann made these and other decisions, but he may have been under similar pressures as current scientists who must balance limited time, resources, and technical help when deciding when to complete and publish studies. Overall, this is an interesting read that provides the perspective of a scientist who, 132 years ago, laid foundational groundwork for the current field of male reproduction.

Fig. 1.

Cross-section through a seminiferous tubule of the newborn mouse²¹. 500/1; “fz.” = follicular (Sertoli) cell; “sp.” = spermatogonium²².

Postfetal Histiogenesis of the Mouse Testicle up to Puberty

By

Dr. Friedrich Hermann

Lecturer at the Anatomical Institute at the University of Erlangen

___

See Panel XXVI.

___

My investigations on this subject were not only motivated by a certain curiosity to enter a field to date considered fairly unknown, but rather by a different reason entirely. As is well known, one of the main challenges precluding the histological analysis of testes in the adult mammal is the fact that two processes coincide in the epithelial wall of the seminiferous tubule: first the actual spermatogenesis¹ (now termed spermiogenesis), that is the formation of seminal elements out of their cellular precursors, and then the process of extensive regeneration² (now termed spermatogenesis), designed to replace the cell elements that have been lost through spermatogenesis (spermiogenesis). If we imagine these two processes as two circular lines, we would have to consider the union of the <round> spermatids with a Benda foot-cell³ (Sertoli cell) as the starting point and the spermatozoon floating in the lumen of the seminiferous tubule as the end point <of spermiogenesis>, while the second would begin with the formation of new spermatogonia and end with the mature <round> spermatids (spermatogenesis). A new challenge arises from the fact that the starting points of the two circular lines do not coincide in the epithelium of the seminiferous tubule, i.e. the new formation of spermatogonia does not correspond to the first phase of spermatogenesis (spermiogenesis), but rather the two circular lines appear somewhat shifted. A brief glance not only at earlier but also recent relevant literature shows that this latter circumstance has caused much confusion and uncertainty regarding the histology of the testicle.

We propose that following the postfetal histiogenesis of the testicle into puberty⁴ can bring clarity to this situation; in the young animal, we can eliminate one circle, the process of spermatogenesis (spermiogenesis), foregrounding the other (spermatogenesis) clearly⁵. This is based on the assumption that the process of growth in the young testicles will follow the same path as that of the regeneration in the adult testicles⁶.

Finally, there was yet another point to prove: In an earlier work (Hermann, 1889), I sided with those authors who see elements in the so-called Benda foot-cells (Sertoli cells), which are completely uninvolved in the spermatogenetic process and only function as supporting elements alongside which the <round> spermatids mature into spermatozoa⁷. I believed to have provided new evidence for this view, as I was able to demonstrate a peculiar nucleolar formation in the nucleus of Benda's foot-cell (Sertoli cell), which occurs in the same typical manner throughout the entire process of seminal formation. This evidence should be even more convincing if it were possible to find nuclei <of Sertoli cells> in very young testes that exhibit the characteristic nucleolar formation, i.e. if it were possible to establish that two different elements are involved in the construction of the testicle from the outset: on the one hand supporting cells and on the other the actual glandular or seminal formation elements⁸.

This work will discuss the extent to which the facts discovered in the investigation correspond to the assumptions presented above.

First, we shall provide an overview of the very sparse information available in the literature about our subject. The first person to study the structure of the sexually immature testicle is La Valette St. George (La Valette-St. George, 1878), who examined it in calves, rabbits, dogs, and humans. According to this author, the sexually immature testicular tubules are embedded in a protein mass completely filling the canal lumen and they contain two kinds of nuclei: firstly, small round or oval nuclei <of Sertoli cells>, and secondly, larger round nuclei <of either spermatogonia or prospermatogonia/gonocytes>⁹, which are distributed unevenly among the smaller ones and are surrounded by a layer of sharply demarcated protoplasm. These structures represent the spermatogonia, which multiply within the former structures, the so-called follicular (Sertoli) cells. Thus, La Valette St. George assumes two elements within the immature seminiferous tubules.

Biondi’s view (Wiedersperg, 1885), also mostly supported by Niessing (Niessing, 1889), stands in complete contrast to La Valette St. George's understanding. Since Biondi completely denies the existence of supporting <Sertoli> cell elements (follicle cells, Benda’s foot-cells), he assumes only elements of the same type in the testes of sexually immature animals. Thus, the illustrations of a calf's testicle that accompany Biondi’s description show nothing more than round <germ> cell nuclei. These are embedded in an intermediate substance, which takes on a peculiarly network-like, jagged structure when immersed in a hardening agent. The rather inadequate drawings by Niessing also reveal nothing of the dual nature of the elements that make up the seminiferous tubules.

My investigations, in which young white mice were studied from the first hours of life up to the 6th week¹⁰, were able to fully confirm La Valette St. George's view.

After the application of the hardening and tincturing method detailed in an earlier work (Niessing, 1889), fine cross-sections through the testes of newborn mice prove that the testicular tubule does not yet have an actual lumen, but rather is filled with a protoplasmic mass, as La Valette St. George has previously noted. Luminous, fine but clear lines, which run through this mass, demonstrate that this is not a structureless protein mass or a homogeneous intermediate substance, as La Valette St. George and Biondi claim, but rather that it is derived from the cell bodies of La Valette St. George's follicle (Sertoli) cells. After examining my preparations, I must insist on interpreting the fine lines as cell boundaries <of Sertoli cells> and thus reject Biondi’s assumption that the peculiar network-like structure of the protoplasmic mass is simply due to the action of hardening agents. The light lines run too clearly and regularly and often allow the complete demarcation of a follicle (Sertoli) cell with its nucleus. It comes as no surprise that the fine lines delimit nucleated protoplasmic fields from the center of the tubule, for the follicular (Sertoli) cells represent relatively large, elongated elements with lobed extensions towards the center of the tubule. Numerically, the follicle (Sertoli) cells dominate in the testes of the newborn mouse. Their nuclei can be seen everywhere, in no particular arrangement, lying on the basement membrane as well as against the center of the tubule. The center itself, however, never contains any follicular (Sertoli) cell nuclei, but rather only the aforementioned extensions. In any case, it can be said with certainty that the follicular (Sertoli) cells are far more numerous in the juvenile testis, as a comparison of the cross-sectional image in Figure 1 with an image of an adult mouse testis in Figure 2 clearly shows. Regarding the nuclei of the follicular (Sertoli) cells, I was pleased to find in them the same peculiar nucleoli formations I described in an earlier essay (Niessing, 1889) as characteristic of the nuclei of Benda's foot-cell (Sertoli cell) in the <adult> testicular epithelium. However, while here the nucleus always has only one, relatively large nucleolus, we see 3 to 4 smaller nucleoli in the nuclei of the juvenile follicle (Sertoli) cells (Fig. 3).

Fig. 2.

Tangential section through a testicular tubule of the adult mouse. 500/1.

Fig. 3.

A spermatogonium²³ and 2 follicle (Sertoli) cells from the newborn mouse at higher magnification. 1500/1.

Surrounding these <Sertoli> cells, we find the second type of cell formations, v. la Valette's St. George's spermatogonia, in the form of strongly contoured, relatively large elements. In a field as complicated as testicular histology, especially with regard to the nomenclature, it is advisable to avoid new terms, but rather to maintain established terms wherever possible¹¹. Nevertheless, it is important to note that the spermatogonia in the adolescent and the <adult> testicle should not be understood as absolutely identical formations. Primarily, adult testicles are significantly larger in size. Furthermore, the juvenile spermatogonia are more reminiscent of young ova. The rounded cell bodies <of postnatal spermatogonia> (Fig. 3) show a clear stratification in an inner dense and a peripheral loosely reticular protoplasm. The nucleus also lacks the real nucleoli of the spermatogonia in the adult testis, which become visible by way of safranine coloration. Instead, it contains only one or more coarse chunks of chromatin in a rather dense chromatic network. Extremely numerous mitoses (Fig. 1) now show the juvenile spermatogonia in active proliferation, with the monaster stage¹² being observed particularly frequently¹³.

The stage described above persists for a long time: The testes of 9, 12, and 14 day old mice have enlarged somewhat as an organ, but their histological structure still shows exactly the same characteristics, and the thickness of the individual seminiferous tubules—on average 0.02 mm—has remained the same as in the newborn animal¹⁴.

On the 15th or 16th day of life, however, this changes suddenly, and we can now observe the first onset of puberty development in the mouse. Let us now turn to the finer processes that take place in the epithelial elements of the seminiferous tubule during this process. Until then, the nuclear divisions of the spermatogonia only yielded equivalent daughter cells, while they now create new cells which differ from the mother cell. These young cells¹⁵ (Fig. 4) are smaller, no longer resemble the egg-like appearance of their mother cells, and are characterized by multiple, well-developed true nucleoli—in short, they represent cells which are exactly the same as the spermatogonia we find in the testes of the adult animal. These spermatogonia (spermatocytes) are produced so rapidly that we find them stratified in three to four concentric layers. Of course, this process of spermatogonia (spermatocyte) formation also impacts the second type of cell, the follicle (Sertoli) cells: The numerous new formations of actual sperm (germ, or sperm-producing) cells force them apart, so that spermatogonia¹⁶ now outnumber follicle (Sertoli) cells, corresponding more and more to the numerical relation between the two types of cells in the adult animal. The position of the follicular (Sertoli) cells within the seminiferous tubule has also changed in that the rapidly growing spermatogonia¹⁷ have pushed them out of their position in the interior of the seminiferous tubules and have gradually pressed them against the membrana propria in a single layer. The above-mentioned extensions gradually disappear: At the beginning, these more or less thinned strips of protoplasm among neighboring groups of spermatogonia can be seen moving towards the center. Later, however, they are completely detached from the nucleated part of the walled follicle (Sertoli) cell, leaving a protein mass condensing more and more until it soon forms an initially bulged lumen inside the originally solid seminiferous tubule. This change of position goes hand in hand with a change in the interior of the follicular (Sertoli) cell’s nucleus, insofar as the multiple nucleoli transform into a single, larger nucleolus, which occurs by a simple fusion, as I was able to directly observe (Fig. 5).

Fig. 4.

Newly formed sperm layers²⁴. Mouse, 16 days old. 500/1.

Fig. 5.

Union of two nucleoli into a single larger one <in the nucleus of a spermatogonium>. Mouse, 16 days old. 1500/1.

Incidentally, not all tubules in the cross-section of the testicle present the picture just described: While many are still in the original, youthful stage, others have already advanced in their development. Very often there are tubules in which the spermatogonia¹⁸ nuclei, arranged in 2 to 3 layers, have transformed into the characteristic tightly coiled clusters of so-called ‘growing cells’¹⁹ (Fig. 6), while a layer of untransformed spermatogonia always remains in the wall. Occasionally, there are seminal tubules containing cells that have already transformed into the next higher cell category, <pachytene> spermatocytes (Fig. 7). It is evident that this rapid growth leads to an enlargement not only of the whole testicle but also of the individual tubules: The diameter of the individual seminiferous tubules is now 0.047 mm on average.

Fig. 6.

Transformation of the spermatogonia into so-called ‘growing cells’²⁵. Mouse, 16 days old. 500/1.

Fig. 7.

Transformation of the ‘growing cells’²⁶ into <pachytene> spermatocytes. Mouse, 16 days old. 500/1.

Once the development of the seminiferous tubule has progressed to the characteristic, loose spiral formations of the <pachytene> spermatocytes, a relatively long pause follows, as there are no other phases of division of the <pachytene> spermatocytes in the period leading up to the 21st day of life. The possibility of such a long phase of nuclear division might be of general histological interest: The fact that the spiral <pachytene> stages far outnumber the other phases of mitosis (meiosis) in the adult testis has led to the conclusion that there must be a resting phase in the spiral <pachytene> stage, though the length had not yet been determined. The examination of the growing testicle delivers a direct answer to this question: We can assume that the <pachytene> spermatocyte remains in the prophase of the monospirem (meiosis) for at least 5 days before it progresses to the metakinesis (diakinesis) and then to the subsequent anaphases²⁰. These only appear on the 21st day and must then take place all the more rapidly, because on this day we already see groups of newly formed <round> spermatids here and there. The main developmental period of the latter spans from the 21st day to the end of the fourth week of life, and the newly formed <round> spermatids do not differ in any way from those of the adult animal: They also contain the head cap (acrosome) and the secondary nucleus, the formation of which I was unfortunately unable to observe here either. The average thickness of these spermatid generations reaches 0.078 mm.

At the end of the 4th week of life, the first process (spermatogenesis, from spermatogonia to round spermatids), which we understood as regenerative in the adult testicle, and which begins with the formation of new spermatogonia and ends with the formation of <round> spermatids, is completed in the young animal. In the 5th week of life, the second process (spermiogenesis) the actual sperm formation process, begins: The spermatid groups are transformed in the manner described earlier (Niessing, 1889) and come into contact with the Benda foot-cells (Sertoli cells). In the 6th week, we find all developmental phases of the spermatozoon up to the mature spermatozoon, which lies in the lumen of the seminiferous tubules.

By the 6th week, the mouse has already reached full sexual maturity, though I am unable to determine whether it actually copulates at this relatively young age, as I have never observed it myself.

Explanation of the Illustrations on Panel XXYI.

All figures are designed with a Camera Lucida, using an apochromatic oil immersion lens from Zeiss; aperture 1.2, focal length 2.0 and ocular 4 and 12. Enlargement 500–1500.