| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Review |
From the Department of Biological Sciences, University of New Hampshire, Durham, New Hampshire.
| Correspondence to: Dr William Berndtson, Department of Biological Sciences, University of New Hampshire, Durham, NH 03824 (e-mail: bill.berndtson{at}unh.edu). |
| Received for publication July 6, 2009; accepted for publication July 8, 2010. |
All approaches for quantifying treatment effects on sperm production
require one or more technical assumptions. Their validity can impact the
integrity of the resulting data, but most had not been subjected to thorough,
critical examination. This review was undertaken to 1) identify the
assumptions required with each of several popular histometric methods for
quantifying treatment effects on sperm production, 2) cite examples in which
each assumption had failed, 3) assess potential consequences of the failure of
each technical assumption, and 4) consider safeguards by which one might avoid
misleading outcomes in the event that a specific assumption did not hold.
Key words: Spermatogenesis, testis, histometric evaluation
This review was undertaken to consider the validity of the assumptions associated with several popular histometric methods for quantifying treatment effects on sperm production, to assess potential consequences of an erroneous assumption, and to suggest safeguards that might be applied in future studies to limit potential inaccurate or misleading findings, should an assumption fail to hold. The assumptions considered herein are ones that may be unavoidable with one or more of the evaluation methods to be considered. Many other assumptions that can and should be avoided are evident within the scientific literature. These include the common practice of assuming that a technique that has been described within the literature can be used without undergoing validation in the investigator's own lab. For example, determinations of sperm production via the enumeration of elongated spermatids in testicular homogenates are often conducted without validation for the species of animal, homogenizer model, volume of the homogenization vessel, sharpness of blades, operating speed, composition of the homogenization fluid, duration of homogenization, volume of fluid used, etc.
Each evaluation method should undergo validation of meaningful rigor prior to its adoption in an individual lab. The literature contains numerous examples in which a new procedure or modifications to an existing one have been adopted after "validation" by a preliminary study of such limited power and sensitivity that it would be incapable of detecting a problem with the new method if one existed. Although such avoidable assumptions can be of considerable consequence, this review will focus on requisite assumptions that must be considered when quantifying sperm production by the most popular histometric methods. The methods for assessing sperm production chosen for this review include determination of the number of germ cells per Sertoli cell or per seminiferous tubular cross section, direct enumeration of degenerating germ cells, assessment of sperm production via volume density and optical dissector approaches, determination of the numbers of elongated spermatids embedded in the apex of individual Sertoli cells, and the determination of daily sperm production (DSP) via enumeration of elongated spermatids in testicular homogenates.
Methods for Quantifying Sperm Production
Germ Cells per Sertoli Cell or per Seminiferous Tubular Cross Section—
Relative changes in sperm production may be assessed via direct enumeration
of germ cells and Sertoli cells within round seminiferous tubular cross
sections at a given stage(s) of spermatogenesis
(Clermont and Morgentaler,
1955). The results are expressed as numbers of germ cells per
single Sertoli cell or per some standard number of Sertoli cells, such as the
average per tubular cross section among control subjects or among all animals
within the experiment.
For many studies, a single stage containing all of the major classes of germ cells (ie, spermatogonia, spermatocytes, and spermatids) is chosen as representative of spermatogenesis as a whole. Each round tubular cross section of the chosen stage(s) is evaluated until the predetermined number of tubules has been evaluated. Next, the number of germ cells of each type and the number of Sertoli cells are determined by direct observation. All nuclei are counted, whether residing entirely within the section or only partially present because of sectioning. As explained subsequently, counts are based on the spherical nuclei of germ cells and on nuclei of Sertoli cells with a spherical nucleolus. Elongated spermatids and the irregularly shaped nuclei of Sertoli cells per se are not counted.
Because cellular associations (eg, stages) are confined to discrete patches in man (Heller and Clermont, 1964; Johnson, 1994), a single tubular cross section would typically contain cells at 2 or more stages of development. This arrangement precludes use of the technique of Clermont and Morgentaler (1955) for the human. The human testis is unique in several other ways, and readers might find the review on human spermatogenesis by Amann (2008) of considerable value. The latter includes discussion of many important factors related to the evaluation of human spermatogenesis that are relevant to but beyond the scope of this review.
Correction for partial nuclei. Data obtained as described above are designated as crude counts (Clermont and Morgentaler, 1955). Adjustment of crude counts is necessary because the inclusion of nuclear fragments causes overestimation of the actual number of nuclei per unit of tissue, and this overestimation is greater for the larger nuclei, which are sectioned more frequently by the microtome. Adjustment to whole-cell equivalency is necessary.
Several equations have been advanced for adjusting crude counts to
whole-cell equivalents or so-called true counts. The Abercrombie formula
(Abercrombie, 1946)
accomplishes this by multiplying the crude count by an Abercrombie correction
factor, which equals the section thickness divided by the sum of the section
thickness plus nuclear diameter of the cell of interest, as follows:
|
|
For hypothetical 1- and 5-µm-diameter nuclei within a 5-µm-thick section, Abercrombie correction factors would be 0.833 and 0.5, respectively. Thus, crude counts would overestimate the number of 1-µm-diameter nuclei by 16.7% (ie, 1.0 – 0.833 = 0.167) and numbers of 5-µm-diameter nuclei by 50% in a 5-µm section.
Because the Abercrombie formula is applicable only for spherical structures (Abercrombie, 1946), elongated spermatids are not counted. A stage containing younger spherical spermatids is usually chosen for evaluation. Similarly, the nuclei of Sertoli cells have an irregular outline with numerous invaginations (Elftman, 1950; Leblond and Clermont, 1952; Lino, 1971; Steinberger and Steinberger, 1977; Johnson et al, 1984b; Johnson, 1986). It has become customary to count only those Sertoli nuclei with a visible nucleolus; the Abercrombie correction is applied to the spherical nucleolus to give a true count for Sertoli cells (Lino, 1971; Berndtson, 1977).
The Sertoli cell correction. After crude counts have been converted to true counts (whole cell equivalents), data are further adjusted by a Sertoli cell correction (Clermont and Morgentaler, 1955). This simply involves reporting germ cell numbers in relation to a single or some standard number of Sertoli cells. This provides an internal adjustment for natural or treatment-induced changes in seminiferous tubular dimensions. It was based on the early belief that Sertoli cell populations were numerically stable in adult animals. Under that assumption, to be discussed later, changes in germ cell numbers would produce proportional changes in the ratio of germ cells to Sertoli cells. Crowding or increased spacing among cells due to changes in seminiferous tubular dimensions, whether due to treatment or tissue processing, would be irrelevant, because it is the ratio of cell types and not their number per unit volume of tissue that is important.
Direct Quantitation of Degenerating Germ Cells— The direct quantification of degenerating germ cells has been applied to assess germ cell losses in both normal and treated animals (Russell and Clermont, 1977; Huckins, 1978; Russell et al, 1981). The basic approach is that of Clermont and Morgentaler (1955) described above, except that one counts degenerating rather than normal cells. Because of this similarity, each of the requisite technical assumptions associated with this basic approach, to be described subsequently, would apply similarly when quantifying degenerating cells.
Volume Density Approaches— The classical approach for quantifying volume density is a method of "random hits" within scanned sections of tissue (Chalkley, 1943; Eschenbrenner et al, 1948). For this, a microscope is fitted with a fixed pointer or pointers. The slide of testicular tissue is moved at random, after which the structure at the tip of the pointer is identified and recorded. Each hit is recorded, even though some hits may fall on poorly defined structures. The process is repeated many times. With adequate sampling, the frequency with which a given structure is hit will be proportional to its relative abundance (or volume density) within the tissue. For example, if Sertoli cell nuclei occupied 2.1% of the testis volume, one would expect Sertoli cell nuclei to be hit 2.1% of the time.
Estimating cell numbers from their volume density. To use the
volume density approach to estimate the number of cells of any given type per
gram of testis, per testis, or per male, one must determine the volume of
testicular parenchyma. Because the specific gravity of testicular tissue of
most species is very close to 1.0 (eg, bull:
Swierstra, 1966; rat:
de Jong and Sharpe, 1977;
stallion: Gebauer et al, 1974;
Johnson and Neaves, 1981;
human: Johnson et al, 1981;
mouse: Mori et al, 1982), this
is usually considered to equal the weight of the testicular parenchyma after
removal of the tunica albuginea. Next, one calculates the total volume of each
testicular component, which will equal the product of its volume density and
total testicular parenchymal volume. For example, if the volume of testicular
parenchyma per testis equaled 2.0 mL and the volume density of Sertoli nuclei
equaled 2.1%, the total volume of Sertoli nuclei per testis would equal 0.042
mL. The final step requires estimating the average volume of a single nucleus
for the cell(s) of interest. For spherical nuclei, this is determined via the
equation characterizing the volume of a sphere, as follows:
|
|
Estimating DSP via volume density approaches. To estimate DSP via the volume density approach, one simply divides the number of germ cells by a time divisor, which equals the estimated number of days of sperm production represented by the cells evaluated. It is customary to use the number of spherical spermatid nuclei per gram of tissue or per testis for this purpose. Spermatocytes or other less developmentally advanced types of germ cells could be used, by projecting the number of sperm that would theoretically result from the division of these younger progenitors. However, the use of younger germ cells is less reliable. Cellular attrition is a normal feature of spermatogenesis (eg, Barr et al, 1971; Berndtson, 1977; Huckins, 1978; Johnson et al, 1983), and the actual yield of spermatids from their younger progenitors could differ substantially from the theoretical expectation. Furthermore, experimental treatments that elicit adverse effects on spermatogenesis can amplify the magnitude of errors associated with estimates of sperm production derived from younger germ cells (Berndtson and Foote, 1997).
The Optical Dissector Approach for Assessing Cell Numbers—
Modern stereological methods have been applied to estimate the number of
testicular cells per unit volume of tissue. These approaches, hereinafter
designated collectively as the optical dissector approach, include features
enabling one to avoid the problems associated with irregularly shaped nuclei.
With this technique, cells are counted directly within a three-dimensional
frame, which Wreford (1995)
characterized as "analogous to counting the number of people in a space
defined by a square marked on the floor and limited by the ceiling
above." The estimated numerical density of a particular cell (est
NV) is determined by the equation:
|
|
Determining the Elongated Spermatid:Sertoli Cell Ratio From the Number of Spermatids Embedded in the Apex of the Sertoli Cell— Procedures for determining the number of germ cells per individual Sertoli cell via the direct enumeration of the respective cells per tubular cross section and/or by volume density approaches have been described. A third alternative involves the direct enumeration of elongated spermatids embedded in the Sertoli cell apex (Russell and Peterson, 1984; Johnson, 1986). For this, one first identifies the stage of the cycle by examination of thick sections (Russell and Peterson, 1984). Next, the Sertoli cells are sectioned obliquely to their long axis and at the proper height. Serial sections may be used to ensure that all of the elongated spermatids embedded within the cytoplasm of an individual Sertoli cell will be enumerated.
Determining DSP via the Numbers of Elongated Spermatids in Testicular Homogenates— During spermiogenesis, the chromatin of developing spermatids condenses and becomes resistant to homogenization of unfixed tissue (Amann and Almquist, 1961, 1962). The number of homogenization-resistant spermatids per testis or per gram of testicular tissue can be determined by first placing a known quantity of testicular tissue into a known quantity of fluid. After homogenization in a blender, the spermatids are enumerated by hemacytometry (Amann and Almquist, 1961, 1962). The relative magnitude of a treatment effect on sperm production can be assessed via a direct comparison of the number of homogenization-resistant spermatids per unit of tissue in untreated vs treated males. DSP can be estimated by dividing the number of homogenization-resistant spermatids by a time divisor that equals the number of days of sperm production represented by these cells.
The Validity of the Technical Assumptions
Numerous assumptions are embodied in the various approaches that might be
used to assess treatment effects on spermatogenesis. Some of these may not
apply or may be of greater potential consequence for one method than for
another. The validity of these assumptions and their potential impact on
experimental findings will now be considered.
Assumption: Section Thickness Is Equivalent to the Microtome Setting— Section thickness must be entered into the Abercrombie formula to permit crude counts to be converted to true counts when determining numbers of germ cells per seminiferous tubular cross section or per Sertoli cell by the method of Clermont and Morgentaler (1955). Based on a report that actual thickness could differ substantially from the nominal microtome setting and also among and within sections prepared at a single setting (Amann, 1970b), it was suggested that the thickness of each section should be determined for use in the Abercrombie formula (Berndtson, 1977). Increased accuracy and precision of modern microtomes may make that earlier recommendation unnecessary (Zhengwei et al, 1990; Wreford, 1995). Moreover, errors due to an incorrect section thickness would likely be randomly distributed or of similar magnitude and direction among all treatment groups, and are likely to be small in relation to those from other sources.
The impact of errors resulting from the use of an inaccurate microtome setting can easily be assessed by introducing different values into the Abercrombie formula (Abercrombie, 1946). For example, assume that preleptotene spermatocytes have a nuclear diameter of 6 µm, and that one counted such nuclei in sections prepared at a 5-µm setting. The Abercrombie correction factor (ie, section thickness divided by section thickness plus nuclear diameter) calculated for these cells would equal 0.4545. If the sections were 10% greater than the microtome setting, or 5.5 µm, the correct correction factor would have been 0.4783.
In this example, the correction factor based on the inaccurate setting was 5% lower than that based on the correct thickness. One would want to avoid a 5% error if possible, but an inaccuracy of 0.5 µm is substantial by modern standards. More importantly, if the instrument were precise (ie, all sections prepared at that setting were of the same 5.5-µm thickness), all errors would be of the same magnitude and direction among all animals and treatment groups. Relative differences in sperm production among treatments would not have been affected. Section thickness errors would alter Abercrombie correction factors for some cell types more than others, as a function of their differing nuclear diameters. This could be of consequence for some investigations, such as those examining the kinetics of spermatogenesis, but readers can easily assess such relationships by entering values of interest into the Abercrombie formula. For the range of nuclear diameters present among the germ cells, potential errors will usually be quite small.
Thick sections may be examined by the optical dissector method. With such an approach, measurements can be obtained within a depth of tissue that is less than the actual section thickness (Meachem et al, 1999). This eliminates the potential for errors of the types that are possible with some alternative evaluation methods because of differences between the microtome setting and actual section thickness.
Assumption: The Sertoli Cell Population Is Numerically Stable in Adult Animals and Is Not Altered by Treatment— The numerical stability of the Sertoli cell population is an assumption of importance when determining numbers of normal or degenerating germ cells per seminiferous tubular cross section, when assessing treatment effects by direct enumeration of elongated spermatids in the apex of individual Sertoli cells, or whenever spermatogenesis is characterized by germ cell:Sertoli cell ratios. Upon review of the early literature (Gondos and Berndtson, 1993), it appears that initial conclusions that Sertoli cells were numerically stable were based on 2 lines of evidence with rats. First, Sertoli cell mitotic activity appeared to cease by about 15–21 days of age (Clermont and Perey, 1957; Steinberger and Steinberger, 1971; Nagy, 1972; Orth, 1982; Gaytan and Aguilar, 1987). Second, Sertoli cells were extremely resistant to many harsh treatments, often surviving well upon exposure to treatments causing the complete obliteration of germ cells (eg, Oakberg, 1959; Clegg, 1963; Bergh, 1981). Despite such evidence, it has since been demonstrated that numbers of Sertoli cells can change in response to either natural or experimental factors. Sertoli cell numbers increase for a period of time prior to puberty (Clermont and Perey, 1957; Steinberger and Steinberger, 1971; Nagy, 1972; Orth, 1982; Gaytan and Aguilar, 1987; Zhengwei et al, 1990). However, a subsequent decline in Sertoli cell number has been reported as a function of age in the adult human (Johnson et al, 1984b) and, despite some inconsistent reports (Johnson and Thompson, 1983; Jones and Berndtson, 1986; Johnson et al, 1991), in the adult stallion. Fluctuations in the numerical size of the Sertoli cell population of postpubertal males have been reported in some seasonally breeding species (eg, Johnson and Thompson, 1983; Johnson and Nguyen, 1986; Johnson et al, 1991). Photoperiod is an important factor controlling natural seasonal breeding patterns in many species, and photoperiod restriction produced a 33% reduction in Sertoli cell numbers in 1 study with adult hamsters (Meachem et al, 2005). Although sound experimental designs (eg, inclusion of appropriate untreated controls at each age or season) may effectively safeguard against the confounding effects of most natural factors, the potential for Sertoli cell numbers to be altered in response to other experimental treatments must also be recognized. Investigators have begun to elucidate the molecular mechanisms associated with Sertoli cell proliferation in prepubertal animals. By applying such information, Meachem et al (1996) extended the normal period of Sertoli cell proliferation, resulting in increased Sertoli cell numbers that persisted into adulthood; the number of Sertoli cells per testis was 118% and 149% greater in 90-day-old rats that had been treated with follicle-stimulating hormone during the first 10 or 15 days of life, respectively. Collectively, the foregoing findings illustrate the potential for Sertoli cell numbers to fluctuate and/or to be permanently altered in response to factors imposed during the prepubertal period or during adulthood.
Assumption: The Number of Sertoli Cells per Seminiferous Tubular Cross Section Can Be Estimated Accurately by Direct Enumeration of Sertoli Nuclei With a Nucleolus— The irregular outline of Sertoli nuclei precludes the application of the Abercrombie formula to adjust crude counts of these nuclei to true counts. To overcome this problem, it is customary to count only those Sertoli nuclei containing a spherical nucleolus, and to apply the Abercrombie formula based on nucleolar dimensions. Some authors have reported the presence of more than 1 nucleolus per Sertoli cell, whereas others have not observed this phenomenon: for example, Johnson et al (1991) vs Wing and Christensen (1982). In the author's opinion, the presence of multiple nucleoli is unlikely to be a serious issue. In practice, a Sertoli cell is only counted once even if more than 1 nucleolus is present. The presence of more than 1 nucleolus would increase the likelihood that at least 1 of these would be visible, so that a given cell would be counted. However, the incidence of multiple nuclei should be similar among all replicates. Unless the number of nucleoli per Sertoli cell was affected by treatment, the presence of multiple nucleoli would not alter relative differences in the number of Sertoli cells per tubular cross section or germ cell:Sertoli cell ratios among treatments. In that context, readers are reminded that the method of Clermont and Morgentaler (1955), which requires adjustment of crude counts to true counts, is intended to assess quantitative changes in sperm production on a relative basis. It is not used to determine actual numbers of specific cells per unit of tissue. Therefore, unless a treatment altered the number of nucleoli per Sertoli cell, the results obtained by this approach would be valid even if some Sertoli nuclei contained more nucleoli than others.
Some investigators have reported that Sertoli nucleoli are not spherical (Johnson et al, 1991). It is possible that the morphology of these structures differs based on the fixatives employed, and it is suggested that the morphology of Sertoli nucleoli be considered before deviating from established tissue processing methods. Caution also should be exercised to distinguish nucleoli from other structures within the nucleoplasm. If one observes such precautions, the shape of nucleoli is unlikely to present an obstacle that would preclude their use for the Abercrombie correction.
Assumption: The Timing of Spermatogenic Events Is Constant and Unaffected by Treatment— At one time, the possibility that a treatment could alter the timing of spermatogenic events was considered unlikely. The length of 1 cycle of the seminiferous epithelium appeared to be constant and characteristic for each species or strain of animals (Clermont and Trott, 1969; Clermont, 1972). Subsequent evidence has provided some reason to question whether the timing of spermatogenic events within normal, untreated males is as constant as originally thought, while also demonstrating that the timing of spermatogenic events can be altered experimentally. Constancy in the timing of spermatogenic events is an assumption of some relevance, although to different degrees, with each method for quantifying spermatogenesis. Thus, this issue is worthy of special consideration.
Although spermatogenesis is a continuous process, researchers have found it useful to divide this into discrete stages, based largely on the different cellular associations (ie, combinations of cells) observed within round seminiferous tubular cross sections at any given point in time. These change continuously, and the interval between the first appearance of a given stage or cellular association until its reappearance at a given point within a seminiferous tubule constitutes 1 cycle of the seminiferous epithelium. A computer program called STAGES (www.cacheriverpress.com) and related publications by Hess (1990) and by Hess and Chen (1992) may be useful to readers with a further interest in this topic.
The frequency of a given stage of the cycle of the seminiferous epithelium should be proportional to its duration. Thus, if the time required for 1 complete cycle of the seminiferous epithelium was constant and the timing of specific spermatogenic events was rigidly controlled, the frequency at which one would observe a given stage should be virtually identical in all males of a given species.
The author is not aware of any studies providing a critical assessment of potential differences in the length of 1 complete cycle of the seminiferous epithelium among individuals of the same strain or species. In most studies, autoradiography has been used to monitor the progression of the most advanced labeled germ cells after injection of 3H-thymidine. However, with this approach, the observations from which cycle length has been calculated have been performed on different individuals, rather than the same individuals, at each time point postinjection (eg, Clermont and Trott, 1969).
Although critical assessments of among-animal variability in cycle length may be lacking, Hess et al (1990) found considerable variation in the frequency of stages among fifteen 90- to 110-day-old Sprague Dawley rats. Indeed, the frequencies for individual rats ranged from 9.9% to 19.4%, from 2.3% to 8.4%, from 1.3% to 3.5%, from 2.6% to 7.5%, from 5.2% to 9.7%, from 5.9% to 12.0%, from 19.2% to 24.2%, from 4.0% to 10.7%, from 1.7% to 4.6%, from 2.3% to 4.1%, from 1.6% to 4.7%, from 6.9% to 10.6%, from 3.3% to 9.3%, and from 5.1% to 8.8% for stages I to XIV, respectively. Clearly, 2- to 3-fold differences in the frequency or duration of individual stages were commonplace among individual rats. It is not clear whether such among-animal differences reflect actual differences in the duration of individual stages, imprecision associated with stage identification, or a combination of these variables. Because one would expect staging errors to be randomly distributed, one would expect among-animal differences attributable to such error to decrease with increasing sampling intensity. The data of Hess et al (1990) were based on the examination of 513–735 tubular cross sections per rat.
Although constancy in the duration of the cycle of the seminiferous epithelium or its component stages among individuals of the same strain or species may be uncertain, experimental alteration in the timing of spermatogenic events has been well documented. Several studies have demonstrated spermatogenic arrest by vitamin A deficiency in the rat, and subsequent progression with a high rate of stage synchronization (ie, with most tubules at the same stage of the seminiferous epithelium at any given point in time) upon vitamin A replacement (Morales and Griswold, 1987; Bartlett et al, 1990; Van Beek and Meistrich, 1990; Van Pelt and De Rooij, 1990).
Because the timing of spermatogenic events might vary among normal rats or because of experimental treatment, it is important to consider how this might impact experimental results with each of the methods one might employ to quantify sperm production rates. Two of the endpoints described herein involved the direct enumeration of cells within round seminiferous tubular cross sections at a specific stage(s) of the cycle of the seminiferous epithelium. Specifically, those involved the basic method of Clermont and Morgentaler (1955) by which one can quantify the number of normal or degenerating germ cells per round seminiferous tubular cross section (or per Sertoli cell). Inherent differences in the frequency or duration of individual stages would likely be of limited consequence for data derived by those approaches. This view is based on the expectation that, despite possible differences in the duration of a stage(s) chosen for evaluation, the specific combination of cells present during each stage should be similar. If a cellular association were to differ, a tubular cross section would be classified as belonging to a different stage. If a treatment caused a major disruption in the timing of spermatogenic events, stage identification might become imprecise. However, spermatogonia, spermatocytes, and spermatids are usually identified, and it is unlikely that such disturbances would remain unnoticed. Thus, constancy of the timing of various spermatogenic events would appear to merit only limited concern during evaluations by this method.
The assumption that the timing of spermatogenic events is constant would presumably merit only limited concern when assessing spermatogenesis via direct examination of the number of elongated spermatids embedded in the Sertoli cell apex. The conditions and criteria required in the selection of individual Sertoli cells for evaluation are quite restrictive (Russell and Peterson, 1984), and they would ensure that only Sertoli cells containing elongated spermatids would be evaluated. Thus, neither inherent among-animal differences in the duration of a particular stage nor treatment-induced alterations in the timing of spermiation would be likely to alter spermatid:Sertoli cell ratios derived by this approach. In contrast, the timing of spermatogenic events could have a substantive impact on the volume density of specific testicular components when assessed via the classical volume density approach or via the optical dissector method. The types of germ cells that are present at each stage differ. For example, elongated spermatids are usually present during only about one-half of the cycle of the seminiferous epithelium in most species. However, if spermiation was delayed, as observed during vitamin A deficiency in rats, elongated spermatids could persist throughout a much greater portion of the cycle, and their volume density and actual number would increase proportionately. This would render time divisors derived with untreated animals invalid and might yield an erroneous conclusion that sperm production had increased, when the release of elongated spermatids had simply been delayed. Also, because the volume density of all of the testicular components must total 100%, a change in the volume density of one component will alter the volume density of all others. Germ cell:Sertoli cell and germ cell:germ cell ratios would be impacted as a consequence.
Hess et al (1990) characterized the frequency of specific stages among males within tissues removed at a single point in time. Because stages follow each other in an orderly progression, the frequency of a given stage might also vary over time within a single individual. This could contribute to corresponding variability in estimates of DSP based on the volume density approach. Fortunately, the use of an adequate number of replicate animals should safeguard against an erroneous or misleading experimental outcome. Procedures for identifying an appropriate number of replicate animals have been described elsewhere (Berndtson, 1989, 1990, 1991, 2008; Berndtson et al, 1989).
Constancy in the timing of spermatogenic events is also a relevant consideration when quantifying spermatogenesis via the homogenization method. Inherent among-animal variations in the duration of 1 cycle of the seminiferous epithelium or of individual stages (Hess et al, 1990) could contribute to actual or apparent differences in sperm production among animals. Fortunately, this issue can be effectively addressed by providing an adequate number of replicate animals per each treatment group. In contrast, treatment-induced alterations, such as those cited with vitamin A deficiency and replenishment, would invalidate time divisors determined with untreated animals, and could have a profound effect on one's experimental outcome, similar to that described with the volume density approach. It is fortunate that experimentally induced arrested development and/or stage synchronization seems to be relatively uncommon. One should, nonetheless, be aware of such a possibility, and should provide safeguards against this when utilizing the homogenization method.
Assumption: Published Time Divisors Are Accurate and Applicable for General Use— DSP can be estimated by dividing the number of spermatids determined via volume density, optical dissector, or homogenization methods by an appropriate time divisor. The validity of the approach(es) used to determine time divisors warrants some scrutiny. As discussed elsewhere (Berndtson, 1977), one should not assume that a time divisor intended for application in converting numbers of homogenization-resistant spermatids to DSP can be determined simply by adding the length of the stages in which elongated spermatids are present during the cycle of the seminiferous epithelium. For example, Amann and Lambiase (1969) demonstrated that elongated stage V spermatids of the rabbit do not survive homogenization. Furthermore, when these investigators developed time divisors via 3 different approaches, they obtained values ranging from 3.0 to 3.5 days. Which of these (ie, 3.0 vs 3.5 days) constituted the most accurate mean time divisor for the rabbit remains unclear. Consensus may be lacking on what constitutes the best method for confirming the accuracy of a time divisor. However, even if such consensus could be achieved or if the accuracy of a given time divisor could be proven, other issues related to its applicability would remain. First, the validity and accuracy of time divisors is predicated on the assumption that the timing and duration of the spermatogenic events is constant. This assumption may not hold, because the frequency (duration) of individual stages may differ among untreated males (Hess et al, 1990) and because the timing of spermatogenic events can be altered experimentally (Morales and Griswold, 1987; Bartlett et al, 1990; Van Beek and Meistrich, 1990; Van Pelt and De Rooij, 1990). For a well-replicated study, published time divisors should approximate the average actual time divisor for untreated subjects. However, treatment-induced alterations in the timing of spermatogenic events could render published time divisors quite inaccurate. Researchers should recognize this potential and are encouraged to provide appropriate safeguards when estimating DSP via volume density, optical dissector, or homogenization methods.
Assumption: Tissue Shrinkage Is Negligible With Modern Tissue Processing Methods, or Is Similar for All Testicular Components— Shrinkage can be substantial for tissues processed by classical approaches. Volumetric shrinkage reported by some investigators included 45.3%–55% (Swierstra, 1966) and 48% (Amann, 1962) for bulls, 34%–44% for stallions (Gebauer, 1973), and 13%–58% for rabbits (Amann, 1970a). One should not assume that such changes in testicular volume would be randomly distributed, because Attal and Courot (1963) reported that shrinkage averaged 15%, 36%, and 43% for 0–3-, 3–7-, and >7-month-old bulls, respectively.
Although the shrinkage associated with classical tissue processing methods
is widely recognized, the presence and/or extent of shrinkage with modern
tissue processing methods has been inconsistent. In some studies, neither the
weight of rat testes (Johnson et al,
1984a) nor the specific gravity of equine testes
(Johnson and Neaves, 1981)
were altered by perfusion with 2% glutaraldehyde in 0.1 M sodium cacodylate
buffer, and seminiferous tubular diameter and volume density were similar in
freshly frozen vs 2% glutaraldehyde-perfused human testes
(Johnson et al, 1983). Also,
the fixation of testicular tissue from adult monkeys by perfusion or immersion
with Bouin fluid elicited less that a 2% change in tissue volume
(Zhengwei et al, 1997). In
another study in which tissues were preserved with Bouin fixative and embedded
in methacrylate (Meachem et al,
1996), 1 g of fresh tissue yielded a processed volume of 979
mm3. In contrast, Mori and Christensen
(1980) reported that the
volume of rat testes fixed by perfusion with 0.1 M collidine-buffered 3%
glutaraldehyde was increased by 20.1% above that in the fresh state.
Subsequent embedding (after dehydration) in glycol methacrylate or Epon caused
22.2% and 11.4% reductions in tissue volume, respectively. In another similar
study (Wing and Christensen,
1982), perfusion caused a 13.4% increase in rat testicular volume,
whereas subsequent dehydration and embedding reduced testicular volume to
45.9% of that after perfusion fixation. Finally, McLachlan et al
(1994) reported that
volumetric shrinkage was consistently 
27% for rat testicular tissue after
perfusion with 5% glutaraldehyde in 0.1 M collidine buffer and embedding in
Epon Araldite. As with classical fixation methods, one should not assume that
changes in tissue volume will be randomly distributed. Zhengwei et al
(1990) preserved testicular
tissue of rats aged 1–15 days by immersion and tissue from
20–70-day-old rats by perfusion with a fixative containing 5%
glutaraldehyde. Shrinkage varied as a function of animal age. Tissue volume
was reduced to 75.6%, 76.1%, 85.2%, and 84.6% of that in the fresh state for
animals aged 1, 5, 10 and 15 days, and to 86.7%–88.8% for rats aged
20–70 days, respectively.
Despite inconsistencies among various studies, the results of Mori and Christensen (1980), Wing and Christensen (1982), Zhengwei et al (1990), and McLachlan et al (1994) confirm the potential for testicular volume to be altered with some modern fixatives and tissue processing procedures. Mori and Christensen (1980) suggested that the volume increase in their study probably resulted from perfusion pressure, and that it may have affected expandable compartments such as blood vessels and lymphatic spaces more than other components. These authors did not attempt to confirm this possibility. Because differential shrinkage has been reported for at least 1 other tissue (Bertram et al, 1986), one cannot exclude the possibility that shrinkage of testicular tissues might not be homogeneous. The fixative employed in a given study would appear to be an important factor impacting the extent of tissue shrinkage, and also the quality of the resulting material. Readers are directed to publications by Harleman and Nolte (1997), Lanning et al (2002), and Latendresse et al (2002) for more extensive discussion of this topic.
Tissue processing–induced changes in testicular volume would be of little concern with several of the methods one might choose to assess sperm production rates. Changes in tissue volume are irrelevant during assessments of the number of normal or degenerating germ cells per seminiferous tubular cross section or per Sertoli cell by the method of Clermont and Morgentaler (1955), because the germ cell:Sertoli cell ratio is independent of tissue volume. Similarly, the number of elongated spermatids embedded in the apex of individual Sertoli cells should not be affected by tissue volume, and tissue shrinkage should not alter results when either fresh or frozen-thawed tissues are used for the homogenization method. Where fixed tissues are to be homogenized, the need to measure and correct for shrinkage can be avoided by homogenizing the entire testis or by determining the fresh volume (or weight) of the sample to be homogenized prior to fixation. In contrast, changes in either the absolute volume of a tissue or differential shrinkage among the various components of the testis could have a profound impact on volume density estimates and any data on cell numbers derived from such estimates. Similarly, shrinkage could impact the number of cells per gram of fresh tissue or per testis determined via the optical dissector approach (McLachlan et al, 1994). Procedures by which one can measure and adjust for tissue shrinkage have been described (Amann, 1970b; Berndtson, 1977; Zhengwei et al, 1990).
Assumption: Approaches Used to Estimate the Volume of Individual Nuclei Are Accurate— Estimates of average nuclear diameter are required to convert crude counts to true counts via the Abercrombie correction when assessing spermatogenesis via the method of Clermont and Morgentaler (1955), and when estimating the size of various cell populations via the volume density approach. Some factors that should be considered when estimating the volume of individual nuclei follow.
Calculating individual cell volumes from linear measurements. The oldest approach for estimating the volume of individual nuclei has been calculation from diameter or length x width measurements. Because of difficulties and potential inaccuracies when applied to structures with a nonspherical profile (eg, elongated spermatids) or irregular outline (eg, Sertoli cells), this approach is usually applied only to spherical nuclei. The linear dimensions (eg, diameter, length, width) of the nuclei are measured with an ocular micrometer (Berndtson, 1977). The resulting values are entered into the equation for calculating the volume of a sphere (Swierstra, 1966).
Whereas the nuclei of many germ cells appear almost perfectly spherical, the spermatogonia of many mammals are ellipsoid. It has been customary to average the length and width measurements of such nuclei to estimate their diameter (eg, Swierstra, 1966; Barr et al, 1971; Lino, 1971; Mori and Christensen, 1980), although the accuracy of this approach appears to have remained relatively untested. One advantage of linear measurements is simplicity; ocular micrometers are relatively inexpensive and the actual measurements can be performed quite rapidly at the light microscopic level. Estimates based on this approach are probably quite accurate if applied to spherical nuclei.
Estimating the volume of irregularly shaped nuclei via reconstruction of serial sections. The developing spermatids of most mammals become elongated and flattened, and Sertoli cell nuclei have an irregular outline with numerous invaginations. The volume of individual nuclei can be estimated via reconstruction of serial sections (eg, Sinha Hikim et al, 1988). Johnson et al (1984b) used this approach to determine human Sertoli cell nuclear volumes. They prepared 0.5-µm serial sections and measured the area occupied by the Sertoli nucleus by tracing its outline on a computerized digitizing tablet. The volume within each section was determined as the product of its area times the section thickness, and these were summed across all serial sections to obtain the total nuclear volume. This approach has considerable merit for some investigations. Disadvantages may include accessibility to the required equipment (electron microscope, computerized digitizing tablet, etc) and labor intensity, a factor likely to confine measurements to a small number of animals and (or) individual cells per testis. Excellent precision and accuracy when measuring individual nuclei may not be meaningful if measurements are limited to a small and potentially unrepresentative sample.
Because the approach is laborious, Johnson and Nguyen (1986) compared the volume of equine Sertoli nuclei estimated from reconstruction of serial sections to that calculated from linear measurements. For the latter, nuclear volume was calculated via the equation for a sphere, for which nuclear diameter was estimated as the average of the nuclear height x width measurement. In this study, nuclear volume was grossly overestimated via the linear measurement approach; it was necessary to multiply these estimates by a correction factor of 0.73 to render values equivalent to those obtained by serial reconstruction. These findings confirm the dangers inherent to estimating the volume of irregular, nonspherical Sertoli nuclei from linear measurements entered into the equation for the volume of a sphere.
Estimating nuclear volume via the point-sampled intercept approach. Zhengwei et al (1990) applied a point-sampled intercept approach of Gunderson and Jensen (1985) to determine the volume of irregularly shaped Sertoli nuclei in the rat. Fixed tissues were placed on the microscope and photomicrographs were taken of the 2 seminiferous tubules closest to the center of the field. These were printed at high magnification and used for point sampling within 1.2- or 1.5-cm grids for young and older animals, respectively. Between 80 and 100 intercepts were assessed per rat, and these were used to provide estimates of the volume fraction of Sertoli cell nuclei within seminiferous tubules, the volume fraction of seminiferous tubules within the testis, and the mean volume of the Sertoli cell nuclei, which in turn were used to calculate the number of Sertoli cells per testis. Some aspects of the point-sampled intercept approach bear a resemblance to the determination of volume density via the recording of random hits, and measurement and correction for tissue shrinkage are important with both approaches. Similarly, differential shrinkage of various testicular components would also impact the accuracy of data obtained by this method. The point-sampled intercept approach differs, however, in the fact that Sertoli nuclear volume is estimated from point sampling rather than via the reconstruction of serial sections or other approaches described previously. Thus, this approach may provide an attractive alternative or supplemental method for assessing the volume of irregularly shaped nuclei.
Changes in nuclear volume during spermatogenesis. Some germ cells persist over several stages of the cycle of the seminiferous epithelium, during which time their nuclear volume can change dramatically. For example, pachytene primary spermatocytes are present within 11 of the 12 stages of the cycle of the seminiferous epithelium of the bull (Berndtson and Desjardins, 1974), during which time they undergo a marked increase in size. It is customary to select round seminiferous tubular cross sections at a particular stage of the cycle for evaluation when using the method of Clermont and Morgentaler (1955). The changes in nuclear size due to progressive development from the beginning to end of a single stage are likely to be relatively minimal and of limited consequence with this evaluation method. In contrast, staging is not applied or possible when determining volume density by the Chalkley (1943) method, and some classes of germ cells (eg, young vs old primary spermatocytes) might encompass a longer period of development. Thus, it is important that the nuclei chosen for individual measurement accurately reflect the corresponding population identified in estimating volume density.
Nuclear dimensions can be obtained via the optical dissector method, and the use of periodic acid–Schiff staining permits identification of spermatids at different steps of spermiogenesis with this method (Meachem et al, 1999). However, because one examines tissue within a defined grid rather than within round tubular cross sections, staging is not normally applied with this method. Thus, the cells on which measurements are taken may include ones present across several stages of the cycle of the seminiferous epithelium. For example, Wreford (1995) reported numbers for cell groupings consisting of stage I–VIII and IX–XIV pachytene spermatocytes, round spermatids at stages I–VIII, and elongated spermatids at steps 15–19 of spermiogenesis in the rat. Similarly, with the preparations they used, Zhengwei et al (1990) grouped round spermatids in steps 1–7 of spermiogenesis into a single category and they reported difficulty in assigning primary spermatocytes to the leptotene, zygotene, and pachytene subclasses. In those instances in which cell nuclei change over time (eg, pachytene primary spermatocytes), determinations of nuclear volume based on a sufficient level of sampling would be expected to yield an accurate estimate of the average volume of the nuclei of cells within that grouping. In contrast, limited sampling could result in the evaluation of a disproportionate number of younger, smaller nuclei vs older, larger nuclei, and an inaccurate estimate of the actual mean nuclear volume for the cells within the intended grouping. Thus, the evaluation of a sufficient volume of material to ensure representative sampling of the cells of interest would be important with this method.
Assumption: Treatment Does Not Alter the Resistance of Germ Cells to Homogenization— The quantification of treatment effects on cell numbers via their enumeration in homogenates of fresh, frozen, or fixed tissues requires an assumption that the resistance of the nuclei of these cells to homogenization is not altered by treatment. To the author's knowledge, experimental alteration of nuclear resistance to homogenization has not been reported. This assumption is cited, nonetheless, because its failure could have a profound effect on the reliability of data derived by the homogenization approach. This assumption is not necessary for any of the other methods considered within this review.
Assumption: A Small Increase in Cellular Degeneration Should Cause a Several-Fold Increase in the Number of Degenerating Cells— Russell (1983) expressed the opinion that "A more sensitive method for determining cell loss [than germ cell:germ cell or germ cell:Sertoli cell ratios] is by identification and quantitation of degenerating cells." This view seemed predicated, in part, on an assumption that a small increase in cell degeneration would cause a several-fold increase in the number of degenerating germ cells. However, for numbers of degenerating germ cells to be meaningful in a quantitative sense, degeneration must not cause either a rapid, premature disappearance of particular cells or their persistence beyond the normal period during which they would otherwise remain present. Huckins (1978) conducted an extensive study of spermatogonial degeneration in adult male rats and reported that the process by which spermatogonia degenerate and disappear from the seminiferous epithelium was "both prolonged and variable." She concluded that "It is not known how long it takes for spermatogonia to degenerate, or how long recognizable necrotic figures persist in the epithelium. This precludes direct quantitation of degenerating cells as a useful measurement." In addition to those limitations, the inherent variability associated with the number of degenerating cells in untreated subjects is quite large, with standard deviations often exceeding 100% of the mean (Berndtson, 2008). As a result of this variability, most investigators will find the number of replicates needed to provide acceptable power and sensitivity for detecting actual treatment effects on sperm production by this approach to be prohibitive (Berndtson, 2008).
Summary, Conclusions, and Recommendations
Each approach for assessing treatment effects on sperm production entails
technical assumptions, none of which should be accepted as infallible. Indeed,
circumstances under which most of these assumptions have failed have been
presented. In the absence of better alternatives, investigators should
consider these assumptions carefully with the objective of providing
reasonable safeguards to prevent misleading or erroneous experimental results.
One way to provide a safeguard would be by assessing treatment effects on
sperm production by each of 2 or more methods that do not share a similar
assumption. Assumptions of relevance and importance with each of the methods
of evaluation that have been considered herein have been summarized in the
Table. To illustrate how this
information might be used, one may recall that the numerical stability of
Sertoli cells in adult animals is an important assumption when sperm
production is assessed by determining the number of germ cells per
seminiferous tubular cross section. However, this assumption is irrelevant
during assessments of sperm production via either the volume density or
homogenization approaches. Indeed, because the latter methods share very few
assumptions with the former method
(Table), either of these would
be quite complementary to the former, and a finding of similar results with
both methods within a given experiment would provide considerable reassurance
that one's results were accurate. In contrast, and despite the fact that they
entail fundamentally different approaches for assessing sperm production, the
volume density and homogenization approaches share the very important
assumption of constancy in the timing of spermatogenic events
(Table). Hence, results are not
mutually reinforcing. If the shared assumptions of constancy in the timing of
spermatogenic events and the frequency of specific stages of the cycle of the
seminiferous epithelium were to fail, this could potentially lead to similar
but inaccurate estimates of DSP with both methods. Through an awareness of the
most important technical assumptions and the judicious application of
complementary evaluation methods in well-replicated experiments of appropriate
design, one should be able to assess treatment effects on spermatogenesis with
great accuracy and confidence.
|
Footnotes
This is Scientific Contribution 2394 from the New Hampshire Agricultural Experiment Station.
References
Abercrombie M. The estimation of nuclear population from microtome sections. Anat Rec. 1946; 94: 239 –247.[CrossRef][Medline]
Amann RP. Reproductive capacity of dairy bulls. III. The effect of ejaculation frequency, unilateral vasectomy, and age on spermatogenesis. Am J Anat. 1962; 110: 49 –67.[CrossRef][Medline]
Amann RP. The male rabbit. IV. Quantitative testicular histology and comparisons between daily sperm production as determined histologically and daily sperm output. Fertil Steril. 1970a; 21: 662 –672.[Medline]
Amann RP. Sperm production rates. In: Johnson AD, Gomes WR, VanDemark NL, eds. The Testis. Vol 1 . New York, NY: Academic Press; 1970b; 433 –482.
Amann RP. The cycle of the seminiferous epithelium in humans: A
need to revisit? J Androl. 2008; 29: 1
–31.
Amann RP, Almquist JO. Reproductive capacity of dairy bulls. I. Technique for direct measurement of gonadal and extra-gonadal sperm reserves. J Dairy Sci. 1961; 44: 1537 –1543.
Amann RP, Almquist JO. Reproductive capacity of dairy bulls. VIII. Direct and indirect measurement of testicular sperm production. J Dairy Sci. 1962;45: 774 –781.
Amann RP, Lambiase JT Jr. The male rabbit. III. Determination of
daily sperm production by means of testicular homogenates. J Anim
Sci. 1969;28: 369
–374.
Attal J, Courot M. Development testiculaire et establissement de la spermatogenese chez le taureau. Ann Biol Anim Biochim Biophys. 1963;3: 219 –241.[CrossRef]
Barr AB, Moore DJ, Paulsen CA. Germinal cell loss during human
spermatogenesis. J Reprod Fertil. 1971; 25: 75
–80.
Bartlett JMS, Weinbauer GF, Nieschlag E. Stability of spermatogenic synchronization achieved by depletion and restoration of vitamin A in rats. Biol Reprod. 1990; 42: 603 –612.[Abstract]
Bergh A. Morphological signs of a direct effect of experimental cryptorchidism on the Sertoli cells in rats irradiated as fetuses. Biol Reprod. 1981; 24: 145 –152.[Abstract]
Berndtson WE. Methods for quantifying mammalian spermatogenesis: a
review. J Anim Sci. 1977; 44: 818
–833.
Berndtson WE. Sampling intensities and replication requirements for
detection of treatment effects on testicular function in bulls and stallions:
a statistical assessment. J Anim Sci. 1989; 67: 213
–225.
Berndtson WE. Replication requirements and number of ejaculates
needed for assessing treatment effects on sperm output and seminal
characteristics of electroejaculated Holstein bulls. J Anim
Sci. 1990;68: 709
–718.
Berndtson WE. A simple, rapid and reliable method for selecting or
assessing the number of replicates for animal experiments. J Anim
Sci. 1991;69: 67
–76.
Berndtson WE. Comparative reliability and sensitivity of different methods for assessing treatment effects on sperm production. Anim Reprod Sci. 2008;105: 5 –22.[Medline]
Berndtson WE, Desjardins C. The cycle of the seminiferous epithelium and spermatogenesis in the bovine testis. Am J Anat. 1974;140: 167 –180.[CrossRef][Medline]
Berndtson WE, Foote RH. Disruption of spermatogenesis in rabbits consuming ethylene glycol monomethyl ether. Reprod Toxicol. 1997;11: 29 –36.[CrossRef][Medline]
Berndtson WE, Neefus C, Foote RH, Amann RP. Optimal replication for histometric analyses of testicular function in rats and rabbits. Fundam Appl Toxicol. 1989; 12: 291 –302.[CrossRef][Medline]
Bertram JF, Sampson PD, Bolender RP. Influence of tissue composition on the final volume of rat liver blocks prepared for electron microscopy. J Electron Micros Tech. 1986; 4: 303 –314.[CrossRef]
Chalkley HW. Method for the quantitative morphologic analysis of
tissues. Natl Cancer Inst. 1943; 4: 47
–53.
Clegg EJ. Studies on artificial cryptorchidism: morphological and
quantitative changes in the Sertoli cells of the rat testis. J
Endocrinol. 1963;26: 567
–574.
Clermont Y. Kinetics of spermatogenesis in mammals: seminiferous
epithelium cycle and spermatogonial renewal. Physiol
Rev. 1972;52: 198
–236.
Clermont Y, Morgentaler H. Quantitative study of spermatogenesis in
the hypophysectomized rat. Endocrinology. 1955; 57: 369
–382.
Clermont Y, Perey B. Quantitative study of the cell population of the seminiferous tubules in immature rats. Am J Anat. 1957; 100: 241 –267.[CrossRef][Medline]
Clermont Y, Trott M. Duration of the cycle of the seminiferous epithelium in the mouse and hamster determined by means of 3H-thymidine and autoradiography. Fertil Steril. 1969;20: 805 –817.[Medline]
de Jong FH, Sharpe RM. The onset and establishment of
spermatogenesis in rats in relation to gonadotrophin and testosterone levels.
J Endocrinol. 1977; 75: 197
–207.
Elftman H. The Sertoli cell cycle in the mouse. Anat Rec. 1950;106: 381 –393.[CrossRef][Medline]
Eschenbrenner AB, Miller E, Lorenz E. Quantitative histologic analysis of the effect of chronic wholebody irradiation with gamma rays on the spermatogenic elements and the interstitial tissue of the testes of mice. Natl Cancer Inst. 1948; 9: 133 –147.[Medline]
Gaytan F, Aguilar E. Quantitative analysis of Sertoli cells in
neonatally oestrogen-treated rats. J Reprod Fertil. 1987; 79: 589
–598.
Gebauer MR. Extra-gonadal transit time, sperm production and sperm reserves in the stallion [PhD thesis]. Fort Collins: Colorado State University; 1973.
Gebauer MR, Pickett BW, Swierstra EE. Reproductive physiology of
the stallion. II. Daily production and output of sperm. J Anim
Sci. 1974;39: 732
–736.
Gondos B, Berndtson WE. Postnatal and prepubertal development. In: Russell LD, Griswold MD, eds. The Sertoli Cell. Clearwater, FL: Cache River Press; 1993; 115 –154.
Gunderson HJG, Jensen EB. Stereological estimation of number and sizes of arbitrary particles using the dissector. J Microsc. 1985;138: 127 –142.[Medline]
Harleman JH, Nolte T. Testicular toxicity: regulatory
guidelines—the end of formaldehyde fixation? Toxicol
Pathol. 1997;25: 414
–417.
Heller CG, Clermont Y. Kinetics of the germinal epithelium in man. Recent Prog Horm Res. 1964; 20: 545 –571.[Medline]
Hess RA. Quantitative and qualitative characteristics of the stages and transitions in the cycle of the seminiferous epithelium: Light microscopic observations of perfusion-fixed and plastic-embedded testes. Biol Reprod. 1990;43: 525 –542.[Abstract]
Hess RA, Chen P. Computer tracking of germ cells in the cycle of the seminiferous epithelium and prediction of changes in cycle duration in animals commonly used in reproductive biology and toxicology. J Androl. 1992;12: 185 –190.
Hess RA, Schaeffer DJ, Eroschenko VP, Keen JE. Frequency of the stages in the cycle of the seminiferous epithelium in the rat. Biol Reprod. 1990;43: 517 –524.[Abstract]
Huckins C. The morphology and kinetics of spermatogonial degeneration in normal adult rats: an analysis using a simplified classification of the germinal epithelium. Anat Rec. 1978; 190: 905 –926.[CrossRef][Medline]
Johnson L. A new approach to quantification of Sertoli cells that avoids problems associated with the irregular nuclear surface. Anat Rec. 1986;214: 231 –237.[CrossRef][Medline]
Johnson L. A new approach to study the architectural arrangement of
spermatogenic stages revealed little evidence of a partial wave along the
length of human seminiferous tubules. J Androl. 1994; 15: 435
–441.
Johnson L, Lebovitz RM, Samson WK. Germ cell degeneration in normal and microwave-irradiated rats: potential sperm production rates at different developmental steps in spermatogenesis. Anat Rec. 1984a; 209: 501 –507.[CrossRef][Medline]
Johnson L, Neaves WB. Age-related changes in the Leydig cell population, seminiferous tubules, and sperm production in stallions. Biol Reprod. 1981; 24: 703 –712.[Abstract]
Johnson L, Nguyen HB. Annual cycle of the Sertoli cell population
in adult stallions. J Reprod Fertil. 1986; 76: 311
–316.
Johnson L, Petty CS, Neaves WB. A new approach to quantification of spermatogenesis and its application to germinal cell attrition during human spermatogenesis. Biol Reprod. 1981; 25: 217 –226.[Abstract]
Johnson L, Petty CS, Neaves WB. Further quantification of human spermatogenesis: germ cell loss during postprophase meiosis and its relationship to daily sperm production. Biol Reprod. 1983; 29: 207 –215.[Abstract]
Johnson L, Thompson DL Jr. Age-related and seasonal variation in the Sertoli cell population, daily sperm production and serum concentrations of follicle-stimulating hormone, luteinizing hormone and testosterone in stallions. Biol Reprod. 1983; 29: 777 –789.[Abstract]
Johnson L, Varner DD, Tatum ME, Scrutchfield WL. Season but not age affects Sertoli cell number in stallions. Biol Reprod. 1991; 45: 404 –410.[Abstract]
Johnson L, Zane RS, Petty CS, Neaves WB. Quantification of the human Sertoli cell population: its distribution, relation to germ cell numbers, and age-related decline. Biol Reprod. 1984b; 31: 785 –795.[Abstract]
Jones LS, Berndtson WE. A quantitative study of Sertoli cell and germ cell populations as related to sexual development and aging in the stallion. Biol Reprod. 1986; 35: 138 –148.[Abstract]
Lanning LL, Creasy DM, Chapin RE, Mann PC, Barlow NJ, Regan KS,
Goodman DG. Recommended approaches for the evaluation of testicular and
epididymal toxicity. Toxicol Pathol. 2002; 30: 507
–520.
Latendresse JR, Warbritton AR, Jonassen H, Creasy DM. Fixation of
testes and eyes using a modified Davidson's fluid: comparison with Bouin's
fluid and conventional Davidson's fluid. Toxicol
Pathol. 2002;30: 524
–533.
Leblond CP, Clermont Y. Definition of the stages of the cycle of the seminiferous epithelium in the rat. Ann N Y Acad Sci. 1952;55: 548 –573.[Medline]
Lino BF. Cell count correction factors for the quantitative histological analysis of the germinal epithelium of the ram. Anat Rec. 1971;170: 413 –420.[CrossRef][Medline]
McLachlan RI, Wreford NG, Meachem SJ, de Kretser DM, Robertson DM. Effects of testosterone on spermatogenic cell populations in the adult rat. Biol Reprod. 1994; 51: 945 –955.[Abstract]
Meachem SJ, McLachlan RI, de Kretser DM, Robertson DM, Wreford NG. Neonatal exposure of rats to recombinant follicle stimulating hormone increases adult Sertoli and spermatogenic cell numbers. Biol Reprod. 1996;54: 36 –44.[Abstract]
Meachem SJ, McLachlan RI, Stanton PG, Robertson DM, Wreford NG. FSH
immunoneutralization acutely impairs spermatogonial development in normal
adult rats. J Androl. 1999; 20: 756
–762.
Meachem SJ, Stanton PG, Schlatt S. Follicle-stimulating hormone
regulates both Sertoli cell and spermatogonial populations in the adult
photoinhibited Djungarian hamster testis. Biol Reprod. 2005; 72: 1187
–1193.
Morales C, Griswold MD. Retinol-induced stage synchronization in
seminiferous tubules of the rat. Endocrinology. 1987; 121: 432
–434.
Mori H, Christensen AK. Morphometric analysis of Leydig cells in
the normal rat testis. J Cell Biol. 1980; 84: 340
–354.
Mori H, Shimizu D, Fukunishi R, Christensen AK. Morphometric analysis of testicular Leydig cells in normal adult mice. Anat Rec. 1982;204: 333 –339.[CrossRef][Medline]
Nagy F. Cell division kinetics and DNA synthesis in the immature
Sertoli cells of the rat testis. J Reprod Fertil. 1972; 28: 389
–395.
Oakberg EF. Initial depletion and subsequent recovery of spermatogonia of the mouse after 20 R of gamma rays and 100, 300, and 600 R of X-rays. Radiat Res. 1959; 11: 700 –719.[CrossRef][Medline]
Orth JM. Proliferation of Sertoli cells in fetal and postnatal rats: a quantitative autoradiographic study. Anat Rec. 1982; 203: 485 –492.[CrossRef][Medline]
Russell LD. Normal testicular structure and methods of evaluation under experimental and disruptive conditions. In: Clarkson TW, Nordberg GF, Sager PR, eds. Reproductive and Developmental Toxicity of Metals. New York, NY: Plenum Press; 1983 .
Russell LD, Clermont Y. Degeneration of germ cells in normal, hypophysectomized and hormone treated hypophysectomized rats. Anat Rec. 1977;187: 347 –366.[CrossRef][Medline]
Russell LD, Malone JP, Karpas SL. Morphological pattern elicited by agents affecting spermatogenesis by disruption of its hormonal stimulation. Tissue Cell. 1981; 13: 369 –380.[CrossRef][Medline]
Russell LD, Peterson RN. Determination of the elongate
spermatid–Sertoli cell ratio in various mammals. J Reprod
Fertil. 1984;70: 635
–641.
Sinha Hikim AP, Bartke A, Russell LD. Morphometric studies on hamster testes in gonadally active and inactive states: light microscopic findings. Biol Reprod. 1988; 39: 1225 –1237.[Abstract]
Steinberger A, Steinberger E. Replication pattern of Sertoli cells in maturing rat testis in vivo and in organ culture. Biol Reprod. 1971;4: 84 –87.[Abstract]
Steinberger A, Steinberger E. The Sertoli cells. In: Johnson AD, Gomes WR, eds. The Testis. Vol IV . New York, NY: Academic Press; 1977 .
Swierstra EE. Structural composition of Shorthorn bull testes and daily spermatozoa production as determined by quantitative testicular histology. Can J Anim Sci. 1966; 46: 107 –119.
Van Beek MEAB, Meistrich ML. A method for quantifying synchrony in testes of rats treated with vitamin A deprivation and readministration. Biol Reprod. 1990; 42: 424 –431.[Abstract]
Van Pelt AMM, De Rooij DG. The origin of the synchronization of the seminiferous epithelium in vitamin A-deficient rats after vitamin A replacement. Biol Reprod. 1990; 42: 677 –682.[Abstract]
Wing T-Y, Christensen AK. Morphometric studies on rat seminiferous tubules. Am J Anat. 1982; 165: 13 –25.[CrossRef][Medline]
Wreford NG. Theory and practice of stereological techniques applied to the estimation of cell number and nuclear volume in the testis. Microsc Res Tech. 1995; 32: 423 –436.[CrossRef][Medline]
Zhengwei Y, McLachlan RI, Bremner WJ, Wreford NG. Quantitative
(stereological) study of normal spermatogenesis in the adult monkey
(Macaca fascicularis). J Androl. 1997; 18: 681
–687.
Zhengwei Y, Wreford NG, de Kretser DM. A quantitative study of spermatogenesis in the developing rat testis. Biol Reprod. 1990;43: 629 –635.[Abstract]
This article has been cited by other articles:
|
|
 |
|
  W. E. Berndtson Replication Needed to Distinguish Alterations in Cell Ratios, the Frequency of Individual Stages of the Cycle of the Seminiferous Epithelium, or the Appearance of Abnormalities in the Testes of Rodents, Rabbits, or Humans J Androl, November 1, 2010; 31(6): 593 - 606. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. E. Berndtson Replication Needed to Detect Alterations in the Composition of Rodent, Rabbit, or Human Testes via Volume Density Approaches J Androl, November 1, 2010; 31(6): 607 - 616. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
