Page 1 of 22 Reproduction Advance Publication first posted on 1 July 2014 as Manuscript REP-14-0279
1 1
Compensatory Meiosis Mechanisms in Human Spermatogenesis
2 3
1
4
Stammler, *1Lutz Konrad
Mareike Borgers,
1
Martin Wolter, 1Anna Hentrich,
2
Martin Bergmann, 1Angelika
5 6
Address: 1Department of Obstetrics and Gynecology, Medical Faculty, Feulgenstr.
7
12, 35392 Giessen, Germany; 2Institute of Veterinary-Anatomy, -Histology and -
8
Embryology, Franfurter Str. 98, 35392 Giessen, Germany
9 10
Short Title: Meiosis efficiency
11 12 13
*Corresponding author, Proof corrections, Reprint request
14
Assistant Professor L. Konrad, PhD
15
Medical Faculty, Department of Obstetrics and Gynecology
16
Feulgenstr. 12
17
D-35392 Giessen, Germany
18
Fax: ..49-(641)-985-45258
19
Phone: ..49-(641)-985-45282
20
e-mail: [email protected]
21 22 23
1 Copyright © 2014 by the Society for Reproduction and Fertility.
Page 2 of 22
2 24
Abstract
25
Disturbances of checkpoints in distinct stages of spermatogenesis (mitosis, meiosis,
26
spermiogenesis) contribute to impaired spermatogenesis, however, the efficiency of
27
entry into meiosis has not been investigated in more detail. In this study, we analyzed
28
azoospermic patients with defined spermatogenic defects by the use of OCT2 for
29
type A spermatogonia, Sarcoma antigen 1 (SAGE1) for mitosis-meiosis transition
30
and Smad3 for pachytene spermatocytes. Especially patients with maturation arrest
31
at the level of primary spermatocytes showed significantly reduced numbers of
32
spermatogonia compared to patients with histologically intact spermatogenesis or
33
patients with hypospermatogenesis. For a detailed individual classification of the
34
patients, we distinguished between “high efficiency of meiotic entry” (high numbers of
35
pachytene spermatocytes) and “low efficiency of meiotic entry” (low numbers of
36
pachytene spermatocytes). Only patients with histologically normal spermatogenesis
37
and patients with hypospermatogenesis showed normal numbers of spermatogonia
38
and a high efficieny of meiotic entry. Of note, only patients with histologically normal
39
spermatogenesis or patients with hypospermatogenesis could compensate low
40
numbers of spermatogonia with a high efficiency of meiotic entry. In contrast, patients
41
with maturation arrest always showed a low efficiency of meiotic entry. This is the
42
first report on patients with impaired spermatogenesis showing that half of the
43
patients with hypospermatogenesis but all patients with maturation arrest cannot
44
compensate reduced numbers in spermatogonia with a highly efficient meiosis. Thus,
45
we suggest that compensatory meiosis mechanisms in human spermatogenesis
46
exist.
47 48
Keywords: Male infertiliy, Spermatogonia, Mitosis-Meiosis-Transition, OCT2, SAGE1,
49
Smad3
2
Page 3 of 22
3 50
Introduction
51
In about half of infertile couples, male infertility plays a role and almost 30% are
52
caused by a male factor alone (Brugh & Lipshultz, 2004). Besides analysis of semen
53
parameters a histological evaluation is indicated especially to confirm obstructive
54
azoospermia in men with testes of normal size and normal levels of reproductive
55
hormones (Dohle et al., 2012).
56
The dynamics of spermatogenesis are crucial for elucidation of impaired
57
spermatogenesis which is caused by many factors like X-ray, chemotherapy,
58
infections, chromosomal defects, Diabetes mellitus or alcohol abusus (Cerilli et al.,
59
2010). Although a lot of aberrations during mitosis (Steger et al., 1998; Bar-Shira
60
Mymon et al., 2003) and meiosis (Gonsalves et al., 2004; Egozcue et al., 2005) have
61
been described, only recently it was shown that also reduced numbers of Sertoli cells
62
or spermatogonia (including the stem cells) contribute to impaired spermatogenesis
63
(Hentrich et al., 2011). However, in humans the stem cell niche is still poorly
64
characterized, but Adark as well as Apale spermatogonia seem to contain
65
subpopulations of spermatogonial stem cells (Dym et al., 2009). Of note, in primates
66
the Apale spermatogonia contain also few spermatogonial stem cells (Hermann et al.,
67
2010).
68
The testicular stem cells in rat and mouse are localized nearly exclusively in a stem
69
cell niche adjacent to the interstitium (Chiarini-Garcia et al., 2001, 2003). The stem
70
cell niche is characterized by a microenvironment which is influenced also by Leydig
71
cells and blood vessels (Shetty et al., 2007; Yoshida et al., 2007). Furthermore, also
72
Sertoli cells and peritubular cells contribute to the self renewal of the spermatogonial
73
stem cells (Dadoune, 2007; de Rooij, 2009).
74
Especially in patients with maturation arrest a significant reduction of dividing
75
spermatogonia during mitosis was observed (Steger et al., 1998; Bar-Shira Mymon et
3
Page 4 of 22
4 76
al., 2003), whereas in hypospermatogenesis accelerated apoptosis rather than
77
proliferative
78
spermatogonial cell numbers (Takagi et al., 2001). Additionally, in cases with
79
maturation arrest reduced recombination rates (up to 50%) were found (Gonsalves et
80
al., 2004; Egozcue et al., 2005) and an increased frequency of genetic abnormalities
81
(Weedin et al., 2011).
82
In this study, we analyzed germ cell numbers in the early phase of spermatogenesis
83
with an emphasis on mitosis-meiosis transition. Furthermore, we classified the
84
deficiencies individually by distinguishing between high and low efficiency of meiosis.
85
Our main finding in this study was that cases with histologically normal
86
spermatogenesis and cases with hypospermatogenesis can compensate reduced
87
numbers of spermatogonia with a highly efficient meiosis suggesting a compensatory
88
mechanism.
dysfunction
was
suggested
to
be
responsible
for
reduced
89
4
Page 5 of 22
5 90
Materials and Methods
91
Patients
92
Testicular biopsies were obtained from 90 patients between 1998 and 2012 and were
93
indicated because of normogonadotropic obstructive or non-obstructive azoospermia.
94
After written informed consent, biopsies were taken under general anesthesia. The
95
Ethics Committee of the Medical Faculty of the Justus-Liebig-University, Giessen,
96
Germany approved the study (75/00 and 56/05). After fixation in Bouins solution and
97
embedding in paraffin, 5 µm sections were stained with hematoxylin and eosin and
98
spermatogenesis was evaluated histologically according to the scoring system of
99
Bergmann & Kliesch (1998, 2010). The patients were classified into histologically
100
normal spermatogenesis (Nsp, n=30; median age 39, range 16-56; mean score 9.93,
101
range 9-10), hypospermatogenesis (Hyp, n=30; median age 37, range 18-63; mean
102
score 8.43, range 6-10) and maturation arrest at the level of spermatocytes (MA,
103
n=30; median age 34.5, range 17-69; mean score 0).
104
Antibodies
105
To classify especially early stages of spermatogenesis (spermatogonia and mitosis-
106
meiosis-transition), we identified several highly robust markers with a good signal-to-
107
noise ratio and distinct expression pattern in human germ cells. For a detailed
108
analysis of impaired spermatogenesis, three markers were chosen (Table 1), OCT2
109
(octamer binding protein2, POU2F2) for early spermatogonia (Lim et al., 2011),
110
SAGE1 (sarcoma antigen1) for dividing and differentiating spermatogonia (Lim et al.,
111
2011) and mitosis-meiosis-transition (Looijenga, 2011), and Smad3 for pachytene
112
spermatocytes (Hentrich et al., 2011).
113
Immunohistochemistry
114
Immunohistochemistry of 5 µm sections of bouin-fixed, paraffin-embedded specimen
115
was performed as published (Konrad et al., 2007). The Envision System from DAKO
5
Page 6 of 22
6 116
(Hamburg, Germany) combined with DAB staining was used according to the
117
manufacturer`s instructions. Counterstaining was done with hematoxylin. The
118
antibodies used for detection are shown in Table 1. Digital images were obtained
119
with the inverse microscope FSX100 (Olympus, Hamburg, Germany) using the
120
Olympus FSX-BSW software. Images were processed with Adobe Photoshop 7.0.
121
Quantification of germ cells
122
Numbers of germ cells were obtained from 26.63±12.58 tubular cross-sections per
123
specimen
124
maturation arrest for each staining. As published recently (Hentrich et al., 2011) and
125
also shown in this study, it was sufficient to use approximately 15-35 nearly round
126
cross-sections for quantification; we never observed outliers.
127
There is no consensus on how quantification of testicular biopsies with emphasis on
128
impaired spermatogenesis should be performed. The number of Sertoli cells is
129
different in patients with maturation arrest or hypospermatogenesis (Hentrich et al.,
130
2011). Thus, the Sertoli cell to germ cell ratio is not applicable for quantification of
131
germ cell numbers.
132
Statistical methods Values from each experiment were used for calculation of the
133
means (per cross-section) and the respective standard errors of the mean (SEM).
134
Differences between the groups were calculated with the test from Mann Whitney. P
135
values below 0.05 were considered statistically significant.
136
Cut-off values to distinguish between normal germ cell numbers versus reduced
137
germ cell numbers were determined with ROC-curves (receiver operating
138
characteristic) and AUC (area under the curve). Besides sensitivity and specificity
139
also the positive Likelihood-Ratio (LR+) was calculated. For all statistical analyses
140
GraphPad Prism6 was used.
representing
normal
spermatogenesis,
hypospermatogenesis,
and
141
6
Page 7 of 22
7 142
Results
143
Identification of germ cell markers for early spermatogenesis
144
We identified expression of OCT2 mainly in spermatogonia Adark but also in very few
145
spermatogonia Apale (Fig. 1A, D, G). Often the OCT2-positive spermatogonia were
146
localized in small groups in close proximity to interstitial arterioles or venules which is
147
in accordance to the characteristics of the stem cell niche in murine testis (Shetty et
148
al., 2007; Yoshida et al., 2007) and thus have been also termed reserve stem cells
149
(Dym et al., 2009). However, we found also OCT2-negative Adark and Apale
150
spermatogonia.
151
We identified SAGE1 in very few spermatogonia Adark but in nearly all Apale and type
152
B spermatogonia as well as in (pre)-leptotene spermatocytes and round spermatids
153
(Fig. 1B, E, H), thus representing mitotic spermatogonia and transition of mitosis into
154
meiosis. For characterization of pachytene spermatocytes, we used Smad3 (Fig. 1C,
155
F, I).
156
Analysis of meiosis efficiency
157
All three markers were used for quantification of germ cell numbers (early
158
spermatogenesis) per cross-section and thus also for individual classification of
159
meiosis efficiency in cases of impaired spermatogenesis. Quantification of OCT2-
160
positive spermatogonia showed nearly identical values for normal spermatogenesis
161
and hypospermatogenesis, but cases with maturation arrest demonstrated a
162
significant reduction of approximately 50% (Table 2). Of note, quantification of
163
SAGE1-positive germ cells (spermatogonia up to (pre)-leptotene spermatocytes)
164
revealed strongly and significantly reduced cell numbers in hypospermatogenesis
165
and in patients with maturation arrest compared to the controls with histologically
166
normal
spermatogenesis
(Table
2).
Similarly
Smad3-positive
pachytene
7
Page 8 of 22
8 167
spermatocytes were reduced by 47% in hypospermatogenesis and by 91% in cases
168
with maturation arrest compared to normal spermatogenesis (Table 2).
169
Individual classification of patients with impaired spermatogenesis
170
For a detailed individual classification of the patients, we determined the cut-off
171
values for all three markers to distinguish between normal and lower germ cell
172
numbers (Table 3). Furthermore, we classified spermatogenesis into “high efficiency
173
of meiotic entry” (high numbers of Smad3-positive spermatocytes, Fig. 2A) and “low
174
efficiency of meiotic entry” (low numbers of Smad3-positive spermatocytes, Fig. 2B).
175
Both groups could be further divided into four subgroups according to the cut-offs for
176
early spermatogonia and dividing/differentiating spermatogonia (Fig. 2). The
177
individual evaluation of each patient revealed that patients with histologically normal
178
spermatogenesis (n=29) or patients with hypospermatogenesis (n=14) showed a
179
high efficieny of meiotic entry and normal numbers of early spermatogonia or
180
dividing/differentiating spermatogonia (Fig. 2A). Remarkably, only patients with
181
histologically normal spermatogenesis (7 from 8) as well as patients with
182
hypospermatogenesis (7 from 14) could compensate low numbers of spermatogonia
183
with a high efficiency of meiotic entry (Fig. 2). In contrast, patients with maturation
184
arrest always showed a low efficiency of meiotic entry and also revealed more often
185
defects in numbers of dividing/differentiating spermatogonia compared to defects in
186
early
187
compensatory mechanisms in meiosis since patients with normal numbers of
188
spermatogonia showed low efficiency of meiosis (Fig. 2B).
spermatogonia
(Fig.
2B).
Especially
subgroup
5
indicates
missing
189 190 191
8
Page 9 of 22
9 192
Discussion
193
Besides the identification and evaluation of markers for early spermatogenesis, the
194
primary aim of this study was the individual classification of defects in the early phase
195
of spermatogenesis in cases of impaired spermatogenesis.
196
Intriguingly, expression of OCT2 was found primarily in clusters of human Adark
197
spermatogonia in a stem cell niche most often adjacent to the interstitial vasculature.
198
This suggests that OCT2 is expressed very early in human spermatogenesis most
199
likely in stem cells as described by Lim et al. (21). However, as also shown by Lim et
200
al. (2011) as well as in this study only a subgroup of Adark spermatogonia is OCT2-
201
positive. Identification of few OCT2-positive Apale spermatogonia in this study is in
202
accordance to results obtained in monkeys (Hermann et al., 2010) which showed that
203
this subgroup also consists of undifferentiated spermatogonia, possibly stem cells.
204
Thus, we believe like Dym et al. (2009) and (Hermann et al., 2010) that both Adark
205
and Apale spermatogonia contain spermatogonial stem cells. In contrast, Ehmcke et
206
al. (2006) described that only the Adark spermatogonia comprise spermatogonial stem
207
cells.
208
In line with our previous findings (Hentrich et al., 2011), we found reduced numbers
209
of spermatogonia in cases with maturation arrest. Moreover, we observed low
210
numbers of early spermatogonia in cases with hypospermatogenesis as well as in
211
maturation
212
spermatogenesis. Thus, without Adark spermatogonia which contribute to stem cell
213
renewal, spermatogenesis is severely impaired (Holstein, 1999).
214
In addition to stem cell markers, we were also interested in identification of marker(s)
215
for dividing and differentiating spermatogonia. We observed SAGE1 expression in
216
very few Adark spermatogonia, but mainly in dividing spermatogonia and (pre)-
217
leptotene spermatocytes indicating entry into meiosis. Similarly Lim et al. (2011) and
arrest,
indicating
deficiencies
already
in
the
early
phase
of
9
Page 10 of 22
10 218
Chen et al. (2011) found SAGE1 in dividing spermatogonia and meiotic entry.
219
Interestingly, SAGE1 expression was not detectable before puberty (Lim et al.,
220
2011), thus further corroborating the importance of SAGE1 for differentiation of
221
spermatogonia.
222
We found that SAGE1-positive spermatogonia were severly reduced in patients with
223
hypospermatogenesis and maturation arrest but only rarely in patients with
224
histologically normal spermatogenesis. Thus, defects in the early phase of
225
spermatogenesis,
226
spermatogenesis. Similarly Bar-Shira Maymon et al. (2003) found a significant
227
reduction of dividing spermatogonia in patients with spermatogenic defects which
228
was also described for hypospermatogenesis and patients with maturation arrest
229
(Steger et al., 1998). However, because reduced numbers of early spermatogonia
230
might contribute to reduced numbers of mitotic spermatogonia, thus, any study on
231
mitotic defects must also include quantification of early spermatogonia in order to
232
discriminate properly between mitotic defects or reduced numbers of spermatogonia
233
before meiosis or a combination of both.
234
Smad3, which is also a key mediator of activin-induced proliferation of Sertoli cells
235
(Itman et al., 2011), is mainly expressed in pachytene spermatocytes in rat (Xu et al.,
236
2003), mouse (Itman et al., 2011) and human testis (Hentrich et al., 2011). The
237
reduced Smad3-positive pachytene spermatocyte numbers in patients with
238
hypospermatogenesis or maturation arrest (Hentrich et al., 2011) was also described
239
without marker expression analysis in cases with reduced sperm cell numbers
240
(Roosen-Runge et al., 1957; Zukerman et al., 1978). In contrast to the retained
241
Smad3 protein expression in cases with impaired spermatogenesis, the BOULE
242
protein (expressed in human leptotene to pachytene spermatocytes in normal
243
spermatogenesis) was completely absent in cases with maturation arrest (Luetjens et
especially
during
mitosis,
contribute
also
to
impaired
10
Page 11 of 22
11 244
al., 2004). Furthermore, alternative splicing of the BOULE gene was supposed to
245
serve as a predictive marker for meiotic efficiency (Kostova et al., 2007).
246
The individual classification of patients with impaired spermatogenesis by
247
categorizing meiotic efficiency in the early phase of spermatogenesis is a promising
248
extension of the classification system of Bergmann & Kliesch (1998, 2010). Although
249
studies are lacking to assess the early phase of spermatogenesis up to pachytene,
250
earlier studies showed up to 50% reduced recombination rates in cases with
251
maturation arrest (Gonsalves et al., 2004; Egozcue et al., 2005). Similarily, a 6-fold
252
increased frequency of genetic abnormalities was observed in cases of maturation
253
arrest compared to hypospermatogenesis (Weedin et al., 2011).
254
In this study we found that patients with histologically normal spermatogenesis
255
showed the lowest and patients with maturation arrest the highest numbers of
256
defects in the early phase of spermatogenesis as well as in the meiotic phase.
257
Especially in the subgroups 2-4 (Fig. 3A) with a high meiotic efficiency but low
258
numbers of spermatogonia, meiotic compensatory mechanisms must exist to
259
ameliorate these defects. Obviously, all cases with maturation arrest are lacking this
260
mechanism in addition to severly reduced numbers of spermatogonia. Interestingly,
261
in cases with maturation arrest reduction of early spermatogonia numbers (54%) is
262
nearly similar to the reduced numbers of dividing/differentiating spermatogonia
263
(42%), suggesting that loss of early spermatogonia might be primarily causative.
264
Of note, in cases of hypospermatogenesis, which is the most heterogenous group
265
(Takagi et al., 2001), we could identify both groups of patients with high or low
266
efficicency of meiosis in almost all combinations with reduced numbers of early
267
spermatogonia or reduced numbers of dividing/differentiating spermatogonia.
268
However, in contrast to cases with maturation arrest, the frequency of low efficiency
269
of meiosis was clearly lower in cases of hypospermatogenesis.
11
Page 12 of 22
12 270
In summary, this study revealed efficiency of meiotic entry as a factor for male
271
infertility and allows for a more detailed analysis of individual cases. We provide a
272
refinement of the established classification system launched by Bergmann & Kliesch
273
(1998, 2010). In line with the diagnosis of maturation arrest, all cases demonstrated
274
reduced numbers of spermatocytes, however, the majority additionally experienced
275
reduced numbers of spermatogonia. Furthermore, we could provide evidence that in
276
cases with normal spermatogenesis as well as in cases with hypospermatogenesis
277
compensatory meiosis mechanisms might exist to attenuate loss of spermatogonia.
278 279 280
Acknowledgments We thank Cornelia Hof for excellent technical assistance and
281
Prof. Kliesch (Center of Andrology and Reproductive Medicine, University of
282
Muenster, Germany) for biopsy sampling. This work was supported by a grant of
283
the Deutsche Forschungsgemeinschaft to L. Konrad (KFO 181/2 P2).
284 285
Disclosure/Conflict of interest
286
The authors declare no conflict of interest.
287
12
Page 13 of 22
13 288
References
289
Bar-Shira Maymon B, Yogev L, Yavetz H, Lifschitz-Mercer B, Schreiber L,
290
Kleiman SE, Botchan A &, Hauser R, Paz G 2003 Spermatogonial proliferation
291
patterns in men with azoospermia of different etiologies. Fertil Steril 80 1175–
292
1180.
293 294
Bergmann M & Kliesch S 1998 Hodenbiopsie. In Andrologie pp 66-71. Eds. W Krause and W Weidner. Stuttgart, Germany, Enke-Verlag.
295
Bergmann M & Kliesch S 2010 Testicular biopsy and histology. In Andrology, edn
296
3, pp 155-168. Eds E Nieschlag, HM Behre, S Nieschlag. Berlin Heidelberg,
297
Germany, Springer-Verlag.
298 299 300 301
Brugh VM 3rd & Lipshultz LL 2004 Male factor infertility: evaluation and management. Med Clin North Am 88 367-385. Cerilli LA, Kuang W & Rogers D 2010 A practical approach to testicular biopsy interpretation for male infertility. Arch Pathol Lab Med 134 1197–1204.
302
Chen YT, Chiu R, Lee P, Beneck D, Jin B & Old LJ 2011 Chromosome X-encoded
303
cancer/testis antigens show distinctive expression patterns in developing gonads
304
and in testicular seminoma. Hum Reprod 26 3232–3243.
305
Chiarini-Garcia H, Hornick JR, Griswold MD & Russell LD 2001 Distribution of
306
type A spermatogonia in the mouse is not random. Biol Reprod 65 1179–1185.
307
Chiarini-Garcia H, Raymer AM & Russell LD 2003 Non-random distribution of
308
spermatogonia in rats: evidence of niches in the seminiferous tubules.
309
Reproduction 126 669–680.
310 311 312 313
Dadoune JP 2007 New insights into male gametogenesis: what about the spermatogonial stem cell niche? Folia Histochem Cytobiol 45 141-147. de Rooij DG 2009 The spermatogonial stem cell niche. Microsc Res Tech 72 580– 585.
13
Page 14 of 22
14 314 315 316 317
Dohle GR, Elzanaty S & van Casteren NJ 2012 Testicular biopsy: clinical practice and interpretation. Asian J Androl 14 88-93. Dym M, Kokkinaki M & He Z 2009 Spermatogonial stem cells: mouse and human comparisons. Birth Defects Res C Embryo Today 87 27–34.
318
Egozcue J, Sarrate Z, Codina-Pascual M, Egozcue S, Oliver-Bonet M, Blanco J,
319
Navarro J, Benet J & Vidal F 2005 Meiotic abnormalities in infertile males.
320
Cytogenet Genome Res 111 337–342.
321 322
Ehmcke J, Wistuba J & Schlatt S 2006 Spermatogonial stem cells: questions, models and perspectives. Human Reprod Update 12 275–282.
323
Gonsalves J, Sun F, Schlegel PN, PJ, Hopps CV, Greene C, Martin RH & Reijo
324
Pera RA 2004 Defective recombination in infertile men. Hum Mol Genet 13 2875–
325
2883.
326
Hentrich A, Wolter M, Szardening-Kirchner C, Lüers GH, Bergmann M, Kliesch
327
S & Konrad L 2011 Reduced numbers of Sertoli, germ, and spermatogonial stem
328
cells in impaired spermatogenesis. Modern Pathol 24 1380-1389.
329
Hermann BP, Sukhwani M, Hansel MC & Orwig KE 2010 Spermatogonial stem
330
cells in higher primates: are there differences from those in rodents?
331
Reproduction 139 479–493.
332 333
Holstein AF 1999 Human spermatogenesis: basic research and clinical issues. Ann Anat 181 427-436.
334
Itman C, Wong C, Hunyadi B, Ernst M, Jans DA & Loveland KL 2011 Smad3
335
dosage determines androgen responsiveness and sets the pace of postnatal
336
testis development. Endocrinology 152 2076–2089.
337
Konrad L, Scheiber JA, Völck-Badouin E, Keilani MM, Laible L, Brandt H,
338
Schmidt A, Aumüller G & Hofmann R 2007 Alternative splicing of TGF-betas
14
Page 15 of 22
15 339
and their high-affinity receptors TβRI, TβRII and TβRIII (Betaglycan) reveal new
340
variants in human prostatic cells. BMC Genomics 8 318.
341
Kostova E, Yeung CH, Luetjens CM, Brune M, Nieschlag E & Gromoll J 2007
342
Association of three isoforms of the meiotic BOULE gene with spermatogenic
343
failure in infertile men. Mol Hum Reprod 13 85-93.
344
Lim J, Goriely A, Turner GD, Ewen KA, Jacobsen GK, Graem N, Wilkie AO &
345
Rajpert-De Meyts E 2011 OCT2, SSX and SAGE1 reveal the phenotypic
346
heterogeneity of spermatocytic seminoma reflecting distinct subpopulations of
347
spermatogonia. J Pathol 224 473–483.
348 349
Looijenga LH 2011 Spermatocytic seminoma: toward further understanding of pathogenesis. J Pathol 224 431-433.
350
Luetjens CM, Xu EY, Rejo Pera RA, Kamischke A, Nieschlag E & Gromoll J 2004
351
Association of meiotic arrest with lack of BOULE protein expression in infertile
352
men. J Clin Endocrinol Metab 89 1926-1933.
353 354
Shetty G & Meistrich ML 2007 The missing niche for spermatogonial stem cells: do blood vessels point the way? Cell Stem Cell 1 361-363.
355
Steger K, Aleithe I, Behre H & Bergmann M 1998 The proliferation of
356
spermatogonia in normal and pathological human seminiferous epithelium: an
357
immunohistochemical study using monoclonal antibodies against Ki-67 protein
358
and proliferating cell nuclear antigen. Mol Hum Reprod 4 227–233.
359
Roosen-Runge EC, Marberger E & Nelson WO 1957 Quantitative investigations on
360
human testicular biopsies. II. Infertility and other conditions. Fertil Steril 8 203–
361
219.
362
Takagi S, Itoh N, Kimura M, Sasao T & Tsukamoto T 2001 Spermatogonial
363
proliferation
and
apoptosis
in
hypospermatogenesis
364
nonobstructive azoospermia. Fertil Steril 76 901–907.
associated
with
15
Page 16 of 22
16 365
Weedin JW, Bennett RC, Fenig DM, Lamb DJ & Lipshultz LI 2011 Early versus
366
late maturation arrest: reproductive outcomes of testicular failure. J Urol 186 621-
367
626.
368 369 370 371
Xu J, Beyer AR, Walker WH & McGee EA 2003 Developmental and stage-specific expression of Smad2 and Smad3 in rat testis. J Androl 24 192–200. Yoshida S, Sukeno M & Nabeshima Y 2007 A vasculature-associated niche for undifferentiated spermatogonia in the mouse testis. Science 317 1722–1726.
372
Zukerman Z, Rodriguez-Rigau LJ, Weiss DB, Chowdhury AK, Smith KD &
373
Steinberger E 1978 Quantitative analysis of the seminiferous epithelium in
374
human testicular biopsies, and the relation of spermatogenesis to sperm density.
375
Fertil Steril 30 448–455.
376 377
16
Page 17 of 22
17 378
Figure Legends
379
Figure 1 Immunohistochemical detection of OCT2 (A, D, G), SAGE1 (B, E, H), and
380
Smad3 (C, F, I) in human biopsies from patients with histologically normal
381
spermatogenesis (A, B, C), patients with hypospermatogenesis (D, E, F) and cases
382
with maturation arrest (G, H, I). OCT2 was found mainly in Adark spermatogonia (Ad)
383
and to a lesser extent in Apale spermatogonia (Ap). Also few OCT2-negative Adark
384
spermatogonia [Ad(-)] could be identified. SAGE1 was expressed in Adark
385
spermatogonia up to type B spermatogonia (B), but also in some (pre)-leptotene
386
spermatocytes (pl). In contrast, Smad3 was localized primarily in the cytoplasm of
387
pachytene spermatocytes (p). The arrows point to the regions given enlarged in the
388
insets. scale bar – 25 µm (A-C), the same magnification was used in (D-I).
389
Counterstaining was done with hematoxylin.
390 391
Figure 2 Classification of patients into high meiosis efficiency (A) and low meiosis
392
efficiency (B). Both groups consist of four subgroups with normal numbers of early
393
spermatogonia (OCT2+) and dividing/differentiating spermatogonia (SAGE1+) or
394
reduced numbers of early spermatogonia (OCT2-) and dividing/differentiating
395
spermatogonia (SAGE1-). The cut-offs from Table 3 were used for classification.
396
Only cases with normal spermatogenesis (Nsp) or hypospermatogenesis (Hyp)
397
demonstrated high meiosis efficiency (A). Interestingly, only Nsp or Hyp could
398
compensate
399
dividing/differentiating spermatogonia (SAGE1-) with a highly efficient meiosis (A). In
400
contrast all cases with maturation arrest (MA) showed low meiosis efficiency besides
401
defects in numbers of early spermatogonia or dividing/differentiating spermatogonia
402
(B).
low
numbers
of
early
spermatogonia
(OCT2-)
and/or
17
Page 18 of 22
A
B
C
D
E
F
G
H
I
Figure 1
Page 19 of 22
HighGmeiosisGefficiency
Number of cases
A
SubgroupGGGGGGG1GGGGGGGGGGGGGGGGG2GGGGGGGGGGGGGGGGGG3GGGGGGGGGGGGGGGG4 GGGGGGOCT2GGGGGGG+GGGGGGGGGGGGGGGGG-GGGGGGGGGGGGGGGGGGG+GGGGGGGGGGGGGGGGGGGGSAGE1GGGGGGG+GGGGGGGGGGGGGGGG+GGGGGGGGGGGGGGGGGGG-GGGGGGGGGGGGGGGG-
B
Number of cases
LowGmeiosisGefficiency
SubgroupGGGGGGG5GGGGGGGGGGGGGGGGGGGG6GGGGGGGGGGGGGGGGGGG7GGGGGGGGGGGGGGGGGGGG8 GGGGGGOCT2GGGGGGG+GGGGGGGGGGGGGGGGGGGG-GGGGGGGGGGGGGGGGGGGG+GGGGGGGGGGGGGGGGGGGGGG GGGSAGE1GGGGGGGG+GGGGGGGGGGGGGGGGGGG+GGGGGGGGGGGGGGGGGGGG-GGGGGGGGGGGGGGGGGGGG-
Page 20 of 22
Table 1 Antibodies used for quantification of distinct germ cell types Protein
Source
Cat-No
Species Clonality
OCT2
Novocastra, Wetzlar,
NCL-OCT-207
Mouse
monoclonal 1:75
Rabbit
polyclonal
1:2000
Rabbit
polyclonal
1:300
Dilution
Germany
SAGE1
Sigma-Aldrich, Munich, HPA003033 Germany
Smad3
EnoGene Biotech, Frankfurt, Germany
Cat-No, catalog number
E021324
Page 21 of 22
Table 2 Numbers of distinct germ cells per cross-section in impaired spermatogenesis OCT2
SAGE1
Early spermatogonia
Mean SEM n P-values Nsp vs Hyp Nsp vs MA Hyp vs MA
Smad3
Dividing/differentiating spermatogonia
Pachytene spermatocytes
Nsp
Hyp
MA
Nsp
Hyp
MA
Nsp
Hyp
MA
2.09 0.25 30
2.05 0.23 30
0.96 0.16 30
15.62 0.82 30
11.16 0.61 30
9.10 0.77 30
16.35 0.92 30
8.64 0.56 30
1.53 0.36 30
ns 0.01 0.0001
0.01 0.001 ns
0.001 0.001 0.001
Nsp, normal spermatogenesis; Hyp, hypospermatogenesis; MA, maturation arrest; ns, not significant; vs, versus
Page 22 of 22
1 1
Table 3 Statistical analysis of the three markers used for individual classification OCT2
SAGE1
Smad3
Cut-Off = 1.06
Cut-Off = 10.84
Cut-Off = 8.6
sens
spec
LR+
sens
spec
LR+
sens
spec
LR+
Nsp/Hyp
23
87
1.8
40
90
4.0
47
97
14
Nsp/MA
57
87
4.3
77
90
7.7
100
97
30
Hyp/MA
57
77
2.4
77
60
1.9
100
53
2.1
2
Values for sensitivity (sens) and specificity (spec) are given in %; LR+, positive
3
likelihood ratio; Nsp, normal spermatogenesis; Hyp, hypospermatogenesis; MA,
4
maturation arrest
5
1
1 1
Compensatory Meiosis Mechanisms in Human Spermatogenesis
2 3
1
4
Stammler, *1Lutz Konrad
Mareike Borgers,
1
Martin Wolter, 1Anna Hentrich,
2
Martin Bergmann, 1Angelika
5 6
Address: 1Department of Obstetrics and Gynecology, Medical Faculty, Feulgenstr.
7
12, 35392 Giessen, Germany; 2Institute of Veterinary-Anatomy, -Histology and -
8
Embryology, Franfurter Str. 98, 35392 Giessen, Germany
9 10
Short Title: Meiosis efficiency
11 12 13
*Corresponding author, Proof corrections, Reprint request
14
Assistant Professor L. Konrad, PhD
15
Medical Faculty, Department of Obstetrics and Gynecology
16
Feulgenstr. 12
17
D-35392 Giessen, Germany
18
Fax: ..49-(641)-985-45258
19
Phone: ..49-(641)-985-45282
20
e-mail: [email protected]
21 22 23
1 Copyright © 2014 by the Society for Reproduction and Fertility.
Page 2 of 22
2 24
Abstract
25
Disturbances of checkpoints in distinct stages of spermatogenesis (mitosis, meiosis,
26
spermiogenesis) contribute to impaired spermatogenesis, however, the efficiency of
27
entry into meiosis has not been investigated in more detail. In this study, we analyzed
28
azoospermic patients with defined spermatogenic defects by the use of OCT2 for
29
type A spermatogonia, Sarcoma antigen 1 (SAGE1) for mitosis-meiosis transition
30
and Smad3 for pachytene spermatocytes. Especially patients with maturation arrest
31
at the level of primary spermatocytes showed significantly reduced numbers of
32
spermatogonia compared to patients with histologically intact spermatogenesis or
33
patients with hypospermatogenesis. For a detailed individual classification of the
34
patients, we distinguished between “high efficiency of meiotic entry” (high numbers of
35
pachytene spermatocytes) and “low efficiency of meiotic entry” (low numbers of
36
pachytene spermatocytes). Only patients with histologically normal spermatogenesis
37
and patients with hypospermatogenesis showed normal numbers of spermatogonia
38
and a high efficieny of meiotic entry. Of note, only patients with histologically normal
39
spermatogenesis or patients with hypospermatogenesis could compensate low
40
numbers of spermatogonia with a high efficiency of meiotic entry. In contrast, patients
41
with maturation arrest always showed a low efficiency of meiotic entry. This is the
42
first report on patients with impaired spermatogenesis showing that half of the
43
patients with hypospermatogenesis but all patients with maturation arrest cannot
44
compensate reduced numbers in spermatogonia with a highly efficient meiosis. Thus,
45
we suggest that compensatory meiosis mechanisms in human spermatogenesis
46
exist.
47 48
Keywords: Male infertiliy, Spermatogonia, Mitosis-Meiosis-Transition, OCT2, SAGE1,
49
Smad3
2
Page 3 of 22
3 50
Introduction
51
In about half of infertile couples, male infertility plays a role and almost 30% are
52
caused by a male factor alone (Brugh & Lipshultz, 2004). Besides analysis of semen
53
parameters a histological evaluation is indicated especially to confirm obstructive
54
azoospermia in men with testes of normal size and normal levels of reproductive
55
hormones (Dohle et al., 2012).
56
The dynamics of spermatogenesis are crucial for elucidation of impaired
57
spermatogenesis which is caused by many factors like X-ray, chemotherapy,
58
infections, chromosomal defects, Diabetes mellitus or alcohol abusus (Cerilli et al.,
59
2010). Although a lot of aberrations during mitosis (Steger et al., 1998; Bar-Shira
60
Mymon et al., 2003) and meiosis (Gonsalves et al., 2004; Egozcue et al., 2005) have
61
been described, only recently it was shown that also reduced numbers of Sertoli cells
62
or spermatogonia (including the stem cells) contribute to impaired spermatogenesis
63
(Hentrich et al., 2011). However, in humans the stem cell niche is still poorly
64
characterized, but Adark as well as Apale spermatogonia seem to contain
65
subpopulations of spermatogonial stem cells (Dym et al., 2009). Of note, in primates
66
the Apale spermatogonia contain also few spermatogonial stem cells (Hermann et al.,
67
2010).
68
The testicular stem cells in rat and mouse are localized nearly exclusively in a stem
69
cell niche adjacent to the interstitium (Chiarini-Garcia et al., 2001, 2003). The stem
70
cell niche is characterized by a microenvironment which is influenced also by Leydig
71
cells and blood vessels (Shetty et al., 2007; Yoshida et al., 2007). Furthermore, also
72
Sertoli cells and peritubular cells contribute to the self renewal of the spermatogonial
73
stem cells (Dadoune, 2007; de Rooij, 2009).
74
Especially in patients with maturation arrest a significant reduction of dividing
75
spermatogonia during mitosis was observed (Steger et al., 1998; Bar-Shira Mymon et
3
Page 4 of 22
4 76
al., 2003), whereas in hypospermatogenesis accelerated apoptosis rather than
77
proliferative
78
spermatogonial cell numbers (Takagi et al., 2001). Additionally, in cases with
79
maturation arrest reduced recombination rates (up to 50%) were found (Gonsalves et
80
al., 2004; Egozcue et al., 2005) and an increased frequency of genetic abnormalities
81
(Weedin et al., 2011).
82
In this study, we analyzed germ cell numbers in the early phase of spermatogenesis
83
with an emphasis on mitosis-meiosis transition. Furthermore, we classified the
84
deficiencies individually by distinguishing between high and low efficiency of meiosis.
85
Our main finding in this study was that cases with histologically normal
86
spermatogenesis and cases with hypospermatogenesis can compensate reduced
87
numbers of spermatogonia with a highly efficient meiosis suggesting a compensatory
88
mechanism.
dysfunction
was
suggested
to
be
responsible
for
reduced
89
4
Page 5 of 22
5 90
Materials and Methods
91
Patients
92
Testicular biopsies were obtained from 90 patients between 1998 and 2012 and were
93
indicated because of normogonadotropic obstructive or non-obstructive azoospermia.
94
After written informed consent, biopsies were taken under general anesthesia. The
95
Ethics Committee of the Medical Faculty of the Justus-Liebig-University, Giessen,
96
Germany approved the study (75/00 and 56/05). After fixation in Bouins solution and
97
embedding in paraffin, 5 µm sections were stained with hematoxylin and eosin and
98
spermatogenesis was evaluated histologically according to the scoring system of
99
Bergmann & Kliesch (1998, 2010). The patients were classified into histologically
100
normal spermatogenesis (Nsp, n=30; median age 39, range 16-56; mean score 9.93,
101
range 9-10), hypospermatogenesis (Hyp, n=30; median age 37, range 18-63; mean
102
score 8.43, range 6-10) and maturation arrest at the level of spermatocytes (MA,
103
n=30; median age 34.5, range 17-69; mean score 0).
104
Antibodies
105
To classify especially early stages of spermatogenesis (spermatogonia and mitosis-
106
meiosis-transition), we identified several highly robust markers with a good signal-to-
107
noise ratio and distinct expression pattern in human germ cells. For a detailed
108
analysis of impaired spermatogenesis, three markers were chosen (Table 1), OCT2
109
(octamer binding protein2, POU2F2) for early spermatogonia (Lim et al., 2011),
110
SAGE1 (sarcoma antigen1) for dividing and differentiating spermatogonia (Lim et al.,
111
2011) and mitosis-meiosis-transition (Looijenga, 2011), and Smad3 for pachytene
112
spermatocytes (Hentrich et al., 2011).
113
Immunohistochemistry
114
Immunohistochemistry of 5 µm sections of bouin-fixed, paraffin-embedded specimen
115
was performed as published (Konrad et al., 2007). The Envision System from DAKO
5
Page 6 of 22
6 116
(Hamburg, Germany) combined with DAB staining was used according to the
117
manufacturer`s instructions. Counterstaining was done with hematoxylin. The
118
antibodies used for detection are shown in Table 1. Digital images were obtained
119
with the inverse microscope FSX100 (Olympus, Hamburg, Germany) using the
120
Olympus FSX-BSW software. Images were processed with Adobe Photoshop 7.0.
121
Quantification of germ cells
122
Numbers of germ cells were obtained from 26.63±12.58 tubular cross-sections per
123
specimen
124
maturation arrest for each staining. As published recently (Hentrich et al., 2011) and
125
also shown in this study, it was sufficient to use approximately 15-35 nearly round
126
cross-sections for quantification; we never observed outliers.
127
There is no consensus on how quantification of testicular biopsies with emphasis on
128
impaired spermatogenesis should be performed. The number of Sertoli cells is
129
different in patients with maturation arrest or hypospermatogenesis (Hentrich et al.,
130
2011). Thus, the Sertoli cell to germ cell ratio is not applicable for quantification of
131
germ cell numbers.
132
Statistical methods Values from each experiment were used for calculation of the
133
means (per cross-section) and the respective standard errors of the mean (SEM).
134
Differences between the groups were calculated with the test from Mann Whitney. P
135
values below 0.05 were considered statistically significant.
136
Cut-off values to distinguish between normal germ cell numbers versus reduced
137
germ cell numbers were determined with ROC-curves (receiver operating
138
characteristic) and AUC (area under the curve). Besides sensitivity and specificity
139
also the positive Likelihood-Ratio (LR+) was calculated. For all statistical analyses
140
GraphPad Prism6 was used.
representing
normal
spermatogenesis,
hypospermatogenesis,
and
141
6
Page 7 of 22
7 142
Results
143
Identification of germ cell markers for early spermatogenesis
144
We identified expression of OCT2 mainly in spermatogonia Adark but also in very few
145
spermatogonia Apale (Fig. 1A, D, G). Often the OCT2-positive spermatogonia were
146
localized in small groups in close proximity to interstitial arterioles or venules which is
147
in accordance to the characteristics of the stem cell niche in murine testis (Shetty et
148
al., 2007; Yoshida et al., 2007) and thus have been also termed reserve stem cells
149
(Dym et al., 2009). However, we found also OCT2-negative Adark and Apale
150
spermatogonia.
151
We identified SAGE1 in very few spermatogonia Adark but in nearly all Apale and type
152
B spermatogonia as well as in (pre)-leptotene spermatocytes and round spermatids
153
(Fig. 1B, E, H), thus representing mitotic spermatogonia and transition of mitosis into
154
meiosis. For characterization of pachytene spermatocytes, we used Smad3 (Fig. 1C,
155
F, I).
156
Analysis of meiosis efficiency
157
All three markers were used for quantification of germ cell numbers (early
158
spermatogenesis) per cross-section and thus also for individual classification of
159
meiosis efficiency in cases of impaired spermatogenesis. Quantification of OCT2-
160
positive spermatogonia showed nearly identical values for normal spermatogenesis
161
and hypospermatogenesis, but cases with maturation arrest demonstrated a
162
significant reduction of approximately 50% (Table 2). Of note, quantification of
163
SAGE1-positive germ cells (spermatogonia up to (pre)-leptotene spermatocytes)
164
revealed strongly and significantly reduced cell numbers in hypospermatogenesis
165
and in patients with maturation arrest compared to the controls with histologically
166
normal
spermatogenesis
(Table
2).
Similarly
Smad3-positive
pachytene
7
Page 8 of 22
8 167
spermatocytes were reduced by 47% in hypospermatogenesis and by 91% in cases
168
with maturation arrest compared to normal spermatogenesis (Table 2).
169
Individual classification of patients with impaired spermatogenesis
170
For a detailed individual classification of the patients, we determined the cut-off
171
values for all three markers to distinguish between normal and lower germ cell
172
numbers (Table 3). Furthermore, we classified spermatogenesis into “high efficiency
173
of meiotic entry” (high numbers of Smad3-positive spermatocytes, Fig. 2A) and “low
174
efficiency of meiotic entry” (low numbers of Smad3-positive spermatocytes, Fig. 2B).
175
Both groups could be further divided into four subgroups according to the cut-offs for
176
early spermatogonia and dividing/differentiating spermatogonia (Fig. 2). The
177
individual evaluation of each patient revealed that patients with histologically normal
178
spermatogenesis (n=29) or patients with hypospermatogenesis (n=14) showed a
179
high efficieny of meiotic entry and normal numbers of early spermatogonia or
180
dividing/differentiating spermatogonia (Fig. 2A). Remarkably, only patients with
181
histologically normal spermatogenesis (7 from 8) as well as patients with
182
hypospermatogenesis (7 from 14) could compensate low numbers of spermatogonia
183
with a high efficiency of meiotic entry (Fig. 2). In contrast, patients with maturation
184
arrest always showed a low efficiency of meiotic entry and also revealed more often
185
defects in numbers of dividing/differentiating spermatogonia compared to defects in
186
early
187
compensatory mechanisms in meiosis since patients with normal numbers of
188
spermatogonia showed low efficiency of meiosis (Fig. 2B).
spermatogonia
(Fig.
2B).
Especially
subgroup
5
indicates
missing
189 190 191
8
Page 9 of 22
9 192
Discussion
193
Besides the identification and evaluation of markers for early spermatogenesis, the
194
primary aim of this study was the individual classification of defects in the early phase
195
of spermatogenesis in cases of impaired spermatogenesis.
196
Intriguingly, expression of OCT2 was found primarily in clusters of human Adark
197
spermatogonia in a stem cell niche most often adjacent to the interstitial vasculature.
198
This suggests that OCT2 is expressed very early in human spermatogenesis most
199
likely in stem cells as described by Lim et al. (21). However, as also shown by Lim et
200
al. (2011) as well as in this study only a subgroup of Adark spermatogonia is OCT2-
201
positive. Identification of few OCT2-positive Apale spermatogonia in this study is in
202
accordance to results obtained in monkeys (Hermann et al., 2010) which showed that
203
this subgroup also consists of undifferentiated spermatogonia, possibly stem cells.
204
Thus, we believe like Dym et al. (2009) and (Hermann et al., 2010) that both Adark
205
and Apale spermatogonia contain spermatogonial stem cells. In contrast, Ehmcke et
206
al. (2006) described that only the Adark spermatogonia comprise spermatogonial stem
207
cells.
208
In line with our previous findings (Hentrich et al., 2011), we found reduced numbers
209
of spermatogonia in cases with maturation arrest. Moreover, we observed low
210
numbers of early spermatogonia in cases with hypospermatogenesis as well as in
211
maturation
212
spermatogenesis. Thus, without Adark spermatogonia which contribute to stem cell
213
renewal, spermatogenesis is severely impaired (Holstein, 1999).
214
In addition to stem cell markers, we were also interested in identification of marker(s)
215
for dividing and differentiating spermatogonia. We observed SAGE1 expression in
216
very few Adark spermatogonia, but mainly in dividing spermatogonia and (pre)-
217
leptotene spermatocytes indicating entry into meiosis. Similarly Lim et al. (2011) and
arrest,
indicating
deficiencies
already
in
the
early
phase
of
9
Page 10 of 22
10 218
Chen et al. (2011) found SAGE1 in dividing spermatogonia and meiotic entry.
219
Interestingly, SAGE1 expression was not detectable before puberty (Lim et al.,
220
2011), thus further corroborating the importance of SAGE1 for differentiation of
221
spermatogonia.
222
We found that SAGE1-positive spermatogonia were severly reduced in patients with
223
hypospermatogenesis and maturation arrest but only rarely in patients with
224
histologically normal spermatogenesis. Thus, defects in the early phase of
225
spermatogenesis,
226
spermatogenesis. Similarly Bar-Shira Maymon et al. (2003) found a significant
227
reduction of dividing spermatogonia in patients with spermatogenic defects which
228
was also described for hypospermatogenesis and patients with maturation arrest
229
(Steger et al., 1998). However, because reduced numbers of early spermatogonia
230
might contribute to reduced numbers of mitotic spermatogonia, thus, any study on
231
mitotic defects must also include quantification of early spermatogonia in order to
232
discriminate properly between mitotic defects or reduced numbers of spermatogonia
233
before meiosis or a combination of both.
234
Smad3, which is also a key mediator of activin-induced proliferation of Sertoli cells
235
(Itman et al., 2011), is mainly expressed in pachytene spermatocytes in rat (Xu et al.,
236
2003), mouse (Itman et al., 2011) and human testis (Hentrich et al., 2011). The
237
reduced Smad3-positive pachytene spermatocyte numbers in patients with
238
hypospermatogenesis or maturation arrest (Hentrich et al., 2011) was also described
239
without marker expression analysis in cases with reduced sperm cell numbers
240
(Roosen-Runge et al., 1957; Zukerman et al., 1978). In contrast to the retained
241
Smad3 protein expression in cases with impaired spermatogenesis, the BOULE
242
protein (expressed in human leptotene to pachytene spermatocytes in normal
243
spermatogenesis) was completely absent in cases with maturation arrest (Luetjens et
especially
during
mitosis,
contribute
also
to
impaired
10
Page 11 of 22
11 244
al., 2004). Furthermore, alternative splicing of the BOULE gene was supposed to
245
serve as a predictive marker for meiotic efficiency (Kostova et al., 2007).
246
The individual classification of patients with impaired spermatogenesis by
247
categorizing meiotic efficiency in the early phase of spermatogenesis is a promising
248
extension of the classification system of Bergmann & Kliesch (1998, 2010). Although
249
studies are lacking to assess the early phase of spermatogenesis up to pachytene,
250
earlier studies showed up to 50% reduced recombination rates in cases with
251
maturation arrest (Gonsalves et al., 2004; Egozcue et al., 2005). Similarily, a 6-fold
252
increased frequency of genetic abnormalities was observed in cases of maturation
253
arrest compared to hypospermatogenesis (Weedin et al., 2011).
254
In this study we found that patients with histologically normal spermatogenesis
255
showed the lowest and patients with maturation arrest the highest numbers of
256
defects in the early phase of spermatogenesis as well as in the meiotic phase.
257
Especially in the subgroups 2-4 (Fig. 3A) with a high meiotic efficiency but low
258
numbers of spermatogonia, meiotic compensatory mechanisms must exist to
259
ameliorate these defects. Obviously, all cases with maturation arrest are lacking this
260
mechanism in addition to severly reduced numbers of spermatogonia. Interestingly,
261
in cases with maturation arrest reduction of early spermatogonia numbers (54%) is
262
nearly similar to the reduced numbers of dividing/differentiating spermatogonia
263
(42%), suggesting that loss of early spermatogonia might be primarily causative.
264
Of note, in cases of hypospermatogenesis, which is the most heterogenous group
265
(Takagi et al., 2001), we could identify both groups of patients with high or low
266
efficicency of meiosis in almost all combinations with reduced numbers of early
267
spermatogonia or reduced numbers of dividing/differentiating spermatogonia.
268
However, in contrast to cases with maturation arrest, the frequency of low efficiency
269
of meiosis was clearly lower in cases of hypospermatogenesis.
11
Page 12 of 22
12 270
In summary, this study revealed efficiency of meiotic entry as a factor for male
271
infertility and allows for a more detailed analysis of individual cases. We provide a
272
refinement of the established classification system launched by Bergmann & Kliesch
273
(1998, 2010). In line with the diagnosis of maturation arrest, all cases demonstrated
274
reduced numbers of spermatocytes, however, the majority additionally experienced
275
reduced numbers of spermatogonia. Furthermore, we could provide evidence that in
276
cases with normal spermatogenesis as well as in cases with hypospermatogenesis
277
compensatory meiosis mechanisms might exist to attenuate loss of spermatogonia.
278 279 280
Acknowledgments We thank Cornelia Hof for excellent technical assistance and
281
Prof. Kliesch (Center of Andrology and Reproductive Medicine, University of
282
Muenster, Germany) for biopsy sampling. This work was supported by a grant of
283
the Deutsche Forschungsgemeinschaft to L. Konrad (KFO 181/2 P2).
284 285
Disclosure/Conflict of interest
286
The authors declare no conflict of interest.
287
12
Page 13 of 22
13 288
References
289
Bar-Shira Maymon B, Yogev L, Yavetz H, Lifschitz-Mercer B, Schreiber L,
290
Kleiman SE, Botchan A &, Hauser R, Paz G 2003 Spermatogonial proliferation
291
patterns in men with azoospermia of different etiologies. Fertil Steril 80 1175–
292
1180.
293 294
Bergmann M & Kliesch S 1998 Hodenbiopsie. In Andrologie pp 66-71. Eds. W Krause and W Weidner. Stuttgart, Germany, Enke-Verlag.
295
Bergmann M & Kliesch S 2010 Testicular biopsy and histology. In Andrology, edn
296
3, pp 155-168. Eds E Nieschlag, HM Behre, S Nieschlag. Berlin Heidelberg,
297
Germany, Springer-Verlag.
298 299 300 301
Brugh VM 3rd & Lipshultz LL 2004 Male factor infertility: evaluation and management. Med Clin North Am 88 367-385. Cerilli LA, Kuang W & Rogers D 2010 A practical approach to testicular biopsy interpretation for male infertility. Arch Pathol Lab Med 134 1197–1204.
302
Chen YT, Chiu R, Lee P, Beneck D, Jin B & Old LJ 2011 Chromosome X-encoded
303
cancer/testis antigens show distinctive expression patterns in developing gonads
304
and in testicular seminoma. Hum Reprod 26 3232–3243.
305
Chiarini-Garcia H, Hornick JR, Griswold MD & Russell LD 2001 Distribution of
306
type A spermatogonia in the mouse is not random. Biol Reprod 65 1179–1185.
307
Chiarini-Garcia H, Raymer AM & Russell LD 2003 Non-random distribution of
308
spermatogonia in rats: evidence of niches in the seminiferous tubules.
309
Reproduction 126 669–680.
310 311 312 313
Dadoune JP 2007 New insights into male gametogenesis: what about the spermatogonial stem cell niche? Folia Histochem Cytobiol 45 141-147. de Rooij DG 2009 The spermatogonial stem cell niche. Microsc Res Tech 72 580– 585.
13
Page 14 of 22
14 314 315 316 317
Dohle GR, Elzanaty S & van Casteren NJ 2012 Testicular biopsy: clinical practice and interpretation. Asian J Androl 14 88-93. Dym M, Kokkinaki M & He Z 2009 Spermatogonial stem cells: mouse and human comparisons. Birth Defects Res C Embryo Today 87 27–34.
318
Egozcue J, Sarrate Z, Codina-Pascual M, Egozcue S, Oliver-Bonet M, Blanco J,
319
Navarro J, Benet J & Vidal F 2005 Meiotic abnormalities in infertile males.
320
Cytogenet Genome Res 111 337–342.
321 322
Ehmcke J, Wistuba J & Schlatt S 2006 Spermatogonial stem cells: questions, models and perspectives. Human Reprod Update 12 275–282.
323
Gonsalves J, Sun F, Schlegel PN, PJ, Hopps CV, Greene C, Martin RH & Reijo
324
Pera RA 2004 Defective recombination in infertile men. Hum Mol Genet 13 2875–
325
2883.
326
Hentrich A, Wolter M, Szardening-Kirchner C, Lüers GH, Bergmann M, Kliesch
327
S & Konrad L 2011 Reduced numbers of Sertoli, germ, and spermatogonial stem
328
cells in impaired spermatogenesis. Modern Pathol 24 1380-1389.
329
Hermann BP, Sukhwani M, Hansel MC & Orwig KE 2010 Spermatogonial stem
330
cells in higher primates: are there differences from those in rodents?
331
Reproduction 139 479–493.
332 333
Holstein AF 1999 Human spermatogenesis: basic research and clinical issues. Ann Anat 181 427-436.
334
Itman C, Wong C, Hunyadi B, Ernst M, Jans DA & Loveland KL 2011 Smad3
335
dosage determines androgen responsiveness and sets the pace of postnatal
336
testis development. Endocrinology 152 2076–2089.
337
Konrad L, Scheiber JA, Völck-Badouin E, Keilani MM, Laible L, Brandt H,
338
Schmidt A, Aumüller G & Hofmann R 2007 Alternative splicing of TGF-betas
14
Page 15 of 22
15 339
and their high-affinity receptors TβRI, TβRII and TβRIII (Betaglycan) reveal new
340
variants in human prostatic cells. BMC Genomics 8 318.
341
Kostova E, Yeung CH, Luetjens CM, Brune M, Nieschlag E & Gromoll J 2007
342
Association of three isoforms of the meiotic BOULE gene with spermatogenic
343
failure in infertile men. Mol Hum Reprod 13 85-93.
344
Lim J, Goriely A, Turner GD, Ewen KA, Jacobsen GK, Graem N, Wilkie AO &
345
Rajpert-De Meyts E 2011 OCT2, SSX and SAGE1 reveal the phenotypic
346
heterogeneity of spermatocytic seminoma reflecting distinct subpopulations of
347
spermatogonia. J Pathol 224 473–483.
348 349
Looijenga LH 2011 Spermatocytic seminoma: toward further understanding of pathogenesis. J Pathol 224 431-433.
350
Luetjens CM, Xu EY, Rejo Pera RA, Kamischke A, Nieschlag E & Gromoll J 2004
351
Association of meiotic arrest with lack of BOULE protein expression in infertile
352
men. J Clin Endocrinol Metab 89 1926-1933.
353 354
Shetty G & Meistrich ML 2007 The missing niche for spermatogonial stem cells: do blood vessels point the way? Cell Stem Cell 1 361-363.
355
Steger K, Aleithe I, Behre H & Bergmann M 1998 The proliferation of
356
spermatogonia in normal and pathological human seminiferous epithelium: an
357
immunohistochemical study using monoclonal antibodies against Ki-67 protein
358
and proliferating cell nuclear antigen. Mol Hum Reprod 4 227–233.
359
Roosen-Runge EC, Marberger E & Nelson WO 1957 Quantitative investigations on
360
human testicular biopsies. II. Infertility and other conditions. Fertil Steril 8 203–
361
219.
362
Takagi S, Itoh N, Kimura M, Sasao T & Tsukamoto T 2001 Spermatogonial
363
proliferation
and
apoptosis
in
hypospermatogenesis
364
nonobstructive azoospermia. Fertil Steril 76 901–907.
associated
with
15
Page 16 of 22
16 365
Weedin JW, Bennett RC, Fenig DM, Lamb DJ & Lipshultz LI 2011 Early versus
366
late maturation arrest: reproductive outcomes of testicular failure. J Urol 186 621-
367
626.
368 369 370 371
Xu J, Beyer AR, Walker WH & McGee EA 2003 Developmental and stage-specific expression of Smad2 and Smad3 in rat testis. J Androl 24 192–200. Yoshida S, Sukeno M & Nabeshima Y 2007 A vasculature-associated niche for undifferentiated spermatogonia in the mouse testis. Science 317 1722–1726.
372
Zukerman Z, Rodriguez-Rigau LJ, Weiss DB, Chowdhury AK, Smith KD &
373
Steinberger E 1978 Quantitative analysis of the seminiferous epithelium in
374
human testicular biopsies, and the relation of spermatogenesis to sperm density.
375
Fertil Steril 30 448–455.
376 377
16
Page 17 of 22
17 378
Figure Legends
379
Figure 1 Immunohistochemical detection of OCT2 (A, D, G), SAGE1 (B, E, H), and
380
Smad3 (C, F, I) in human biopsies from patients with histologically normal
381
spermatogenesis (A, B, C), patients with hypospermatogenesis (D, E, F) and cases
382
with maturation arrest (G, H, I). OCT2 was found mainly in Adark spermatogonia (Ad)
383
and to a lesser extent in Apale spermatogonia (Ap). Also few OCT2-negative Adark
384
spermatogonia [Ad(-)] could be identified. SAGE1 was expressed in Adark
385
spermatogonia up to type B spermatogonia (B), but also in some (pre)-leptotene
386
spermatocytes (pl). In contrast, Smad3 was localized primarily in the cytoplasm of
387
pachytene spermatocytes (p). The arrows point to the regions given enlarged in the
388
insets. scale bar – 25 µm (A-C), the same magnification was used in (D-I).
389
Counterstaining was done with hematoxylin.
390 391
Figure 2 Classification of patients into high meiosis efficiency (A) and low meiosis
392
efficiency (B). Both groups consist of four subgroups with normal numbers of early
393
spermatogonia (OCT2+) and dividing/differentiating spermatogonia (SAGE1+) or
394
reduced numbers of early spermatogonia (OCT2-) and dividing/differentiating
395
spermatogonia (SAGE1-). The cut-offs from Table 3 were used for classification.
396
Only cases with normal spermatogenesis (Nsp) or hypospermatogenesis (Hyp)
397
demonstrated high meiosis efficiency (A). Interestingly, only Nsp or Hyp could
398
compensate
399
dividing/differentiating spermatogonia (SAGE1-) with a highly efficient meiosis (A). In
400
contrast all cases with maturation arrest (MA) showed low meiosis efficiency besides
401
defects in numbers of early spermatogonia or dividing/differentiating spermatogonia
402
(B).
low
numbers
of
early
spermatogonia
(OCT2-)
and/or
17
Page 18 of 22
A
B
C
D
E
F
G
H
I
Figure 1
Page 19 of 22
HighGmeiosisGefficiency
Number of cases
A
SubgroupGGGGGGG1GGGGGGGGGGGGGGGGG2GGGGGGGGGGGGGGGGGG3GGGGGGGGGGGGGGGG4 GGGGGGOCT2GGGGGGG+GGGGGGGGGGGGGGGGG-GGGGGGGGGGGGGGGGGGG+GGGGGGGGGGGGGGGGGGGGSAGE1GGGGGGG+GGGGGGGGGGGGGGGG+GGGGGGGGGGGGGGGGGGG-GGGGGGGGGGGGGGGG-
B
Number of cases
LowGmeiosisGefficiency
SubgroupGGGGGGG5GGGGGGGGGGGGGGGGGGGG6GGGGGGGGGGGGGGGGGGG7GGGGGGGGGGGGGGGGGGGG8 GGGGGGOCT2GGGGGGG+GGGGGGGGGGGGGGGGGGGG-GGGGGGGGGGGGGGGGGGGG+GGGGGGGGGGGGGGGGGGGGGG GGGSAGE1GGGGGGGG+GGGGGGGGGGGGGGGGGGG+GGGGGGGGGGGGGGGGGGGG-GGGGGGGGGGGGGGGGGGGG-
Page 20 of 22
Table 1 Antibodies used for quantification of distinct germ cell types Protein
Source
Cat-No
Species Clonality
OCT2
Novocastra, Wetzlar,
NCL-OCT-207
Mouse
monoclonal 1:75
Rabbit
polyclonal
1:2000
Rabbit
polyclonal
1:300
Dilution
Germany
SAGE1
Sigma-Aldrich, Munich, HPA003033 Germany
Smad3
EnoGene Biotech, Frankfurt, Germany
Cat-No, catalog number
E021324
Page 21 of 22
Table 2 Numbers of distinct germ cells per cross-section in impaired spermatogenesis OCT2
SAGE1
Early spermatogonia
Mean SEM n P-values Nsp vs Hyp Nsp vs MA Hyp vs MA
Smad3
Dividing/differentiating spermatogonia
Pachytene spermatocytes
Nsp
Hyp
MA
Nsp
Hyp
MA
Nsp
Hyp
MA
2.09 0.25 30
2.05 0.23 30
0.96 0.16 30
15.62 0.82 30
11.16 0.61 30
9.10 0.77 30
16.35 0.92 30
8.64 0.56 30
1.53 0.36 30
ns 0.01 0.0001
0.01 0.001 ns
0.001 0.001 0.001
Nsp, normal spermatogenesis; Hyp, hypospermatogenesis; MA, maturation arrest; ns, not significant; vs, versus
Page 22 of 22
1 1
Table 3 Statistical analysis of the three markers used for individual classification OCT2
SAGE1
Smad3
Cut-Off = 1.06
Cut-Off = 10.84
Cut-Off = 8.6
sens
spec
LR+
sens
spec
LR+
sens
spec
LR+
Nsp/Hyp
23
87
1.8
40
90
4.0
47
97
14
Nsp/MA
57
87
4.3
77
90
7.7
100
97
30
Hyp/MA
57
77
2.4
77
60
1.9
100
53
2.1
2
Values for sensitivity (sens) and specificity (spec) are given in %; LR+, positive
3
likelihood ratio; Nsp, normal spermatogenesis; Hyp, hypospermatogenesis; MA,
4
maturation arrest
5
1
