Molecular Human Reproduction Vol.13, No.2 pp. 85–93, 2007
Advance Access publication November 17, 2006
doi:10.1093/molehr/gal101
Association of three isoforms of the meiotic BOULE gene
with spermatogenic failure in infertile men
E.Kostova, C.H.Yeung, C.M.Luetjens, M.Brune, E.Nieschlag and J.Gromoll1
Institute of Reproductive Medicine of the University, Domagkstrasse, Muenster, Germany
1
To whom correspondence should be addressed at: Institute of Reproductive Medicine of the University, Domagkstrasse 11,
D-48149 Muenster, Germany. E-mail: [email protected]
The complex process of spermatogenesis requires the expression and precise coordination of a multitude of genes. Abnormal
function of such genes is frequently associated with male infertility. Among these candidates is the human BOULE gene that is a
possible fundamental mediator of meiotic transition. In this study, we describe for the first time the existence of three BOULE
transcript variants (B1, B2 and B3). We investigated their tissue specificity and mRNA transcript levels in 23 testis biopsies from
infertile men. B1, B2 and B3 differed solely in their N-terminal sequences, which are encoded by three alternatively spliced exons 1.
In humans, all three isoforms are exclusively expressed in the testes in a relative proportion of 80:220:1 for B1, B2 and B3, respectively. RT–PCR quantification revealed significantly reduced mRNA expression of all three variants in testicular biopsies with
meiotic arrest (MA) compared with those with qualitatively complete spermatogenesis. Alteration of the B1/B2 and B1/B3 transcript ratios was correlated with reduced meiotic capacity of spermatocytes to produce round spermatids as assessed by flow
cytometry. Furthermore, BOULE mRNA reduction in biopsies with MA paralleled the absence of BOULE protein as analysed by
immunohistochemistry. In conclusion, the relative proportions of B1, B2 and B3 may serve as predictive markers for meiotic efficiency and thus the probability of finding haploid cells in the human testis. Among the three isoforms, B2 might have the major
role for meiotic completion.
Key words: BOULE/isoforms/meiotic arrest/mRNA quantification
Introduction
During spermatogenesis, a series of mitotic and meiotic cell divisions
leads to the production of haploid germ cells, which undergo dramatic
morphological changes giving rise to spermatozoa. This highly complex
process requires the expression and precise coordination of many
genes. Dysfunction of such genetic factors is associated with disturbed spermatogenesis and is suspected to be a frequent cause of
male infertility (Vogt, 1997; Krausz et al., 2000; Maurer and Simoni,
2000; Foresta et al., 2001; Kostova and Gromoll, 2001; Quintana-Murci
et al., 2001; Choi et al., 2004). The recent description of several
candidate infertility genes was the starting point for thorough investigations aimed at clarifying their role in reproduction. Among these
candidates is the DAZ (Deleted in Azoospermia) gene family consisting
of two autosomal genes, BOULE and DAZL (DAZ-like) and the Y
chromosomal DAZ gene cluster. BOULE has been proposed as the
progenitor of the DAZ gene family and is widely distributed, ranging
from worms to humans (Haag, 2001; Xu et al., 2001; Tung et al.,
2006). DAZL arose from BOULE in an ancestor of vertebrates, and the
Y chromosomal DAZ subsequently originated by duplication of
DAZL and transposition to the Y chromosome (Saxena et al., 2000;
Xu et al., 2001).
All DAZ members are RNA-binding proteins specifically expressed
in the germline and are essential for germ cell development (reviewed
in Reynolds and Cooke, 2005). In Drosophila, male boule mutants are
sterile, and their germ cells are arrested at the spermatocyte stage,
demonstrating the requirement of boule for meiosis (Eberhart et al.,
1996). In Caenorhabditis elegans, dazl is only needed for female
reproduction. Hermaphrodites deficient in dazl are sterile because of
blocked oogenesis (Karashima et al., 2000). In vertebrates, depletion
of Xdazl mRNA in Xenopus oocytes results in tadpoles lacking primordial germ cells (Houston and King, 2000). Mouse Dazl-null
mutants are sterile in both sexes, lacking both spermatozoa and
oocytes (Ruggiu et al., 1997).
The function of the DAZ proteins should be highly conserved,
because human BOULE and Xenopus Xdazl transgenes can rescue the
spermatogenic defect of Drosophila boule mutants, and human DAZ/
DAZL partially rescues the mouse Dazl-null phenotype (Houston
et al., 1998; Slee et al., 1999; Vogel et al., 2002; Xu et al., 2003).
To date, several infertility-related genes have been investigated in
an attempt to identify reliable molecular markers that can predict the
presence of haploid cells in the testes of infertile men. Thus, the
expression levels of human BOULE and CDC25A mRNAs have been
shown to be positively correlated with the success of sperm retrieval
during testicular sperm extraction (TESE) for assisted reproduction
(Lin et al., 2005; Cheng et al., 2006). In addition, BOULE protein was
found in the cytoplasm of primary spermatocytes, and its lack was
associated with spermatogenic arrest at the meiotic stage in a large
group of patients, independent of the factors causing infertility (Xu
et al., 2001; Luetjens et al., 2004). Owing to its essential role for
meiotic transition and thus for haploid germ cell production, BOULE
may be considered as a candidate molecular marker for the prediction
of complete spermatogenesis.
In this study, we describe for the first time the existence of three
BOULE isoforms and investigate their mRNA levels in testicular
biopsies of patients with qualitatively complete spermatogenesis or
© The Author 2006. Published by Oxford University Press on behalf of the European Society of Human Reproduction and Embryology. All rights reserved. For
85
Permissions, please email: [email protected]
E.Kostova et al.
with meiotic arrest (MA). We further analyse meiotic efficiency in
relation to the relative proportion of each transcript variant and speculate
about the physiological importance of the three isoforms.
Materials and methods
Patients and testicular samples
Twenty-three testis biopsies (21 unilateral and 2 bilateral) from 22 patients
attending the Institute of Reproductive Medicine (IRM) in Münster were
investigated in this study. The patients were either azoospermic (n = 13) or
severely oligozoospermic (<0.1 × 106/ml) (n = 9). All testicular biopsies were
obtained in an attempt to retrieve sperm for IVF. All patients underwent a complete clinical and physical examination and gave informed written consent to
the procedures and to further studies with their redundant testicular tissue
material. The clinical studies were approved by the Ethics Committee of the
University and the State Medical Board (Muenster, Germany; no. 4INie1).
Hormone values were measured in peripheral blood. FSH, LH and prolactin
were analysed by immunofluorometric assays (Autodelfia, Wallac, Freiburg,
Germany). Serum testosterone was measured using commercial radioimmunoassay kits (DRG Instruments GmbH, Marburg, Germany).
The testicular biopsies analysed in this study were subdivided by histological
evaluation into three groups: those with complete spermatogenesis (quantitatively normal or reduced to various degrees, n = 15), where the seminiferous
tubules contained elongated spermatids; those with MA (n = 4), defined as
biopsies in which primary spermatocytes represented the most advanced
spermatogenic stage and those with Sertoli cell only (SCO, n = 4), where
seminiferous tubules were completely devoid of germ cells. All biopsies were
denoted by patient number, biopsy site (left, L; right, R) and histological
diagnosis in the case of MA and SCO.
RNA extraction, RT and quantitative PCR
Commercially derived testis total RNA (BioChain, Hayward, CA, USA) from
two healthy men was used as control to analyse the normal distribution of
BOULE isoforms. Total RNA from various macaque tissues (testis, epididymis, prostate, liver, kidney, breast, heart and brain) and from human testis
biopsies (frozen tissue) was isolated using the Ultraspec® reagent (AMS
GmbH, Wiesbaden, Germany), following the manufacturer’s protocol.
To remove DNA traces in the RNA preparation, samples were treated with
DNase. One microgram of RNA (per 20 μl RT reaction) of each tissue sample
was reverse-transcribed using the Superscript II system (Stratagene, Heidelberg,
Germany). The efficiency of cDNA synthesis was verified by PCR amplification
of the b-actin gene.
Tissue-specific expression of BOULE isoforms 1, 2 and 3 was investigated
by PCR amplification using cDNA from human and macaque testes and various
macaque control tissues. The forward primers were specific for exon 1A, 1B or
1C, and the reverse primer was located in exon 2, which is common for the
three transcript variants. The primer sequences and amplicon sizes were
BOULE 1 (247 bp), fw 5′-ACAAGCAGCGCTGCTGCCTTG-3′; BOULE 2
(367 bp), fw 5′-GTTGAGGCTTCCCGCCACTG-3′; BOULE 3 (518 bp for
human and 290 bp for macaque), fw 5′-CCAGTGAGGGCACATTGCTTTG3′; rev 5′-TCAATTCCTCCTACAAAGATGC-3′.
Quantitative PCR analysis of BOULE 1, 2 and 3 was performed by TaqMan
methodology using ABI Prism 7000 Sequence Detection System (Applied
Biosystems, Darmstadt, Germany), according to the manufacturer’s guidelines.
Three real-time PCR assays were designed to amplify sequence stretches on
the boundary of exons 1 and 2 of each isoform. The primer pairs and FAMlabelled (6-carboxy-fluorescein) probes for each assay were BOULE 1 (amplicon
size, 106 bp), fw 5′-AGGTGGTTTCAAACTTTAGGCTTCT-3′, rev 5′-GTTT
GCATCTGGTTTGATGTTTGC-3′, FAM probe 5′-CCGGACTCGGTTTCC-3′;
BOULE 2 (amplicon size, 145 bp), fw 5′-CAGTGCCGCAACTTGCT-3′, rev
5′-GGCACTTGTTGGGTTATTCAAAGG-3′, FAM probe 5′-TTTGCGTCGCAGTCACT-3′; BOULE 3 (amplicon size, 104 bp), fw 5′-ATCACGAGGTCAGGAGATCGA-3′, rev 5′-GCACAGGTGACACAGGATTAGG-3′, FAM
probe 5′-CCAAAATGCAAACATCAAAC-3′. All BOULE assays had equal
efficiencies (characterized by the slope of the standard curves) allowing comparison of the mRNA levels of the three different transcripts. The 18S gene
(commercial assay, Applied Biosystems) was used as endogenous control to
normalize for variations in cDNA amounts. Each PCR contained cDNA
86
derived from 100 ng (BOULE assays) or 166 pg (18S assay) total RNA. PCR
conditions were 10 min 95°C, 40 × 15 s 92°C and 1 min 60°C (annealing and
extension). Samples were analysed in triplicate. Relative gene-expression analysis
was performed according to the 2−ΔΔCt method (Livak and Schmittgen, 2001).
Decreases in the transcript amounts of BOULE 1, 2 and 3 in the MA and SCO
groups were calculated relative to the values obtained from the biopsies with
complete spermatogenesis.
Flow cytometry
The cell types in the testicular biopsies (HC, elongated spermatids; 1C, round
spermatids; 2C, diploid germ and somatic cells and 4C, tetraploid cells, spermatocytes and dividing 2C cells) investigated in this study were analysed by
flow cytometry. The cell ploidy is reflected by the DNA content, which was
measured after propidium iodide (PI) staining. Testicular tissue specimens (5–
20 mg) from the biopsies were minced in 0.5 ml PI staining solution [0.1 mg/
ml RNase, 0.1% Triton X-100, 25 μg/ml PI in phosphate-buffered saline (PBS)
with 1 mg/ml bovine serum albumin (BSA)]. To release cells from the germinal epithelium, samples were mixed several times by aspiration using an electronic dispenser (EDOS 5222, Eppendorf, Hamburg, Germany). After
centrifugation (1 min at 40 × g), the supernatant was transferred to a new tube.
To release HC cells, the tissue pellet was resuspended in new 0.5 ml PI staining solution and homogenized twice for 3 s at 5000 r.p.m. with an Ultra-Turrax
T8 (IKA GmbH Staufen, Germany). Both fractions were pipetted together,
thoroughly mixed and allowed to stain for 30 min in the dark, at room temperature.
Two hundred microlitres of each sample was mixed with 50 μl fluorescentlabelled count control beads (Cyto-Cal; Duke Scientific Corporation, California, CA, USA) and analysed in a flow cytometer (Beckman Coulter FC500,
Krefeld, Germany) at 488-nm laser excitation. For each sample, ∼10 000 particles were analysed. The HC, 1C, 2C and 4C cell populations for each fraction
were identified by the DNA staining intensity after elimination of cell debris
using Percoll purified donor’s ejaculated spermatozoa (HC standard) to set
the lower threshold and subtraction of aggregates using the red fluorescence
channel auxiliary and intensity dot plot. The numbers of cells per
gram testicular tissue were computed and expressed as percentage for each cell
population.
Immunohistochemistry
Bouin-fixed, paraffin-embedded biopsies from all patient testes were routinely
sectioned at 4 μm. After deparaffinization and rehydration, sections were
blocked with 5% (v/v) goat serum, 0.1% (v/v) BSA in Tris-buffered saline
(TBS, pH 7.4) for 30 min. Polyclonal primary antibodies against the peptide
ETQEDAQKILQEAEKLNYKDKKLN (common to all three isoforms) were
used for detection of human BOULE protein (provided by Prof. R. Reijo, San
Francisco, CA, USA). Tissue sections were incubated at 1:300 dilution at room
temperature for 1 h. After washing, DAKO LSAB2 System (Dako Diagnostika,
Hamburg, Germany) was applied, according to the manufacturer’s protocol.
Briefly, slides were incubated with biotin-labelled goat anti-rabbit immunoglobulins and then with streptavidin conjugated to horseradish peroxidase,
15 min each, at room temperature followed by incubation with diaminobenzidine (DAB) substrate for 15 min. The sections were counterstained with haematoxylin and mounted with DAKO-Faramount (Dako Diagnostika). BOULE
protein staining was analysed microscopically (Axioskop, Zeiss, Oberkochen,
Germany) with ×25 or ×40 objectives. Digital images at ×400 magnification
were acquired with a CCD camera (Axiocam, Zeiss).
Genetic analysis
Analysis of the genomic organization of BOULE, nucleotide sequence alignments and protein motif searches were performed using e-net databases and
software tools: www.ncbi.nlm.nih.gov, www.ncbi.nih.gov/BLAST, www.ebi.
ac.uk/clustalw, www.genomatix.de/products/Gene2Promoter and http://motif.
genome.jp.
Statistical analysis
The software program SigmaStat 32 was used to analyse the data. Hormone
values were analysed by one-way analysis of variance (ANOVA) on Ranks,
and multiple pairwise comparison was done with Bonferroni’s t-test. The differences in the relative transcript numbers for BOULE 1, 2 and 3 in the three
BOULE gene and spermatogenic failure
histological groups (complete spermatogenesis, MA and SCO) were analysed
for statistical significance by the Kruskal–Wallis test and Dunn’s multiple
pairwise comparison. The correlation between the mRNA ratio for B1/B2 or
B1/B3 and the 1C/4C cell ratio was studied by linear regression. For all tests, a
P-value <0.05 was considered statistically significant.
Results
Genomic organization of BOULE
All previous studies dealing with BOULE assumed the existence of a
single isoform of this gene. In an attempt to study the expression and
function of human BOULE, we screened the GenBank database
(NC_000002 and several cDNA entries) to derive the coding sequence
of the gene and to analyse its genomic organization. Among the several published cDNA clones (partial and complete sequences) displaying homology to BOULE, we identified three different transcripts.
Two of them were described as BOULE isoform 1 and 2 (NM_197970
and NM_033030). A third unnamed cDNA sequence, highly similar
to these isoforms, we designated isoform 3 (AK093453). RT–PCR
amplification of the coding region of all three variants, followed by
sequencing, revealed some nucleotide mismatches for BOULE 3. The
sequence identified in our screening was submitted to the GenBank
(AM295005). Further analysis showed that BOULE 1, 2 and 3 differed solely in their 5′ ends, specifically in exon 1, which is encoded
by distinct genomic sequences. Each ‘variable’ exon 1 is spliced to a
common set of constant downstream exons 2–11. According to their
position in the genomic structure of BOULE, we assigned exon 1A to
isoform 1, exon 1B to isoform 2 and exon 1C to isoform 3 (Figure 1A).
Among them, exon 1A displayed the shortest sequence spanning 232
nucleotides (nt). Exons 1B and 1C consisted of 295nt and 382nt,
respectively. To investigate whether exons 1A, 1B and 1C showed
any similarity in their sequences, we performed ClustalW analysis.
The nucleotide alignment displayed no identity between the three exons.
A GenBank search in other animal models showed a related organization in higher primates (chimpanzee) and mice. Analogous structures with respect to the three first exons could be found at genomic
level. Nucleotide alignment of 1A, 1B and 1C to the genomic contig
of mouse Boule (NC_000067) showed high similarities corresponding
to 64% for 1A, 83% for 1B and 43% for 1C. In chimpanzees, identities reached 99% for all three exons (NW_104290). To verify that
these isoforms are expressed, as in humans, we screened the cDNA
data bank (EMBL). In mice, all three alternatively spliced variants
were transcribed (AK016403, AF272859 and NM_029267). In chimpanzees, a single mRNA sequence, corresponding to BOULE 1, is
currently described (XM_516011).
The distinct first exons alter the translational start site and lead to a
modification in the N-terminal part of the corresponding proteins.
Exons 1A and 1C contribute to the coding sequence of BOULE,
whereas 1B is non-coding, and the translational start site for this isoform is located in exon 2 (Figure 1A). The alternative start codon in
BOULE 1 and 3 has led to the recruitment of additional 12 and 6
amino acids (aa), respectively, without shifting the open reading
frame (ORF) (Figure 1C). To explore whether the extended N termini
of BOULE 1 and 3 may add new features to the protein, we screened
the initial aa sequences (exon 1 and partially exon 2) for characteristic
elements. No specific motifs were found.
Figure 1. (A) Genomic organization of human BOULE. BOULE 1, 2 and 3 comprise 11 exons. The three respective first exons, 1A, 1B and 1C, are alternatively
spliced to a common set of ‘constant’ exons (2–11). The nucleotide (nt) numbers for all exons and amino acid (aa) numbers for exons 1 and 2 are given below the
genomic organization. The sequence-coding region is represented by dark grey and the non-coding sequence by pale grey. (B) Genomic organization of human
DAZL. Its structure resembles closely that of BOULE but displays a single first exon. (C) N-terminal part (exon 1 and partially exon 2) of the proteins encoded by
BOULE 1, 2 and 3.
87
E.Kostova et al.
Another member of the DAZ gene family, the DAZL gene, closely
resembled this genomic structure but has only one first exon. Like
BOULE 3, the ORF of DAZL starts in exon 1 and encodes a single aa
(Figure 1B). However, no sequence similarity between the DAZL
exon 1 and BOULE exon 1C could be found, and neither did alignment of the DAZL exon 1 with 10-kb genomic sequence upstream of
BOULE exon 2 (including exons 1A and 1B) show a significant
match.
BOULE mRNA expression and tissue specificity
Previous data showed that BOULE mRNA expression (without discrimination between the different variants) is restricted to germ cells
in humans and mice (Xu et al., 2001). To determine whether all three
isoforms of BOULE are expressed and are germ cell-specific, we isolated total RNA from human testes and several other control tissues
from the closely related monkey species Macaca fascicularis. Using
RT–PCR analysis, products specific for BOULE 1, 2 and 3 were only
obtained from human and macaque testes and, in lower amount, from
macaque epididymis (Figure 2). The faint expression in the epididymis may be associated with the presence of spermatozoa and a
small number of other spermatogenic cells released from the testes.
The PCR amplicons with the expected size were sequenced to verify
their identity.
Quantification of BOULE 1, 2 and 3 in human testes
RT–PCR analysis showed distinct signal intensities for each BOULE
isoform. To estimate their relative proportions in human testes of
healthy men, we performed TaqMan-based quantification, which
showed most abundant expression for BOULE 2, followed by BOULE
1 and 3. The relative ratio for B1:B2:B3 was 80:220:1.
To estimate whether the expression of one or more isoforms was
correlated with spermatogenic efficiency, which would vary among
the azoospermic or severe oligozoospermic TESE patients, we analysed
the mRNA levels of BOULE 1, 2 and 3 and quantified the populations
of various germ cell types in the testicular biopsies with qualitatively
complete and disturbed spermatogenesis.
Clinical parameters and histopathological groups
Thirteen of the 22 patients included in this study were azoospermic,
and 9 were severely oligozoospermic (<0.1 × 106/ml) as diagnosed by
routine semen analysis. Four patients displayed qualitatively complete
spermatogenesis in one testicle and mixed atrophy (n = 3) or MA (n = 1)
Figure 2. RT–PCR analysis of BOULE 1, 2 and 3 mRNA expression. Specific
cDNA products were observed from human and macaque testes and macaque
epididymis. The amplified sequences span the borders from exon 1 and exon 2.
The size of the expected products (marked with arrows) was 247 bp for B1,
367 bp for B2 and 518 bp for B3. The PCR product of B3 generated from
macaque testes (arrowhead) displayed different size (290 bp) owing to nucleotide exchanges altering the 3′ splice site of its exon 1, as proven by genomic
DNA amplification and sequence analysis (data not shown). M, DNA size ladder.
88
Table I. Clinical parameters of the patients in each histopathological group
Number of patients
FSH (IU/l)
LH (IU/l)
Testosterone (nmol/l)
Prolactin (nmol/l)
Successful sperm
retrieval
Complete
spermatogenesis
Meiotic arrest
Sertoli cell only
15a
5.6 ± 2.7b,c
3.4 ± 1.2c
15.8 ± 4.6
163 ± 80.9
14 (93%)
4a
14.8 ± 8.0b
6.5 ± 3.8
24.7 ± 4.5
192 ± 32.1
2 (50%)
4
23.0 ± 8.0c
10.1 ± 6.8c
16.3 ± 10.0
195 ± 86.3
0 (0%)
Values are means ± SD.
One patient (no. 13) was included in both groups because his bilateral testis
biopsies used in the study displayed different histology (left, meiotic arrest;
right, complete spermatogenesis).
b
Significant difference, P = 0.029 for complete spermatogenesis versus meiotic
arrest group.
c
Significant difference, P < 0.01 (FSH) and P = 0.014 (LH) for complete
spermatogenesis versus Sertoli cell only (SCO) group.
a
in the other. One patient with MA and another with SCO in one testis
displayed mixed atrophy on the contralateral side. Therefore, the testicular biopsies investigated in this study [unilateral from 21 patients
and bilateral from 1 patient (no. 13)] were grouped according to histological diagnosis. The groups were as follow: (i) 15 with complete
spermatogenesis, (ii) 4 with MA and (iii) 4 with SCO. In the group
with complete spermatogenesis, five patients were vasectomized, two
had maldescended testes, five had other obstructions and three were
classified as idiopathic. Three patients with MA were idiopathic, and
one had maldescended testes. Among the SCO patients, one was idiopathic; one had bilateral maldescended testes; one was diagnosed with
Klinefelter syndrome and the fourth had Hodgkin disease. In none of
the patients were Y chromosomal microdeletions detected. In the first
group, spermatozoa were successfully retrieved at TESE from 14 of
15 biopsies but failed in 2 of 4 testes with MA and in all SCO biopsies. Significant differences between the histopathological groups were
found for the FSH and LH hormone values (Table I). Patients with MA
and those with SCO diagnosis displayed significantly higher FSH
levels compared with the group with qualitatively complete spermatogenesis (P = 0.029 and P < 0.01, respectively). Significantly different
LH levels were measured for the SCO group versus the group with
complete spermatogenesis (P = 0.014).
Quantification of BOULE isoforms in testicular biopsies of
infertile patients
In this investigation, we aimed to analyse BOULE mRNA expression
in testes with qualitatively complete spermatogenesis and compare the
transcript levels with those measured in blocked meiosis or SCO
tubules. We were particularly interested to see whether a single isoform was affected or whether an overall alteration might be observed
in spermatogenic failure. The mRNA levels of BOULE 1, 2 and 3
were studied in the group with MA as relative changes in transcript
numbers compared with the group with qualitatively complete spermatogenesis. SCO samples were included as negative controls.
All BOULE variants were detected in the biopsies with complete
spermatogenesis. The mean ratios between BOULE 1, 2 and 3 were
98:172:1, which deviated only slightly from the values estimated in
commercially derived testicular RNA from healthy men (80:220:1).
The TaqMan quantification in samples with MA showed a significant
5-fold overall decrease in the transcript levels when compared with
complete spermatogenesis (Figure 3 and Appendix). The reduction of
BOULE 1, 2 and 3 was 3.5-, 5- and 7.7-fold, respectively, and corresponded to ratio of 209:190:1. One of the four samples (13_L_MA)
BOULE gene and spermatogenic failure
Figure 3. BOULE mRNA expression in testicular biopsies of infertile patients
with qualitatively complete (Complete) spermatogenesis, meiotic arrest (MA)
or Sertoli cell only (SCO). Total BOULE transcripts, BOULE 1, 2 and 3, are
significantly reduced in testes with MA and hardly detectable in SCO biopsies
compared to controls with qualitatively complete spermatogenesis. Values are
mean ± SD. Numbers on the graph represent the mean fold decrease (MFD)
compared with the complete spermatogenesis group. Significant difference,
*P = 0.05–0.01; **P = 0.01–0.001.
from this group displayed apparently higher mRNA levels, resembling the values from the group with complete spermatogenesis. In
SCO biopsies, mRNA expression was hardly detectable, and in one
sample it was completely lacking (23_R_SCO). The decrease was
479-, 567-, 430- and 462-fold for total BOULE, B1, B2 and B3,
respectively (Figure 3).
Flow cytometry
For verification of the histological diagnosis and correlation of the
BOULE mRNA values to the testicular cell populations, we used flow
cytometric analysis. On the basis of the DNA amount (staining intensity), four different cell fractions were identified: elongated spermatids (HC), round spermatids (1C), diploid cells (2C) (spermatogonia
and somatic cells) and meiotic spermatocytes (4C). Few cells of the
4C fraction represented dividing 2C cells. Using tissue material from
the same testicular biopsies as for the mRNA quantification, we measured the percentage of HC, 1C, 2C and 4C cells (Figure 4). Testes
with qualitatively complete spermatogenesis showed the presence of
all four cell types, although the percentage of cell numbers fluctuated
between samples. The mean distribution for each fraction was 11, 35,
38 and 15% for HC, 1C, 2C and 4C, respectively. Furthermore, one
patient from this group (2_R) had strongly reduced haploid cell populations (6% HC and 12% 1C), which correlated with diminished
BOULE 1, 2, and 3 mRNA expression. In testes with MA only 2C
(∼84%) and 4C (∼16%) cells were identified. In contradiction of the
histological diagnosis, one biopsy with MA (13_L_MA) displayed
low numbers of 1C (9%) cells and in another (19_L_MA) only a few
4C cells could be found (2%). In SCO patients, 2C cells (∼98%) represented the major cell population and only a few 4C (∼2%) cells were
counted.
To examine whether BOULE mRNA was correlated with the cell
populations present in the testes, the 1C/4C ratio, representative of the
extent of meiosis completion, was investigated in relation to the transcript levels. 1C/4C values were positively associated with the transcript amounts of BOULE 1, 2 and 3 (r = 0.444, P = 0.03) (Figure 5A).
Thus, reduced transcript levels in patients with MA and SCO were
associated with zero 1C/4C values. An obvious deviation from this
pattern and the initial histological evaluation was found in two biopsies, namely in 13_L_MA and 2_R. In 13_L_MA, diagnosed with
MA, the elevated mRNA amount of BOULE 1, 2 and 3 paralleled the
Figure 4. Testicular cell populations determined by flow cytometry for 15 testicular biopsies with qualitatively complete spermatogenesis, 4 with meiotic
arrest (MA) and 4 with Sertoli cell only (SCO). Four cell populations were
identified: elongated and elongating spermatids (HC), round spermatids (1C),
diploid cells (germ cells and somatic cells) (2C) and spermatocytes and dividing spermatogonia (4C). R, sample from right testes and L, sample from left
testes.
presence of 1C cells (9%). In 2_R, a decreased 1C/4C ratio was
associated with strongly diminished transcripts.
Because mRNA amounts of BOULE 1, 2 and 3 were altered differentially, we tested the probability whether a modified B1 : B2 : B3
ratio can reflect the efficiency of haploid cell production. Linear
regression analysis showed a high negative correlation of B1/B2
(r = 0.670, P = 0.002) and B1/B3 (r = 0.718, P < 0.001) ratios with the
1C/4C cell numbers, both in the group with complete spermatogenesis
and that with MA (Figure 5B and C). B2/B3 was not correlated significantly with 1C/4C (data not shown).
BOULE protein expression
The association between mRNA quantity and BOULE protein expression was investigated by immunohistochemistry on testicular tissue
sections from all patient biopsies. Discrimination between BOULE 1,
2 and 3 was not possible because the antibody used was directed to an
epitope common to all isoforms. All biopsies with complete spermatogenesis showed BOULE staining. As previously reported (Luetjens
et al., 2004), BOULE protein was found in germ cells of the first meiotic division, detectable from leptotene to diplotene, with highest
expression in pachytene spermatocytes (Figure 6A). No staining was
found in round and elongating spermatids. The protein was exclusively localized in the cytoplasm. Peritubular cells and blood capillaries frequently showed non-specific staining. Three of the four biopsies
with MA contained meiotic spermatocytes that completely lacked
BOULE expression (Figure 6B). In one biopsy (17_L_MA), no
tubules containing meiotic cells could be found, although several
tissue sections were analysed. All sections from SCO biopsies showed
complete absence of spermatogenic cells and solely non-specific
staining in peritubular cells and blood vessels (Figure 6C).
Discussion
In this study, we describe for the first time the existence of three
BOULE isoforms and show that they are expressed in different
amounts during complete spermatogenesis. Furthermore, their mRNA
levels are significantly decreased in azoospermic patients with MA
and paralleled the lack of BOULE protein. Altered B1/B2 and B1/B3
transcript ratios correlated with reduced spermatogenic efficiency.
89
E.Kostova et al.
Figure 5. Correlation between mRNA transcript levels for BOULE 1, 2 and 3
and the 1C/4C ratio. Decreased 1C/4C values are associated with reduced
amounts of mRNA for each isoform (A) and an increase in the B1/B2 (B) and
B1/B3 (C) transcript ratios. The r and P values for the linear regression (B and
C) are shown on the graph. Complete, complete spermatogenesis; MA, meiotic
arrest; SCO, Sertoli cell only.
All previous investigations dealing with BOULE are based on the
sequence identified by Xu et al. (2001), corresponding to B2. In this
study, we describe two further transcript variants. Each isoform consists of a variable exon 1 and a constant set of common exons 2–11. The
three first exons show no apparent similarity and modify the N-terminal
90
part of the proteins by introduction of alternative start codons in the
sequence of B1 and B3. Analysis of their protein sequences did not
show any specific motifs in their N termini. However, we cannot
exclude the possibility that the aa extensions might moderate the
structure and/or function of the proteins.
Recent literature describes several examples of gene families with
similar genomic organization. For instance, the α and γ clusters of the
neural protocadherin (Pcdh) family display structures with variable
first exons, which are spliced to a group of common exons (Wu et al.,
2001; Zhang et al., 2004). A genome-wide search identified many
other mammalian genes containing multiple variable first exons, for
example, glucuronosyltransferase (UGT1), plectin, nitric oxide synthase (NOS1) and glucocorticoid receptor (GR) (Zhang et al., 2004).
This genomic organization of variable and constant regions seems to
be highly prevalent in mammalian genomes and provides a genetic
mechanism for cell- and tissue-specific regulation of gene expression,
as well as developmentally regulated gene expression. In the case of
human BOULE, all three isoforms are specifically expressed in testes.
This renders unlikely the probability that they are required for tissuespecific expression, as it was demonstrated for Drosophila boule,
where three brain-specific isoforms gained a function in the nervous
system (Joiner and Wu, 2004). Because the spermatogenic process is
precisely regulated and involves the interplay of multiple factors, it is
possible that the transcription of BOULE variants is either developmentally regulated or stage-specifically controlled. However, the
age-related expression was not investigated in this study, and the technique used did not allow discrimination of the different germ cell
types. To test this, single cell laser-capture combined with RT–PCR
will be useful. Furthermore, it is not clear how the age- or cell-specific
expression of BOULE 1, 2 and 3 would be controlled if this was the
case. Recent studies on the Pcdh gene family have demonstrated that
the transcription of each specific gene variant is controlled by a distinct promoter (Tasic et al., 2002; Wang et al., 2002). Similarly, our
preliminary analysis of the genomic sequence of BOULE, using
promoter-finder software tools, suggested the existence of three
different promoter regions, each preceding one variable first exon
(1A, 1B or 1C). On the basis of these data, we hypothesize that the
expression of B1, B2 and B3 may be regulated in a similar way. However,
it is also possible that the expression is driven by a single promoter
upstream of exon 1A. To check this assumption, thorough promoter
analysis is necessary.
In evolutionary terms, the acquisition of variable first exons in the
genomic sequence of BOULE should be an ancient event because not
only humans and primates but also lower mammals, such as mice, display analogous BOULE organization. Screening of the GenBank data
showed the existence of even more complex boule variants in
Drosophila (NM_079265, NM_168318, NM_168320, NM_168319),
where not only the first exon but also several other exons are alternatively spliced. The existence of multiple BOULE isoforms raises the
question which of them might be the progenitor form of the gene.
Exon 1B showed highest similarity between humans and mice (83%),
rendering isoform 2 as the presumable progenitor.
In this study, we showed that all three variants are exclusively
expressed in germ cells and seem to be essential for completion of
spermatogenesis. Similar association between BOULE mRNA and
spermatogenic failure in azoospermic men was found by Lin et al.
(2005), although they did not discriminate between BOULE 1, 2 and
3. Among the three isoforms, B2 displayed most abundant expression
in complete spermatogenesis. In testicular biopsies with MA, all transcripts were significantly lower, but BOULE 2 and 3 were affected
more strongly than BOULE 1. Furthermore, the increased B1/B2 and
B1/B3 ratios reflected less efficient production of haploid germ
cells when the groups with MA and complete spermatogenesis were
BOULE gene and spermatogenic failure
Figure 6. Immunohistochemical staining of BOULE in testis tissue sections counterstained with haematoxylin. No discrimination between BOULE 1, 2 and 3 was
possible. BOULE is expressed in the cytoplasm of spermatocytes in patients with complete spermatogenesis (arrows) (A) and is completely lacking from spermatocytes in testes with meiotic arrest (MA) (arrowheads) (B) and Sertoli cell only (C). Non-specific brown staining was detected in peritubular cells, Leydig cells and
blood vessels. Scale bars, 20 μm.
considered together. It is worth noting that when the latter group, in
which spermatogenesis differed quantitatively, was considered alone,
a stronger correlation (r = 0.777, P < 0.001) could be obtained
between B1/B2 and 1C/4C, reflecting spermatogenic efficiencies. In
the SCO group, BOULE mRNA was still detectable, although its
expression was greatly reduced and approached the TaqMan detection
limit. The presence of transcripts is likely to be associated with low
numbers of germ cells, considering that disruption of spermatogenesis
is frequently non-homogenous, with focal depletion of germ cells in
some tubules and intact spermatogenesis in others (Devroey et al.,
1995; Silber, 2000). Furthermore, the mRNA decrease was correlated
with the absence of BOULE protein in arrested spermatocytes. This
was consistent with previous findings where lack of BOULE protein
expression in azoospermic patients was associated with spermatogenic arrest at the meiotic stage (Luetjens et al., 2004). Considering
these findings together, we suggest the following hypothesis about the
physiological relevance of each isoform: B3 might have a minor role
in spermatogenesis (in adult men) in view of its many-fold lower
expression, compared with B1 and B2; the extra aa length of B1 may
be causing some sterical hindrance to the active site of the protein,
rendering B1 less effective in function than B2. The resulting higher
B1/B2 ratio would lead to less effective overall function of BOULE
and result in a blocked meiosis. Thus, B2 may play the most important
role, and B1 may have supporting or regulatory function in adult
human spermatogenesis. However, this hypothesis is based on the
mRNA expression levels and may not necessarily reflect the protein
variants in the testis.
Presently, the factors that regulate BOULE expression at the mRNA
and protein levels are not known. However, there must be a general
molecular mechanism regulating the expression of BOULE and other
genes involved in the meiotic cell cycle. Analogous histopathology
with lack of BOULE protein and transcript reduction was apparent in
the biopsies from the group with MA, although the spermatogenic
damage has a different aetiology. This suggests that other meiotic factors upstream of BOULE might control its expression. Studies in Drosophila describe a multitude of genes that are essential for the
mitosis–meiosis transition, for example, aly, comr, achi/vis, can, mia,
sa and nht (Perezgasga et al., 2004). The aly gene family is conserved
from plants to humans and is essential for the transcriptional activation of several genes, including twine and boule (White-Cooper et al.,
2000). Thus, analogous genes might be involved in the transcription
and/or translation of human BOULE.
In this study, some discrepancies between histological diagnosis
and flow cytometry data were observed. Thus, in 1 patient with MA,
as evaluated by histology, round spermatids were detected by flow
cytometry. This could be explained either by a failure to determine the
spermatids in the histology or by the presence of different cell types in
the tissue pieces used in the two distinct analyses. In another patient
with complete spermatogenesis, strongly reduced amounts of haploid
cells were detected. In both cases, the cell type estimation by flow
cytometry correlated precisely with the mRNA amount of BOULE,
confirming the quantitative RT–PCR method as a valuable tool for
investigating histopathology in TESE samples. Because the testicular
histology may be variable throughout the testis, fine needle aspiration
biopsies (FNAB) from multiple tissue locations would provide more
precise diagnosis. In addition, recent studies demonstrate that FNAB
is less invasive than standard testicular biopsy and, in combination
with the RT–PCR technique, is an accurate diagnostic tool (Bettella et
al., 2005; Letsas et al., 2006).
In conclusion, we identified three BOULE transcript variants and
have measured a significant decrease in their mRNA levels in biopsies
from patients with MA. An altered ratio between the isoforms (B1/B2
and B1/B3) was negatively associated with meiotic efficiency both in
qualitatively complete spermatogenesis and MA. Considering this
observation and the most abundant mRNA expression of B2, we
suggest a major physiological role for this isoform in complete spermatogenesis. Although the observed association needs confirmation
by further studies, we propose that BOULE transcript levels and
their proportions provide valuable molecular markers for the
presence of haploid germ cells in testes of patients with disturbed
spermatogenesis.
Acknowledgements
The authors thank Prof. Dr M. Bergmann at the University of Giessen for contributing biopsy material and Prof. Dr R. Reijo at the University of California,
San Francisco, CA, USA, for providing us with BOULE antibodies. E. Lahrmann
and N. Terwort are thanked for technical assistance, Dr J. Wistuba for assistance in the immunohistochemistry and Prof. M. Simoni for comments on the
manuscript. Susan Nieschlag and Dr T. Cooper are acknowledged for language
editing. This work was supported by the DFG (grant no. GR 1547/7-1).
References
Bettella A, Ferlin A, Menegazzo M, Ferigo M, Tavolini IM, Bassi PF and
Foresta C (2005) Testicular fine needle aspiration as a diagnostic tool in
non-obstructive azoospermia. Asian J Androl 7,289–294.
Cheng YS, Kuo PL, Teng YN, Kuo TY, Chung CL, Lin YH, Liao RW, Nan Lin JS
and Lin YM (2006). Association of spermatogenic failure with decreased
CDC25A expression in infertile men. Hum Reprod 21,2346–2352.
Choi JM, Chung P, Veeck L, Mielnik A, Palermo GD and Schlegel PN (2004)
AZF microdeletions of the Y chromosome and in vitro fertilization outcome.
Fertil Steril 81,337–341.
91
E.Kostova et al.
Devroey P, Silber S, Nagy Z, Liu J, Tournaye H, Joris H, Verheyen G and Van
Steirteghem A (1995) Ongoing pregnancies and birth after intracytoplasmic
sperm injection with frozen-thawed epididymal spermatozoa. Hum Reprod
10,903–906.
Eberhart CG, Maines JZ and Wasserman SA (1996) Meiotic cell cycle requirement for a fly homologue of human Deleted in Azoospermia. Nature
381,783–785.
Foresta C, Moro E and Ferlin A (2001) Y chromosome microdeletions and
alterations of spermatogenesis. Endocr Rev 22,226–239.
Haag ES (2001) Rolling back to BOULE. Proc Natl Acad Sci USA
13,6983–6985.
Houston DW and King ML (2000) A critical role for Xdazl, a germ plasmlocalized RNA, in the differentiation of primordial germ cells in Xenopus.
Development 3,447–456.
Houston DW, Zhang J, Maines JZ, Wasserman SA and King ML (1998) A
Xenopus DAZ-like gene encodes an RNA component of germ plasm and is a
functional homologue of Drosophila boule. Development 2,171–180.
Joiner ML and Wu CF (2004) Nervous system function for the testis RNAbinding protein boule in Drosophila. J Neurogenet 18,341–363.
Karashima T, Sugimoto A and Yamamoto M (2000) Caenorhabditis elegans
homologue of the human azoospermia factor DAZ is required for oogenesis
but not for spermatogenesis. Development 5,1069–1079.
Kostova E and Gromoll J (2001) The DAZ gene family and its role in
spermatogenesis- clinical and experimental aspects. Andrologia
33,239–240.
Krausz C, Quintana-Murci L and McElreavey K (2000) Prognostic value of Y
deletion analysis: what is the clinical prognostic value of Y chromosome
microdeletion analysis? Hum Reprod 15,1431–1434.
Letsas KP, Andrikoula M and Tsatsoulis A (2006) Fine needle aspiration
biopsy-RT-PCR molecular analysis of thyroid nodules: a useful preoperative diagnostic tool. Minerva Endocrinol 31,179–182.
Lin YM, Kuo PL, Lin YH, Teng YN and Nan Lin JS (2005) Messenger RNA
transcripts of the meiotic regulator BOULE in the testis of azoospermic men
and their application in predicting the success of sperm retrieval. Hum
Reprod 20,782–788.
Livak KJ and Schmittgen TD (2001) Analysis of relative gene expression data
using real-time quantitative PCR and the 2(-Delta Delta C(T)) method.
Methods 25,402–408.
Luetjens CM, Xu EY, Reijo Pera RA, Kamischke A, Nieschlag E and Gromoll J
(2004) Association of meiotic arrest with lack of BOULE protein expression
in fertile men. J Clin Endocrinol Metab 4,1926–1933.
Maurer B and Simoni M (2000) Y chromosome microdeletion screening in
infertile men. J Endocrinol Invest 23,664–670.
Perezgasga L, Jiang J, Bolival B Jr, Hiller M, Benson E, Fuller MT and WhiteCooper H (2004) Regulation of transcription of meiotic cell cycle and terminal differentiation genes by the testis-specific Zn-finger protein matotopetli. Development 131,1691–1702.
Quintana-Murci L, Krausz C, Heyer E, Gromoll J, Seifer I, Barton DE, Barrett T,
Skakkebaek NE, Raijpert-Demeyts E, Mitchell M et al. (2001) The relationship
92
between the Y chromosome DNA haplotypes and Y chromosome deletions
leading to male infertility. Hum Genet 108,55–58.
Reynolds N and Cooke HJ (2005) Role of the DAZ genes in male fertility.
Reprod Biomed Online 10,72–80.
Ruggiu M, Speed R, Taggart M, McKay SJ, Kilanowski F, Saundres F, Dorin J
and Cooke HJ (1997) The mouse DAZLa gene encodes a cytoplasmic protein essential for gametogenesis. Nature 389,73–77.
Saxena R, de Vries JW, Repping S, Alagappan RK, Skaletsky H, Brown LG, Ma P,
Chen E, Hoovers JM and Page DC (2000) Four DAZ genes in two clusters found
in the AZFc region of the human Y chromosome. Genomics 67,256–267.
Silber SJ (2000) Microsurgical TESE and the distribution of spermatogenesis
in non-obstructive azoospermia. Hum Reprod 15,2278–2284.
Slee R, Grimes B, Speed RM, Taggart M, Maguire SM, Ross A, McGill NI,
Saunders PT and Cooke HJ (1999) A human DAZ transgene confers partial
rescue of the mouse DAZL null phenotype. Proc Natl Acad Sci USA
96,8040–8045.
Tasic B, Nabholz CE, Baldwin KK, Kim Y, Rueckert EH, Ribich SA, Cramer P,
Wu Q, Axel R and Maniatis T (2002) Promoter choice determines splice site
selection in protocadherin alpha and gamma pre-mRNA splicing. Mol Cell
10,21–33.
Tung JY, Luetjens CM, Wistuba J, Xu EY, Reijo Pera RA and Gromoll J
(2006) Evolutionary comparison of the reproductive genes, DAZL and
BOULE, in primates with and without DAZ. Dev Genes Evol 3,158–168.
Vogel T, Speed RM, Ross A and Cooke HJ (2002) Partial rescue of the DAZL
knockout mouse by the human DAZL gene. Mol Hum Reprod 8,797–804.
Vogt PH (1997) Human Y chromosome deletions in Yq11 and male fertility.
Adv Exp Med Biol 424,17–30.
Wang X, Su H and Bradley A (2002) Molecular mechanisms governing Pcdhgamma gene expression: evidence for a multiple promoter and cis-alternative
splicing model. Genes Dev 16,1890–1905.
White-Cooper H, Leroy D, MacQueen A and Fuller MT (2000) Transcription
of meiotic cell cycle and terminal differentiation genes depends on a conserved chromatin associated protein, whose nuclear localisation is regulated.
Development 127,5463–5473.
Wu Q, Zhang T, Cheng JF, Kim Y, Grimwood J, Schmutz J, Dickson M,
Noonan JP, Zhang MQ, Myers RM et al. (2001) Comparative DNA
sequence analysis of mouse and human protocadherin gene clusters.
Genome Res 11,389–404.
Xu EY, Moore FL and Reijo RP (2001) A gene family required for human
germ cell development evolved from an ancient meiotic gene conserved in
metazoans. Proc Natl Acad Sci USA 98,7414–7419.
Xu EY, Lee DF, Klebes A, Turek PJ, Kornberg TB and Reijo RP (2003)
Human BOULE gene rescues meiotic defects in fertile flies. Hum Mol
Genet 12,169–175.
Zhang T, Haws P and Wu Q (2004) Multiple variable first exons: a mechanism
for cell- and tissue-specific gene regulation. Genome Res 14,79–89.
Submitted on August 3, 2006; resubmitted on October 12, 2006; accepted on
October 20, 2006
25.37
26.63
23.59
25.58
35.6
36.84
29.96
ND
MA
16_R_MA
17_R_MA
13_L_MA
19_L_MA
SCO
20_L_SCO
21_R_SCO
22_L_SCO
23_R_SCO
0.32
0.10
0.20
—
0.24
0.17
0.12
0.01
0.12
0.16
0.12
0.11
0.09
0.07
0.15
0.15
0.01
0.14
0.26
0.11
0.12
0.00
0.15
35.33
35.27
28.74
ND
26.71
26.94
22.94
25.15
22.32
24.85
22.51
23.03
24.07
21.39
22.17
21.37
22.25
21.81
22.63
21.53
22.70
22.19
23.16
0.23
0.03
0.16
—
0.16
0.08
0.16
0.04
0.19
0.08
0.19
0.15
0.23
0.18
0.08
0.16
0.04
0.29
0.13
0.032
0.109
0.04
0.239
Ct
SD
BOULE 2
ND
ND
35.88
ND
33.96
34.38
30.77
32.88
30.11
32.53
30.07
30.02
31.36
29.29
28.91
28.92
29.45
29.40
30.56
27.69
30.59
29.12
31.30
—
—
0.07
—
0.26
0.08
0.12
0.30
0.13
0.13
0.08
0.05
0.18
0.01
0.21
0.04
0.10
0.02
0.22
0.18
0.11
0.04
0.16
Ct
SD
BOULE 3
18.44
18.07
17.03
18.32
17.68
17.53
17.22
17.48
17.02
16.38
16.14
17.04
18.11
16.27
15.83
15.42
16.16
16.07
17.95
18.04
17.49
19.10
19.36
Ct 18S
0.06
0.07
0.14
0.19
0.28
0.10
0.05
0.04
0.24
0.08
0.05
0.33
0.08
0.10
0.17
0.03
0.10
0.04
0.34
0.04
0.13
0.75
0.50
SD
17.16
18.77
12.93
—
7.69
9.10
6.36
8.10
6.57
8.58
7.54
5.96
6.23
6.37
7.31
7.80
6.64
6.67
5.72
4.40
5.81
3.89
4.32
6.25 ± 1.32
ΔCt =
B1 – 18S
0.32
2.33
1.29
−0.29
−0.02
0.12
1.06
1.55
0.39
0.41
−0.53
−1.85
−0.44
−2.36
−1.94
Mean
SD
MFD
1.44
2.85
0.11
1.85
Mean
SD
MFD
10.91
12.52
6.68
—
Mean
SD
MFD
ΔΔCt =
ΔCtB1 – ΔCt
B1 mean
0.8011
0.1989
0.4099
1.2212
1.0140
0.9202
0.4785
0.3407
0.7649
0.7500
1.4406
3.6133
1.3566
5.1337
3.8238
1.4845
1.4784
—
0.3694
0.1387
0.9244
0.2784
0.4277
0.3445
3.5
0.0005
0.0002
0.0098
0.0000
0.0026
0.0048
567
2−ΔΔCt
16.89
17.20
11.70
—
9.02
9.41
5.72
7.67
5.30
8.47
6.37
5.98
5.96
5.12
6.34
5.96
6.08
5.74
4.67
3.49
5.21
3.09
3.81
5.44 ± 1.33
ΔCt =
B2 – 18S
−0.14
3.03
0.93
0.54
0.52
−0.32
0.90
0.52
0.64
0.30
−0.77
−1.95
−0.23
−2.35
−1.63
Mean
SD
MFD
3.58
3.97
0.28
2.23
Mean
SD
MFD
11.45
11.76
6.26
—
Mean
SD
MFD
ΔΔCt =
ΔCtB2 – ΔCt
B2 mean
1.1057
0.1224
0.5249
0.6870
0.6990
1.2483
0.5371
0.6990
0.6410
0.8113
1.7033
3.8548
1.1755
5.0982
3.0987
1.4671
1.4225
—
0.0835
0.0640
0.8236
0.2132
0.2961
0.3579
5.0
0.0004
0.0003
0.0130
0.0000
0.0034
0.0064
430
2−ΔΔCt
—
—
18.85
—
16.27
16.85
13.54
15.40
13.09
16.15
13.94
12.98
13.25
13.02
13.08
13.51
13.29
13.33
12.60
9.65
13.09
10.02
11.95
12.86 ± 1.52
0.23
3.29
1.08
0.12
0.39
0.16
0.22
0.65
0.43
0.47
−0.26
−3.21
0.23
−2.84
−0.91
Mean
SD
MFD
3.41
3.99
0.68
2.54
Mean
SD
MFD
—
—
5.99
—
Mean
SD
MFD
ΔCt = B3 – 18S ΔΔCt =
ΔCtB3 – ΔCt
B3 mean
0.8507
0.1020
0.4741
0.9202
0.7631
0.8981
0.8606
0.6388
0.7423
0.7220
1.1947
9.2535
0.8507
7.1437
1.8812
1.8197
2.6470
—
0.0939
0.0631
0.6227
0.1715
0.2378
0.2606
7.7
0.0000
0.0000
0.0158
0.0000
0.0039
0.0079
462
2−ΔΔCt
MA, Meiotic arrest; MFD, mean fold decrease; ND, not detectable; SCO, Sertoli cell only; SD, standard deviation.
The 18S gene was used for normalization. The MFD for each isoform in the groups with MA and SCO is calculated relative to the mean expression in the group with complete spermatogenesis (calibrator). All samples
were analysed in triplicate.
23.59
24.96
23.67
23.01
24.34
22.64
23.14
23.22
22.80
22.73
23.68
22.44
23.30
22.99
23.67
Ct
SD
BOULE 1
Complete
spermatogenesis
1_R
2_R
3_R
4_R
5_L
6_R
7_L
8_R
9_R
10_R
11_R
12_L
13_R
14_L
15_R
Biopsies
Appendix 1. Threshold cycle (Ct) values for BOULE 1, 2 and 3 estimated by TaqMan quantification and data analysis using the 2−ΔΔCt method (Livak and Schmittgen, 2001)
BOULE gene and spermatogenic failure
93