Download The human testis-specific proteome defined by

Molecular Human Reproduction, Vol.20, No.6 pp. 476–488, 2014
Advanced Access publication on March 5, 2014 doi:10.1093/molehr/gau018
ORIGINAL RESEARCH
The human testis-specific proteome
defined by transcriptomics and
antibody-based profiling
D. Djureinovic1,2, L. Fagerberg 3, B. Hallström3, A. Danielsson1,2,
C. Lindskog1,2, M. Uhlén 3, and F. Pontén 1,2,*
1
Science for Life Laboratory, Rudbeck Laboratory, Uppsala University, Uppsala SE-751 85, Sweden 2Department of Immunology, Genetics
and Pathology, Rudbeck Laboratory, Uppsala University, Uppsala SE-751 85, Sweden 3Science for Life Laboratory, KTH – Royal Institute
of Technology, Stockholm SE-171 21, Sweden
*Correspondence address. Science for Life Laboratory, Department of Immunology, Genetics and Pathology, Uppsala University,
751 85 Uppsala, Sweden. Tel: +46-18-611-3846. E-mail: [email protected]
Submitted on November 25, 2013; resubmitted on February 20, 2014; accepted on February 21, 2014
abstract: The testis’ function is to produce haploid germ cells necessary for reproduction. Here we have combined a genome-wide transcriptomics analysis with immunohistochemistry-based protein profiling to characterize the molecular components of the testis. Deep sequencing (RNA-Seq) of normal human testicular tissue from seven individuals was performed and compared with 26 other normal human tissue types.
All 20 050 putative human genes were classified into categories based on expression patterns. The analysis shows that testis is the tissue with the
most tissue-specific genes by far. More than 1000 genes show a testis-enriched expression pattern in testis when compared with all other analyzed
tissues. Highly testis enriched genes were further characterized with respect to protein localization within the testis, such as spermatogonia, spermatocytes, spermatids, sperm, Sertoli cells and Leydig cells. Here we present an immunohistochemistry-based analysis, showing the localization
of corresponding proteins in different cell types and various stages of spermatogenesis, for 62 genes expressed at .50-fold higher levels in testis
when compared with other tissues. A large fraction of these genes were unexpectedly expressed in early stages of spermatogenesis. In conclusion,
we have applied a genome-wide analysis to identify the human testis-specific proteome using transcriptomics and antibody-based protein
profiling, providing lists of genes expressed in a tissue-enriched manner in the testis. The majority of these genes and proteins were previously
poorly characterised in terms of localization and function, and our list provides an important starting point to increase our molecular understanding of human reproductive biology and disease.
Key words: immunohistochemistry / RNA sequencing / spermatogenesis / testis / tissue specificity
Introduction
The main function of the testis is to produce haploid germ cells necessary for our reproduction. This is achieved through a process that
requires high fidelity throughout the life of fertile males. Spermatogenesis is a stepwise and highly coordinated series of events that start in
cells located in the basal region of the seminiferous duct, where selfrenewing spermatogonial stem cells mitotically divide into spermatogonia. After several rounds of mitosis, the spermatogonia develop
into primary and secondary spermatocytes that divide by meiosis to
form spermatids. The final stage of spermatogenesis occurs in the
luminal region of the seminiferous duct where spermatids differentiate
into sperm. The various phases of spermatogenesis are distinguishable
due to the morphologically different cell types present at each stage,
since each cell type being closely associated with a particular phase
of spermatogenesis (Russell et al., 1990). In addition to germ cells,
the Sertoli cell is intimately involved in all stages of spermatogenesis.
This somatic cell, which lies interspersed throughout the seminiferous
duct, regulates both the development and the number and quality of
spermatogenic cells (Griswold 1998). Approximately 90% of the cell
mass in the testis is occupied by seminiferous ducts. In addition to
the generation of sperm, the testis is crucial for synthesis of androgens
that are necessary for spermatogenesis but also play a key role in
the development of male sex characteristics. The main production
of androgens, primarily testosterone, is localized in the testicular
Leydig cell.
The highly specialized functions, specific events and cell structure of
the testis endow a uniqueness to this particular tissue type. Although
spermatogenesis has been studied in depth and the different stages of
sperm maturation have been well defined, a large fraction of the genes
involved in the different stages are yet unknown. A detailed knowledge
regarding the molecular repertoire of the different cell types in the
testis is essential to understand the complex interaction under normal
and pathological conditions.
& The Author 2014. Published by Oxford University Press on behalf of the European Society of Human Reproduction and Embryology. All rights reserved.
For Permissions, please email: [email protected]
477
The human testis-specific proteome
The development of gene expression profiling techniques has allowed
for characterization of tissues with high resolution. Several previous
studies, based on expressed sequence tags (ESTs) and various microarray studies, have identified an unusual and diverse set of genes
expressed in testes. A large number of these genes were shown to be
unique to spermatogenic cells in both mouse (Kerr et al., 1994; Hoog,
1995) and humans (Pawlak et al., 1995; Penttilä et al., 1995). Studies
using Next Generation Sequencing technology (RNA-Seq) have also
been employed on testicular tissue. In a recent study, 10 different
types of human normal tissues showed that a higher fraction of genes
and more diverse mRNA populations (less dominance of a few highly
expressed genes) were expressed in the testis when compared with
other tissue types (Ramsköld et al., 2009). Although RNA-Seq and
other gene expression profiling techniques offer several advantages
(Mortazavi et al., 2008; Wang et al., 2009), the acquired data from the
transcriptomics analyses based on a heterogeneous tissue material
does not provide information regarding the expression pattern of corresponding proteins, e.g. what particular cell types express a specific
protein or in what subcellular compartment a specific protein is
expressed.
Here we analyzed genes expressed in normal human testis and these
data are compared with the transcriptome of 26 other human tissue and
organ types based on recently published RNA-seq data (Fagerberg et al.,
2014). Using this approach we not only detected all genes expressed in
the adult testis, but also identified genes enriched in the testis versus
other organs and tissue types. In order to translate the mRNA expression data to protein levels and to determine cell type-specific expression,
the proteins corresponding to the tissue-specific genes were studied
using the immunohistochemistry images at the Human Protein Atlas
(www.proteinatlas.org) (Uhlén et al., 2005, 2010; Pontén et al., 2011).
The testis-specific transcriptome and the identified cell type-specific proteins in normal human testis can serve as a starting point for further
disease-specific research.
Materials and Methods
Patient characteristics
The tissue samples used for transcript profiling of human testis consisted of
histologically normal tissue from surgical specimens, including testis from
four patients with cancer (two seminoma (Patient 1, age 34 years and
Patient 2, age 37 years), one embryonal cancer (Patient 3, age 26 years)
and one atypical plasmocytoma (Patient 4, age 56 years), two patients with
only scar tissue (Patient 5, age 62 years and Patient 6, age 50 years) and
one patient with focal inflammation (Patient 7, age 34 years). The corresponding histology of each case is shown in Supplementary data, Fig. S1.
Transcript profiling (RNA-Seq)
The use of human tissue samples was approved by the Uppsala Ethical Review
Board (Ref # 2011/473). Tissues samples, collected within the infrastructure
of an established biobank, were embedded in Optimal Cutting Temperature
(O.C.T.) compound and stored at 2808C. A hematoxylin– eosin-stained
frozen section (4 mm) was prepared from each sample using a cryostat and
the CryoJanew Tape-Transfer System (Instrumedics, St. Louis, MO, USA).
Each slide was examined by a pathologist to ensure accurate tissue morphology. Three sections (10 mm) were cut from each frozen tissue block and
homogenized prior to extraction of total RNA, using the RNeasy Mini Kit
(Qiagen, Hilden, Germany) according to the manufacturer’s instructions.
The extracted RNA samples were analyzed using either an Agilent 2100 Bioanalyzer system (Agilent Biotechnologies, Palo Alto, USA) with the RNA 6000
Nano Labchip Kit or an Experion automated electrophoresis system
(Bio-Rad Laboratories, Hercules, CA, USA) with the standard-sensitivity
RNA chip. Only samples of high-quality RNA (RNA Integrity Number
≥7.5) were used in the following mRNA sample preparation for sequencing.
mRNA sequencing was performed on Illumina HiSeq2000 and 2500
machines (Illumina, San Diego, CA, USA) using the standard Illumina
RNA-Seq protocol with a read length of 2 × 100 bases.
Analysis of data
The raw reads from 20 050 transcripts obtained from the sequencing system
were trimmed for low-quality ends with the software Sickle, using a phred
quality threshold of 20. All reads shorter than 54 bp after the trimming
were discarded. The processed reads were mapped to the GRCh37
version of the human genome with Tophat v2.0.3 (Trapnell et al., 2010).
Potential PCR duplicates were eliminated by applying the MarkDuplicates
module of Picard 1.77. To obtain quantification scores for all human genes,
FPKM (fragments per kilobase of exon model per million mapped reads)
values were calculated with Cufflinks v2.0.2 (Trapnell et al., 2010), which corrects for transcript length and the total number of mapped reads from the
library to compensate for different read depths for different samples. The
gene models from Ensembl build 69 (Flicek et al., 2012) were used in Cufflinks. In addition to Cufflinks, HTSeq v0.5.1 was run to calculate read
counts for each gene, which were used for analyses of differentially expressed
genes utilizing the DESeq package (Anders and Huber 2010; Flicek et al.,
2012). All data were analyzed with R Statistical Environment (RCoreTeam
2013) with the addition of the package ‘gplots’. A network analysis was performed using Cytoscape 3.0 (Shannon et al., 2003). For analyses performed in
this study where a log2-scale of the data was used, pseudo-counts of +1
were added to the data set.
Specificity classification
The average FPKM value of all individual samples for each tissue was used to
estimate the gene expression level. A cutoff value of 1 FPKM was used as the
detection limit. Each of the 20 050 gene transcripts were classified into one of
eight categories based on the FPKM levels: (i) ‘Not detected’—,1 FPKM
in testis; (ii) ‘highly testis enriched’—50-fold higher FPKM level in testis
compared with all other tissues; (iii) ‘moderately testis enriched’—5-fold
higher FPKM level in testis compared with all other tissues; (iv) ‘group
enriched’—5-fold higher average FPKM level in a group of 2 – 7 tissues
including testis compared with all other tissues; (v) ‘expressed in all low’—
detected in 27 tissues and at least one tissue ,10 FPKM; (vi) ‘expressed
in all high’—detected in 27 tissues and all tissues .10 FPKM; (vii) ‘testis
enhanced’—5-fold higher FPKM level in testis compared with the average
FPKM value of all 27 tissues; (viii) ‘mixed’—genes expressed in 1 – 26
tissues and in none of the above categories. The ‘testis-specific score’ was
defined as the Testis FPKM divided by the maximum FPKM value in any of
the other 26 tissues.
Hierarchical clustering
FPKM values for all 20 050 genes were used to calculate a correlation matrix
based on Spearman’s rank correlation coefficient for each pairwise combination of individual samples (n ¼ 7). The correlation matrix was used for
unsupervised hierarchical clustering analyses of individual samples using
the average linkage method to measure distances between clusters. The
results were visualized in a heatmap of all pairwise correlation coefficients.
478
Djureinovic et al.
Gene ontology analysis
Data availability
A gene ontology (GO) analysis (Ashburner et al., 2000) was performed using
the GOrilla tool (Eden et al., 2009) in order to determine overrepresented
GO categories in the gene set of tissue-enriched genes. For the cellular component analysis the GOSlim GOA associations were used to determine
whether genes were extracellular, intracellular or membrane bound. The
number of genes for each term was counted, allowing a gene to be associated
with more than one term. A list of all genes analyzed in this study was used as
the background list in GOrilla.
All the data (FPKM values for all the samples) will be available for downloads
without any restrictions (www.proteinatlas.org/about/download). The
primary data (reads) are available through the Array Express Archive (www.
ebi.ac.uk/arrayexpress/) under the accession number: E-MTAB-1733. The
transcript profiling data (FPKM values) for each gene in each cell and tissue
will also be available in the next version of the Human Protein Atlas (www.
proteinatlas.org).
Results
Tissue profiling
Tissue microarrays (TMAs) were generated as previously described (Kampf
et al., 2012), containing 1 mm cores of 46 different normal tissues in triplicates and 216 cancer tissues, representing the 20 most common cancers
(Pontén et al., 2008). The samples were received from the Department of
Pathology, Uppsala University Hospital, Sweden, approved by the Research
Ethics Committee at Uppsala University (Ref # Ups 02 – 577). The immunohistochemical (IHC) staining was performed as previously described (Kampf
et al., 2012). In brief, 4 mm sections of the TMA blocks were cut and mounted
on adhesive slides, baked at 608C for 45 min, followed by deparaffinization in
xylene and hydration in graded alcohols. Blocking for endogenous peroxidase
was performed in 0.3% hydrogen peroxide, and for antigen retrieval, a pressure boiler (Decloaking Chamber, Biocare Medical, Walnut Creek, CA,
USA) was used, boiling the slides for 4 min at 1258C in Citrate buffer, pH 6
(Lab Vision, Freemont, CA, USA). Automated immunohistochemistry was
performed using an Autostainer 480 instrumentw (Lab Vision). Primary antibodies diluted in UltraAb Diluentw (Lab Vision) and the secondary reagent
UltraVision LP HRP polymerw (Lab Vision) were incubated for 30 min
each at room temperature (RT). The slides were developed for 10 min at
RT, adding Diaminobenzidine (Lab Vision) as chromogen, followed by counterstaining with Mayers hematoxylin (Histolab) and mounting with Pertexw
(Histolab). The immunohistochemically stained and mounted slides were
scanned using an Aperio ScanScope XT Slide Scanner (Aperio Technologies,
Vista, CA, USA) for generation of high-resolution digital whole slide images,
followed by annotation by certified pathologists. In brief, the manual score of
IHC-based protein expression was determined as the fraction of positive cells
defined in different tissues: 0 ¼ 0 – 1%, 1 ¼ 2 – 25%, 2 ¼ 26 – 75%, 3.75%
and intensity of immunoreactivity: 0 ¼ negative, 1 ¼ weak, 2 ¼ moderate
and 3 ¼ strong staining. All of the tissues used as donor blocks were acquired
from the archives at the Department of Pathology of Uppsala University Hospital in agreement with approval from the Research Ethics Committee at
Uppsala University (Ref # Ups 02-577).
All annotation data together with validation of the antibodies is publically
available at the website www.proteinatlas.org (Uhlen et al., 2010; Pontén
et al., 2011), with the current version containing protein expression profiles
from 21 984 antibodies, corresponding to 16 621 genes.
Antibodies
Primary antibodies used for shown examples of IHC stainings include
HPA027850 (DMRT1), HPA011122 (PASD1), HPA003473 (PAGE1),
HPA036490 (DMRTB1), HPA059199 (C3ORF22), HPA019777 (DAZL),
HPA041915 (TEX101), HPA036124 (SHCBP1L), HPA049690 (TPTE),
HPA011284 (FMR1NB), HPA044387 (TNP1), HPA035105 (SMCP),
HPA021803 (ACTL7B), HPA038719 (ACRV1), HPA056530 (TMCO5A),
HPA056386 (PRM2), HPA042666 (GAPDHS), HPA05949 (AKAP4),
HPA049917 (C2ORF57), HPA048927 (C10ORF40), HPA054126 (SLCO6A1),
HPA034604 (FATE1), HPA043059 (DEFB119) and HPA028615 (INSL3)
have been developed within the Human Protein Atlas and are available
from AtlasAntibodies, Stockholm, Sweden.
The transcriptomic analysis of testicular tissue
The transcriptomes of 27 different human organ- and tissue-types were
analyzed using next generation sequencing (RNA-Seq) based on specimens from altogether 95 individuals (Fagerberg et al., 2014). The normalized mRNA levels were determined for each sample, calculated as FPKM
values where 1 FPKM roughly corresponds to one mRNA molecule per
average cell in the sample (Hebenstreit et al., 2011). Testicular tissues
from seven individuals (age 26– 62 years) were included in the analysis.
Microscopical evaluation of cryosections showed normal appearing testicular tissue with intact spermatogenesis, with only rare inflammation
and without contamination of tumor cells (Supplementary data, Fig. S1).
A total of 15 224 genes were detected in testis, with a mean expression value of .1 FPKM (with a variation from 15 397 genes in individual
2 to 15 039 genes in individual 3). Approximately 77% of all putative
protein coding genes (n ¼ 20 050) were detected in testis and of the
27 different tissues analyzed, the testis had the largest number of
detected genes.
The distribution of the genes detected in testis ranged from the
maximum of 4956 FPKM (the mitochondrially encoded cytochrome
c oxidase III gene) down to 1 FPKM to give a dynamic range of 5000.
The 30 genes with the highest levels of expression in the testis are
listed in Supplementary data, Table SI, which shows that a majority of
these genes are those with ‘house-keeping’ functions, such as genes
coding for mitochondrial and ribosomal proteins.
Analysis of the biological variation between the seven individual testis
samples was performed for the expression levels of all the protein-coding
genes in pairwise correlation plots and hierarchical clustering analysis.
The correlation overall was high with Spearman’s coefficients from
0.99 (Fig. 1A) to 0.91 (Fig. 1B). The average correlation coefficient for
all the seven testis samples was 0.96 (Supplementary data, Table S2,
Supplementary data, Fig. S2). These results demonstrate high technical
reproducibility between the testis samples, but also show low interindividual variation across the genome-wide expression pattern,
although variations due to isoform differences of the transcripts cannot
be ruled out.
As expected, similar comparisons between testis and other tissue
types show a higher degree of variation. The comparison of the testis
samples with data from the other 26 tissue types shows that the tissue
with the highest similarity to testis was the ovary with a Spearman coefficient of 0.75 (Fig. 1C), while the lowest correlation is found between
testis and liver with a Spearman coefficient of 0.65 (Fig. 1D).
Classification of the genes expressed in testis
The transcriptomics data obtained from the 27 tissues enabled us to
classify all the 20 050 protein-encoding genes into four major categories
The human testis-specific proteome
479
Figure 1 Correlation between samples and tissues. Scatter plot of average fragments per kilobase of exon model per million mapped reads (FPKM) values
for all detected genes comparing the two samples of normal testis with the highest correlation (A) and the two individuals with lowest correlation (B). Similar
analysis with a scatter plot showing the average FPKM values for the seven testis samples compared with the most similar other tissue type, ovary (C) and
most dissimilar other tissue type, liver (D). Color code: not detected (grey), highly testis enriched, moderately testis enriched, group enriched or testis
enhanced (blue), mixed (green), expressed in all, low (red), expressed in all, high (dark red). The correlation scores are given as pairwise Spearman’s correlation coefficients between the respective individuals or tissues.
based on their expression levels in testis (Fig. 2A). The major categories
included (i) genes that were not detected in testis (23%), (ii) genes that
showed a mixed expression pattern, being expressed in several, but
not in all tissue types (21%), (iii) genes that were expressed in all
tissues (46%) and (iv) genes that were categorized to have elevated
levels of expression in testis when compared with the average expression
levels in other tissue types (10%). The latter tissue-specific genes (n ¼
1934) were further divided into four subcategories depending on
degree of tissue specificity. Three hundred and sixty-four genes were
defined as highly enriched in testis (Supplementary data, Table SIII), characterized as the highest level of tissue-specificity with at least 50-fold
higher FPKM level in testis compared with any other tissue type. The
top 50 highly enriched genes are displayed in Table I, with the tissuespecific score calculated as the ratio between the average FPKM in the
seven testis tissues and the maximal FPKM in all the other tissues.
Seven hundred and sixty genes in testis were defined as moderately
enriched in testis (Supplementary data, Table SIV), with at least 5-fold
higher FPKM level in testis compared with all other tissues and 254
genes were defined as group enriched (Supplementary data, Table
SIV), with 5-fold higher average FPKM level in a group of 2 –7 tissues
including testis compared with all other tissues. In addition, 556 genes
were identified as testis enhanced (Supplementary data, Table SIV),
defined as having a 5-fold higher FPKM level in testis when compared
with the average FPKM value of all the 27 tissues.
From the transcriptomics analysis, it is possible to estimate the relative
mRNA pool for each of the categories, as shown in Fig. 2B. Almost
80% of the transcripts in the testis correspond to genes with potential
‘house-keeping’ functions expressed also in all other tissues analyzed.
This suggests that despite the highly specialized function of testis, most
of the transcription apparatus is still devoted to ‘house-hold’ functions
necessary for all types of cells (Fig. 2C, Supplementary data, Table SI).
An analysis of the testis-enriched genes shows that the mRNA pool constitutes 14% of the transcripts, with most of these corresponding to
genes highly (6%) or moderately (5%) enriched in testis. The expression
480
Djureinovic et al.
Figure 2 Classification of all human protein-coding genes and gene expression levels for highly abundant, highly enriched and group-enriched genes in
testis. Pie chart showing the distribution of gene expression comprising all 20 050 human putative protein-coding genes in testis (A). The pie chart is divided
into four major categories based on the transcript abundance in analyzed tissues, including not detected in testis (grey), expressed in all tissues (red color
shades), mixed expression with genes expressed in a large subset, but not all tissues (green) and genes with elevated expression in testis (blue color shades).
The genes with elevated expression in testis are further subdivided dependent on the degree of specificity as highly testis-enriched genes (light blue), moderately enriched in testis (blue), group enriched in testis (darker blue) and genes enhanced in testis (dark blue). Pie chart displaying the distribution of mRNA
molecules for the genes expressed in testis using the same color codes (B). Nearly 80% of all mRNA molecules are derived from housekeeping genes. The
FPKM values are shown across all 27 different tissue types for the 30 genes with most abundant expression in the testis. Each color corresponds to a gene
(C). Similar diagrams across all 27 different tissue types showing FPKM values for the top 30 highly testis-enriched genes. Each color corresponds to a gene
(D) and the top 30 testis group-enriched genes. Each color corresponds to a gene (E).
levels across the 27 analyzed tissues for the 30 most abundant highly
enriched genes are shown in Fig. 2D and in Fig. 2E the top 30 groupenriched genes are shown. The shared expression of these groupenriched genes is distributed among the other tissues, but with a slight
overrepresentation of shared genes with the brain.
The relation of the group-enriched genes is presented in a network
plot in Fig. 3A, which shows the amount of genes shared with a particular
tissue type (up to three different tissues). The tissue type with most
group-enriched genes in common with testis is the brain (n ¼ 48), followed by lung (n ¼ 33), whereas the remaining 24 other tissue types
only show a few group-enriched genes. A GO-based analysis of the
group of enriched genes shared between testis and lung shows an
enrichment for genes related to cilia function and movement. For the
group-enriched genes shared between testis and brain there was no
clear pattern of enriched GO-terms.
A GO-based analysis of all the 1124 genes identified as highly enriched
(n ¼ 364) and moderately enriched (n ¼ 760) in the testis indicated
a clear overrepresentation of genes associated with reproductive
process (154 genes) and spermatogenesis (105 genes). The cellular
localization of the corresponding proteins encoded by the highly and
moderately testis enriched genes were distributed as intra-cellular
(38%), membrane bound (5%) and extracellular (6%), whereas for half
of the proteins (50%) the localization was unknown (Supplementary
data, Fig. S3A).
481
The human testis-specific proteome
Table I The top 50 highly enriched genes in the testis.
Gene name
Gene description
Testis FPKM
Testis-specific score
.............................................................................................................................................................................................
PRM2
Protamine 2
2447
2520
PRM1
Protamine 1
2174
1489
TNP1
Transition protein 1 (during histone to protamine replacement)
1377
1353
PHF7
PHD finger protein 7
889
79
ANKRD7
Ankyrin repeat domain 7
567
505
TSACC
TSSK6-activating co-chaperone
488
604
INSL3
Insulin-like 3 (Leydig cell)
406
114
BOD1L2
Biorientation of chromosomes in cell division 1-like 2
405
4049
SMCP
Sperm mitochondria-associated cysteine-rich protein
344
470
LELP1
Late cornified envelope-like proline-rich 1
335
3346
TCP11
t-complex 11 homolog (mouse)
317
468
HMGB4
High mobility group box 4
307
2400
TUBA3C
Tubulin, alpha 3c
301
1244
C11orf 20
Chromosome 11 open reading frame 20
279
287
LDHC
Lactate dehydrogenase C
269
167
SPATA8
Spermatogenesis associated 8
250
1288
DEFB119
Defensin, beta 119
248
447
GSG1
Germ cell associated 1
244
110
FATE1
Fetal and adult testis expressed 1
179
300
ODF1
Outer dense fiber of sperm tails 1
177
1685
SPATA22
Spermatogenesis associated 22
172
172
AC007557.1
HCG1816075; uncharacterized protein
156
1146
PGK2
Phosphoglycerate kinase 2
149
1485
GAPDHS
Glyceraldehyde-3-phosphate dehydrogenase, spermatogenic
147
68
ATPAF1-AS1
ATPAF1 antisense RNA 1
145
815
AKAP4
A kinase (PRKA) anchor protein 4
145
1446
ACSBG2
acyl-CoA synthetase bubblegum family member 2
141
109
SHCBP1L
SHC SH2-domain binding protein 1-like
132
1171
ZPBP
Zona pellucida-binding protein
132
445
DDX4
DEAD (Asp-Glu-Ala-Asp) box polypeptide 4
127
691
BRDT
Bromodomain, testis specific
125
111
C10orf62
Chromosome 10 open reading frame 62
122
367
TPD52L3
Tumor protein D52-like 3
121
1210
CAPZA3
Capping protein (actin filament) muscle Z-line, alpha 3
115
68
GTSF1L
Gametocyte-specific factor 1-like
114
83
SPATA4
Spermatogenesis associated 4
114
68
UBQLN3
Ubiquilin 3
112
1124
DKKL1
Dickkopf-like 1
112
348
CETN1
Centrin, EF-hand protein, 1
110
1097
TPTE
Transmembrane phosphatase with tensin homology
105
170
TMCO5A
Transmembrane and coiled-coil domains 5A
105
193
ZPBP2
Zona pellucida binding protein 2
104
207
ACTL7A
Actin-like 7A
104
1036
EPPIN
Epididymal peptidase inhibitor
103
192
ADAD1
Adenosine deaminase domain containing 1 (testis specific)
103
1025
TEX101
Testis expressed 101
96
95
CCDC62
Coiled-coil domain containing 62
94
138
Continued
482
Djureinovic et al.
Table I Continued
Gene name
Gene description
Testis FPKM
Testis-specific score
.............................................................................................................................................................................................
DYDC1
DPY30 domain containing 1
91
71
C17orf64
Chromosome 17 open reading frame 64
85
266
SYCP3
Synaptonemal complex protein 3
84
105
Figure 3 Network plot of enriched genes, characterization of the highly enriched genes and cell type-specific expression for the 62 highly enriched genes
previously not well characterized. Network plot of the testis-enriched genes showing the relation of group-enriched genes (A). The light blue circle shows
the highly and moderately testis-enriched genes (n ¼ 1124). Dark blue nodes show the number of genes that are group enriched in two or three different
tissue types (grey circles). The size of each blue node is related to the square root of the number of genes enriched in a particular combination of tissues. Pie
chart based on the 364 highly testis-enriched genes, showing the distribution of preexisting knowledge with regard to function and cell types these genes are
expressed in (B). Sixty-two percent of the highly testis-enriched genes are previously poorly characterized. Pie chart showing the cell type localization based
on protein profiling using immunohistochemistry for 62 of the previously uncharacterized highly testis-enriched genes (C).
Cell type distribution of the testis specific
genes
With the large amount of unknown genes based on the GO analysis, we
further investigated how well characterized the 364 highly enriched genes
in testis were by using gene summary information from Uniprot and
NCBI data regarding function and the cell type specificity. Similar to
the GO analysis, a large fraction (62%) of the highly enriched genes
was not well characterized and only for 16% of the genes (n ¼ 60) the
specific cell type expressing the gene was known (Fig. 3B). The cell typespecific expression in relation to the different stages of the spermatogenesis of these previously known genes is shown in Supplementary data,
483
The human testis-specific proteome
Fig. S3B and Table SV. According to the analysis of previous literature,
more than one-third of these genes (40%) were expressed in more
than one cell type and of the cell type specifically expressed genes over
half were expressed in the later stages of spermatogenesis, including
spermatids (21%) and sperm (30%). Relatively few testis-specific genes
have previously been identified as expressed in early stages of the spermatogenesis, in Sertoli or Leydig cells.
The PASD1 gene encodes a putative protein with nuclear expression in
spermatogonia. PASD1 is a proposed transcription factor and has also
been classified as a cancer testis antigen (Liggins et al., 2004). The
DMRTB1 gene contains a DNA-binding domain but the putative protein
is unknown with regard to function and expression pattern (Ottolenghi
et al., 2002). Expression of the putative protein encoded by chromosome
3 open reading frame 22 (C3ORF22) is also found in spermatogonia and
shows a cytoplasmic expression pattern.
Antibody-based profiling of the testis
For the remaining highly enriched genes where cell type previously was
unknown (n ¼ 305), the expression patterns were evaluated, for the
corresponding proteins with available antibodies (n ¼ 260), using the
online Human Protein Atlas (HPA) database. The HPA provides immunohistochemically stained TMA cores from 46 different normal human
tissues, including three cores of human testicular tissue. The IHC staining
was visually evaluated with regard to testis specificity and cellular distribution. The antibody-based profiling allows further determination of localization within different subcellular compartments and cell types within
the testis, such as different morphologically defined stages during spermatogenesis, i.e. spermatogonia, spermatocytes, spermatids and sperm,
and other testis specific cell types such as Sertoli cells and Leydig cells.
The expression pattern for genes with fully congruent RNA-Seq and
protein expression data (n ¼ 62) was analyzed in detail, and a large
number of genes expressed in early phases of spermatogenesis were
identified (Fig. 3C and Supplementary data, Table SVI). The protein expression pattern in testis corresponding to these highly testis-enriched
genes are displayed in Supplementary data, Fig. S4. Almost half of
these genes showed expression in the earliest stages of spermatogenesis,
including spermatogonia (38%) and spermatocytes (11%). Only a few
genes were found to be localized to Sertoli cells and none of the highly
enriched genes where cell type previously was unknown in testis was
found to be expressed in Leydig cells.
Genes expressed in spermatocytes
Spermatocytes, derived from type B spermatogonia, can be subdivided
into primary spermatocytes that enter the first meiosis and secondary
spermatocytes that enter the second meiosis to produce haploid
spermatids. Several of the testis-specific proteins localized to spermatocytes were involved in testicular differentiation, proliferation and
meiosis. Examples of proteins expressed in spermatocytes as revealed
by the immunohistochemistry analysis are shown in Fig. 4B. The
deleted-in-azoospermia-like (DAZL) gene encodes a well-known RNAbinding protein that is essential for gametogenesis in both males and
females. In the testis, this protein is localized to the nucleus of
spermatogonia, but relocates to the cytoplasm in spermatocytes
during meiosis (Reijo et al., 2000). The testis expressed 101 (TEX101)
gene, has previously been reported to be associated with various
phases of spermatogenesis, but more precise knowledge regarding
the specific function and expression pattern is unknown (Teng et al.,
2006). Two additional examples of highly testis-enriched genes showing
cytoplasmic expression predominantly in spermatocytes are the two
cancer testis antigens transmembrane phosphatase with tensin homology
(TPTE) (Tapparel et al., 2003) and fragile X mental retardation 1 neighbor (FMR1NB) with previously no information regarding what cell type in
testis this gene is expressed in as well as the uncharacterized SHC
SH2-domain binding protein 1-like (SHCBP1L) gene.
Genes expressed in spermatogonia
Genes expressed in spermatids
Spermatogonia are diploid cells that form the basal layer of the seminiferous duct and present the initial phase of the spermatogenesis. The
spermatogonia can be divided into two subtypes, type A cells that
have stem cell-like properties and maintain the spermatogonia population and type B cells that produce primary spermatocytes. Both type
A and B cells result from mitotic divisions (de Rooij and Russell 2000).
In Fig. 4A, five examples of proteins with specific expression in
spermatogonia are shown as revealed by the immunohistochemistry
analysis. Doublesex- and mab-3-related transcription factor 1 (DMRT1)
encodes a protein that is described to primarily be expressed in the
nuclei of spermatogonia (Jørgensen et al., 2012). This transcription
factor plays a key role in male sex determination and differentiation by controlling testis development and male germ cell proliferation (Matson et al.,
2010). Also PAS domain-containing 1 (PASD1), P antigen family member
1 (PAGE1) and DMRT-like family B with proline-rich C-terminal,
1 (DMRTB1) show nuclear expression restricted to spermatogonia
located along the basement membrane of seminiferous ducts. The function of PAGE1 is unknown, but the PAGE1 gene belongs to a family of
genes that has been described to be expressed in a variety of tumors,
but with a testis restricted expression in normal tissues (cancer testis antigens) (Brinkmann et al., 1998). For the PASD1 and DMRTB1 genes
there is previously only evidence of existence at the transcript level.
Spermatids are derived from secondary spermatocytes and these cells
are subdivided into early, round spermatids that are transcriptionally
active and late, elongated spermatids that are transcriptionally inert
(Kierszenbaum and Tres, 1975). The formation of the acrosome, a testisspecific organelle required for sperm– egg interactions, is formed during
the meiotic development of the late spermatid into mature spermatozoa.
The highly testis-enriched proteins with known function and predominant expression in spermatids are involved in conversion of nucleosomal
chromatin and sperm development and maturation. Five examples of
proteins showing expression in spermatids, including two proteins with
expression restricted to the acrosome, are shown in Fig. 4C. The transition protein 1 (TNP1) gene encodes a spermatid-specific product of
the haploid genome which replaces histone and is itself replaced in the
mature sperm by the protamines. TNP1 protein is expressed in late
stage, elongating spermatids and is involved in the conversion of nucleosomal chromatin to the compact, non-nucleosomal form found in the
sperm nucleus (Wouters-Tyrou et al., 1998). The proteins encoded by
sperm mitochondria-associated cysteine-rich protein (SMCP) and actinlike protein 7B (ACTL7B) show a similar expression pattern in spermatids.
SMCP encodes a sperm mitochondria-associated cysteine-rich protein
that localizes to the capsule associated with the mitochondrial outer
membranes and is thought to function in the organization and
484
Djureinovic et al.
Figure 4 Examples of immunohistochemistry-based protein profiling of highly testis-enriched genes, showing expression patterns in cell types representing different stages of spermatogenesis. Tissue sections showing parts of normal seminiferous ducts with proteins expressed in spermatogonia (A), spermatocytes (B), spermatids (C) and sperm (D). Note differences in expression pattern between DMRT1, PASD1, PAGE1 and DMRTB1 (nuclei) and
C3ORF22 (cytoplasm) in spermatogonia. In spermatids TNP1, SMCP and ACTL7B are expressed in the nuclei, whereas ACRV1 and TMCO5A show a
distinct acrosomal pattern of expression.
Figure 5 IHC analysis of highly enriched genes in testis that are expressed in Sertoli cells and Leydig cells. Membranous positivity of Sertoli cells shown by
antibody towards SLCO6A1, whereas FATE1 and DEFB119 show cytoplasmic staining pattern in both spermatogenic cells and Sertoli cells. INSL3 shows
cytoplasmic expression restricted to the Leydig cells.
485
The human testis-specific proteome
stabilization of the helical structure of the sperm’s mitochondrial sheath
(Hawthorne et al., 2006). The ACTL7B gene, with evidence of existence
at the transcript level, encodes a putative protein of unknown function
that is suggested to be a member of a family of actin-related proteins
which share significant amino acid sequence identity to conventional
actins (Chadwick et al., 1999). The two examples of proteins showing
a distinct acrosomal expression pattern, include the well-characterized
acrosomal vesicle protein 1 (ACRV1) and the uncharacterized transmembrane and coiled-coil domains 5A (TMCO5A) gene which only has
evidence of existence at the transcript level. The ACRV1 gene encodes
an acrosomal vesicle protein, which arises within the acrosomal vesicle
during spermatogenesis, and is associated with the acrosomal membranes and matrix of mature sperm (Herr et al., 1990). ACRV1 may be
involved in binding or penetration of the oocyte zona pellucidum by
the mature sperm.
Genes expressed in sperm
Following the maturation of late stage spermatids, the testicular end stage
of spermatogenesis results in the formation of sperm. Sperm consists of
uniflagellar sperm cells that are localized along the lumen of seminiferous
ducts. Specific proteins localized to mature sperm are involved in functions including sperm motility and male fertility. Examples of proteins
highly enriched in testis that are expressed in sperm are shown in
Fig. 4D. The protamine 2 (PRM2) gene shows the highest expression
level (2447 FPKM) of all the 364 highly testis-enriched genes (Wykes
et al., 1995). Protamines substitute for histones in sperm and represent
the major DNA-binding proteins in sperm nuclei. The function of protamines is to package sperm DNA into a highly condensed, stable and inactive complex. The glyceraldehyde-3-phosphate dehydrogenase,
spermatogenic (GAPDHS) gene, with specific expression in the testis,
has a similar function as the ubiquitously expressed GAPDH in regulating
glycolysis. The encoded enzyme catalyzes an important energy-yielding
step in carbohydrate metabolism and during spermatogenesis, this
enzyme may play an important role in regulating the switch between different energy-producing pathways, required for sperm motility (Welch
et al., 2000) The A kinase (PRKA) anchor protein 4 (AKAP4) gene
encodes for a protein, suggested to be a major structural component
of sperm fibrous sheath, which also plays a role in sperm motility
(Turner et al., 2001). The additional two examples include the chromosome 2 open reading frame 57 (C2orf 57) and chromosome 10 open
reading frame 40 (C10orf40), both genes of unknown function and only
evidence of existence at the transcript level.
Genes expressed in Sertoli and Leydig cells
The interaction between germinal cells and Sertoli cells is essential both
for the development of testis and for the progress of the spermatogenesis (Griswold, 1998). Of the highly testis-enriched genes in testis, only
a few were specifically localized to Sertoli or Leydig cells when compared
with the germ cell-specific proteins (Fig. 5). The solute carrier organic
anion transporter family, member 6A1 (SLCO6A1) with currently only
evidence at transcript level has previously been reported to be strongly
expressed in the human testis but not specifically in Sertoli cells. For the
remaining examples of genes we found to be associated with Sertoli cells,
the expression did not seem to be restricted to this particular cell type.
For example, fetal and adult testis expressed (FATE1) protein is shown to
be expressed in spermatogonia, primary spermatocytes and Sertoli cells
which also has been reported previously (Yang et al., 2005). Another
example was defensin, beta 119 (DEFB119), a member of the betadefensin family assumed to be involved in antibacterial activity. The
secreted DEFB119 protein, previously reported to be expressed in the
testis and epididymis, showed an expression pattern suggesting that
this protein also is expressed in Sertoli cells (Radhakrishnan et al.,
2005). Only one of the highly enriched genes in testis, the insulin-like 3
(Leydig cell) (INSL3) gene, showed specific expression in the androgenproducing Leydig cells. The INSL3 gene encodes a well-known marker of
Leydig cells, suggested to be involved in the development of urogenital
tract and intra-abdominal testicular descent (Burkhardt et al., 1994).
Discussion
Here we demonstrate the power of combining genome-wide RNA and
protein analysis to define the tissue-specific proteome of human testis.
The analysis shows that 77% of all putative protein coding genes
are expressed in the testis, which is more than in any of the other
26 tissues analyzed. In addition, the testis also has by far more
tissue-enriched genes than any other human tissue with .1000 genes
showing an at least 5-fold higher expression. One explanation to the
large abundance of enriched genes in testis, when compared with
other tissue types, could be that testis harbors the only male cell type
that undergoes meiosis to generate haploid cells. Meiosis involves
several specific events such as reduction of the number of chromosomes,
condensation of nucleus and removal of excess cytoplasm within the
sperm. In addition, the sperm cells can be described as an independent
unicellular organism that has acquired unique morphological features and
structures that are necessary for sperm motility and fertilization of the
egg. It is not unlikely that many of the genes here identified as testis specific could show shared expression with the cells developing into the
haploid egg in females. An important extension of this could therefore
be to include additional specialized tissue types from females including
an enrichment of oocytes. More work is needed to explore the expression pattern of the development of the egg in females to establish the differences between the development of the haploid cell in males and
females.
It is noteworthy that the term “testis specific” is used for the genes with
enriched or enhanced expression in testis, but in reality a vast majority of
these genes have transcripts present also in other tissues. A tissuespecific score has therefore been obtained for all genes to define the
ratio of mRNA levels in testis when compared with the maximum
mRNA levels in all other tissues. In this analysis we have focused on
the highly enriched genes with a ratio over 50 and found that most of
these were expressed in germ cells of the seminiferous epithelium,
whereas the two somatic cell types in testis, Leydig and Sertoli cells,
only expressed a few highly enriched genes. This supports the assumption that the cell types specifically involved in spermatogenesis account
for the large amount of specific genes. It is therefore likely that the expression profiles of Sertoli and Leydig cells more resemble other somatic cell
types and hence do not contain as many specific genes as germ cells. For
example, suggested Sertoli cell markers such as the anti-Mullerian
hormone (AMH) (Josso et al., 1977) was also expressed in the brain,
the follicle-stimulating hormone receptor (Heckert and Griswold,
2002) was expressed in the ovary and the inhibin alpha (Merchenthaler
et al., 1987) was expressed in the adrenal gland, ovary and placenta
and accordingly these genes were categorized as group enriched or
486
enhanced in testis. It should also be pointed out that genes with low levels
of expression may still exist with a specific expression in Sertoli or Leydig
cells, but given that 90% of the mRNA from testicular tissue is derived
from germ cells, such genes may not readily be detected. A more
in-depth analysis of the testis genes detected in the categories moderately enriched, group enriched and enhanced would be of interest to generate a more complete view of expression patterns in testicular tissue.
Altogether the tissue-specific genes, including testis-enriched, groupenriched and testis-enhanced genes, constitute 10% of all the human
genes and 14% of the transcript pool. An analysis of the most expressed
genes in testis also support that few of the testis-specific genes are
expressed at high mRNA levels. Despite the high number of tissuespecific genes in testis, an in-depth analysis of the mRNA pool derived
from testicular tissue demonstrates that 80% of the transcripts are
encoded by genes expressed in all 27 tissues (‘house-keeping’) (Fagerberg et al., 2014), well consistent with that a large fraction of all expressed
proteins is shared between different tissues (Pontén et al., 2009). The list
of all these genes with elevated expression in testis can be downloaded
from the Human Protein Atlas portal (www.proteinatlas.org) and are
also provided as Supplementary data, Table SIV.
An interesting category of proteins is the genes with shared elevated
expression with at least another tissue (group enriched). The network
analysis demonstrates that the tissues with most shared tissue-enriched
genes are the brain (n ¼ 48) and the lung (n ¼ 33). A functional analysis
shows that several of the genes shared with the lung are involved in functions involving motility of cilia and sperm, while the genes shared with the
brain show less enrichment for any specific functions. A subset of genes
expressed in the testis and brain are RNA-binding proteins evidently
involved in the regulation of transcription and another subset are
expressed during development, such as the AMH gene that encodes
a growth factor involved in male sexual differentiation by causing the
regression of Mullerian ducts.
It has earlier been suggested that tissue-specific genes are overrepresented for genes coding for secreted and transmembrane spanning proteins, summarized as the inside –outside rule (Ramsköld et al., 2009;
Fagerberg et al., 2010). In contrast, the analysis of the testis-specific
proteins shows that a large majority (76%) of the tissue-specific genes
with known subcellular localization are intracellular. This suggests that
the inside –outside rule for tissue specificity does not apply for the
testis and that most of the testis-specific functions are expressed inside
the cell which is not surprising since meiosis, histone replacement and
sperm differentiation mainly reflect intracellular events.
Combining the data from previously well-characterized testis-specific
proteins with present data regarding subcompartment localization in
testis suggests that approximately half of the testis-specific genes are
expressed in late stage spermatogenesis, sperm (30%) and spermatids
(21%), whereas only a few are expressed in earlier stages, spermatogonia
(3%) and spermatocytes (4%). Interestingly, a majority of the genes
analyzed here with previous unknown knowledge regarding cell-type
localization were expressed in early stage spermatogenesis. A substantial
subset of these proteins belong to the family of cancer testis antigens. At
present .100 cancer testis antigens genes have been identified and a
similar study including data from cancer tissues could be of interest to
identify additional cancer testis antigens. These genes are important
since they are restricted to germ cells in normal tissues, but frequently
expressed in various cancer tissues. In addition, the corresponding
proteins appear immunogenic and recognized by the immune system
Djureinovic et al.
of cancer patients and thus believed to be of use in the development
of antigen-specific cancer vaccines (van der Bruggen et al., 1991; Chen
et al., 1997).
In conclusion, the analysis has identified genes of interest to study
testis biology. These genes are interesting starting points for studying
mechanisms for infertility and basic cell biology. In this work we have
focused our protein analysis on genes with corresponding unambiguously staining antibodies and further analysis is needed for the identified
testis-specific genes not yet characterized on the protein level. The
ultimate aim of such an effort should be to provide a complete map of
expression patterns of all genes expressed in a tissue-specific manner.
Supplementary data
Supplementary data are available at http://molehr.oxfordjournals.org/.
Acknowledgements
Pathologists and staff at the Department of Clinical Pathology, Uppsala
University Hospital are greatly acknowledged for providing the tissues
used in the study, in particular, the authors would like to thank Simin
Tahmasebpoor for excellent help with preparing frozen tissues for
RNA extraction. We thank Christer Höög for valuable comments on
the manuscript. The authors also wish to thank the staff of the Human
Protein Atlas project in both Sweden and India for their efforts generating
the Human Protein Atlas.
Authors’ roles
F.P. and M.U. conceived and designed the study. F.P., M.U. and D.D.
wrote the paper. L.F. and B.H. performed the bioinformatics analysis.
D.D. and C.L. performed the antibody-based protein profiling and
data comparison. A.D. prepared the RNA samples.
Funding
This work was supported by the Knut and Alice Wallenberg Foundation.
Conflict of interest
None declared.
References
Anders S, Huber W. Differential expression analysis for sequence count data.
Genome Biol 2010;11:R106.
Ashburner M, Ball CA, Blake JA, Botstein D, Butler H, Cherry JM, Davis AP,
Dolinski K, Dwight SS, Eppig JT et al. Gene ontology: tool for the unification
of biology. The Gene Ontology Consortium. Nat Genet 2000;25:25 – 29.
Brinkmann U, Vasmatzis G, Lee B, Yerushalmi N, Essand M, Pastan I. PAGE-1,
an X chromosome-linked GAGE-like gene that is expressed in normal and
neoplastic prostate, testis, and uterus. Proc Natl Acad Sci USA 1998;
95:10757 – 10762.
Burkhardt E, Adham IM, Brosig B, Gastmann A, Mattei MG, Engel W.
Structural organization of the porcine and human genes coding for a
Leydig cell-specific insulin-like peptide (LEY I-L) and chromosomal
localization of the human gene (INSL3). Genomics 1994;20:13 – 19.
The human testis-specific proteome
Chadwick BP, Mull J, Helbling LA, Gill S, Leyne M, Robbins CM, Pinkett HW,
Makalowska I, Maayan C, Blumenfeld A et al. Cloning, mapping, and
expression of two novel actin genes, actin-like-7A (ACTL7A) and
actin-like-7B (ACTL7B), from the familial dysautonomia candidate
region on 9q31. Genomics 1999;58:302– 309.
Chen YT, Scanlan MJ, Sahin U, Türeci O, Gure AO, Tsang S, Williamson B,
Stockert E, Pfreundschuh M, Old LJ. A testicular antigen aberrantly
expressed in human cancers detected by autologous antibody screening.
Proc Natl Acad Sci USA 1997;94:1914– 1918.
De Rooij DG, Russel LD. All you wanted to know about spermatogonia but
were afraid to ask. J Androl 2000;21:776 – 798.
Eden E, Navon R, Steinfeld I, Lipson D, Yakhini Z. GOrilla: a tool for
discovery and visualization of enriched GO terms in ranked gene lists.
BMC Bioinformatics 2009;10:48.
Fagerberg L, Jonasson K, von Heijne G, Uhlén M, Berglund L. Prediction of the
human membrane proteome. Proteomics 2010;10:1141 –1149.
Fagerberg L, Hallström BM, Oksvold P, Kampf C, Djureinovic D, Odeberg J,
Habuka M, Tahmasebpoor S, Danielsson A, Edlund K et al. Analysis of
the human tissue-specific expression by genome-wide integration of
transcriptomics and antibody-based proteomics. Mol Cell Proteomics
2014;13:397 – 406.
Flicek P, Amode MR, Barrell D, Beal K, Brent S, Carvalho-Silva D, Clapham P,
Coates G, Fairley S, Fitzgerald S et al. Ensembl 2012. Nucleic Acids Res 2012;
40(Database issue): D84 –D90.
Griswold MD. The central role of Sertoli cells in spermatogenesis. Semin
Cell Dev Biol 1998;9:411 – 416.
Hawthorne SK, Goodarzi G, Bagarova J, Gallant KE, Busanelli RR, Olend WJ,
Kleene KC. Comparative genomics of the sperm mitochondria-associated
cysteine-rich protein gene. Genomics 2006;87:382 – 391.
Hebenstreit D, Fang M, Gu M, Charoensawan V, van Oudenaarden A,
Teichmann SA. RNA sequencing reveals two major classes of gene
expression levels in metazoan cells. Mol Syst Biol 2011;7:497.
Heckert LL, Griswold MD. The expression of the follicle-stimulating
hormone receptor in spermatogenesis. Recent Prog Horm Res 2002;
57:129– 148.
Herr JC, Flickinger CJ, Homyk M, Klotz K, John E. Biochemical and
morphological characterization of the intra-acrosomal antigen SP-10
from human sperm. Biol Reprod 1990;42:181 – 193.
Hoog C. Expression of a large number of novel testis-specific genes during
spermatogenesis coincides with the functional reorganization of the
male germ cell. Int J Dev Biol 1995;39:719 – 726.
Jørgensen A, Nielsen JE, Jensen MB, Græm N, Rajpert-De Meyts E.
Analysis of meiosis regulators in human gonads: a sexually dimorphic
spatio-temporal expression pattern suggests involvement of DMRT1 in
meiotic entry. Mol Hum Reprod 2012;18:523 –534.
Josso N, Picard JY, Tran D. The anti-Mullerian hormone. Birth Defects Orig
Artic Ser 1977;13:59 – 84.
Kampf C, Olsson I, Ryberg U, Sjöstedt E, Pontén F. Production of tissue
microarrays, immunohistochemistry staining and digitalization within the
human protein atlas. J Vis Exp 2012. doi: 10.3791/3620.
Kerr SM, Vambrie S, McKay SJ, Cooke HJ. Analysis of cDNA sequences from
mouse testis. Mamm Genome 1994;5:557 – 565.
Kierszenbaum AL, Tres LL. Structural and transcriptional features of the
mouse spermatid genome. J Cell Biol 1975;65:258– 270.
Liggins AP, Brown PJ, Asker K, Pulford K, Banham AH. A novel diffuse large
B-cell lymphoma-associated cancer testis antigen encoding a PAS domain
protein. Br J Cancer 2004;91:141 – 149.
Matson CK, Murphy MW, Griswold MD, Yoshida S, Bardwell VJ,
Zarkower D. The mammalian doublesex homolog DMRT1 is a
transcriptional gatekeeper that controls the mitosis versus meiosis
decision in male germ cells. Dev Cell 2010;19:612 – 624.
487
Merchenthaler I, Culler MD, Petrusz P, Negro-Vilar A. Immunocytochemical
localization of inhibin in rat and human reproductive tissues. Mol Cell
Endocrinol 1987;54:239– 243.
Mortazavi A, Williams BA, McCue K, Schaeffer L, Wold B. Mapping and
quantifying mammalian transcriptomes by RNA-Seq. Nat Methods 2008;
5:621 – 628.
Ottolenghi C, Fellous M, Barbieri M, McElreavey K. Novel paralogy relations
among human chromosomes support a link between the phylogeny of
doublesex-related genes and the evolution of sex determination.
Genomics 2002;79:333 – 343.
Pawlak A, Toussaint C, Lévy I, Bulle F, Poyard M, Barouki R, Guellaën G.
Characterization of a large population of mRNAs from human testis.
Genomics 1995;26:151 – 158.
Penttilä TL, Yuan L, Mali P, Höög C, Parvinen M. Haploid gene expression:
temporal onset and storage patterns of 13 novel transcripts during rat
and mouse spermiogenesis. Biol Reprod 1995;53:499 – 510.
Pontén F, Jirström K, Uhlen M. The Human Protein Atlas—a tool for
pathology. J Pathol 2008;216:387– 393.
Pontén F, Gry M, Fagerberg L, Lundberg E, Asplund A, Berglund L, Oksvold P,
Björling E, Hober S, Kampf C. A global view of protein expression in human
cells, tissues, and organs. Mol Syst Biol 2009;5:337.
Pontén F, Schwenk JM, Asplund A, Edqvist PH. The Human Protein Atlas as a
proteomic resource for biomarker discovery. J Intern Med 2011;
270:428– 446.
Radhakrishnan Y, Hamil KG, Yenugu S, Young SL, French FS, Hall SH.
Identification, characterization, and evolution of a primate beta-defensin
gene cluster. Genes Immun 2005;6:203 – 210.
Ramsköld D, Wang ET, Burge CB, Sandberg R. An abundance of ubiquitously
expressed genes revealed by tissue transcriptome sequence data. PLoS
Comput Biol 2009;5:e1000598.
RCoreTeam. A Language and Environment for Statistical Computing. Vienna,
Austria: R Foundation for Statistical Computing, 2013.
Reijo RA, Dorfman DM, Slee R, Renshaw AA, Loughlin KR, Cooke H,
Page DC. DAZ family proteins exist throughout male germ cell
development and transit from nucleus to cytoplasm at meiosis in
humans and mice. Biol Reprod 2000;63:1490 – 1496.
Russell LD, Ettlin RA, Sinha Hikim AP, Clegg ED. Histological and
Histopathological Evaluation of the Testis. Clearwater, FL, USA: Cache
River Press, 1990.
Shannon P, Markiel A, Ozier O, Baliga NS, Wang JT, Ramage D, Amin N,
Schwikowski B, Ideker T. Cytoscape: a software environment for
integrated models of biomolecular interaction networks. Genome Res
2003;13:2498 – 2504.
Tapparel C, Reymond A, Girardet C, Guillou L, Lyle R, Lamon C, Hutter P,
Antonarakis SE. The TPTE gene family: cellular expression, subcellular
localization and alternative splicing. Gene 2003;323:189 – 199.
Teng X, Yang J, Xie Y, Ni Z, Hu R, Shi L, Lin Z, Hu L, Zhao G, Ding X et al. A
novel spermatogenesis-specific uPAR gene expressed in human and
mouse testis. Biochem Biophys Res Commun. 2006;342:1223 – 1227.
Trapnell C, Williams BA, Pertea G, Mortazavi A, Kwan G, van Baren MJ,
Salzberg SL, Wold BJ, Pachter L. Transcript assembly and quantification
by RNA-Seq reveals unannotated transcripts and isoform switching
during cell differentiation. Nat Biotechnol 2010;28:511 – 515.
Turner RM, Musse MP, Mandal A, Klotz K, Jayes FC, Herr JC, Gerton GL,
Moss SB, Chemes HE. Molecular genetic analysis of two human sperm
fibrous sheath proteins, AKAP4 and AKAP3, in men with dysplasia of
the fibrous sheath. J Androl 2001;22:302 – 315.
Uhlén M, Björling E, Agaton C, Szigyarto CA, Amini B, Andersen E,
Andersson AC, Angelidou P, Asplund A, Asplund C et al. A human
protein atlas for normal and cancer tissues based on antibody
proteomics. Mol Cell Proteomics 2005;4:1920– 1932.
488
Uhlen M, Oksvold P, Fagerberg L, Lundberg E, Jonasson K, Forsberg M,
Zwahlen M, Kampf C, Wester K, Hober S et al. Towards a knowledgebased Human Protein Atlas. Nat Biotechnol 2010;28:1248 – 1250.
van der Bruggen P, Traversari C, Chomez P, Lurquin C, De Plaen E,
Van den Eynde BJ, Knuth A, Boon T. A gene encoding an antigen
recognized by cytolytic T lymphocytes on a human melanoma. Science
1991;254:1643 – 1647.
Wang Z, Gerstein M, Snyder M. RNA-Seq: a revolutionary tool for
transcriptomics. Nat Rev Genet 2009;10:57 – 63.
Welch JE, Brown PL, O’Brien DA, Magyar PL, Bunch DO, Mori C,
Eddy EM. Human glyceraldehyde 3-phosphate dehydrogenase-2 gene
Djureinovic et al.
is expressed specifically in spermatogenic cells. J Androl 2000;
21:328 – 338.
Wouters-Tyrou D, Martinage A, Chevaillier P, Sautière P. Nuclear basic
proteins in spermiogenesis. Biochimie 1998;80:117 – 128.
Wykes SM, Nelson JE, Visscher DW, Djakiew D, Krawetz SA. Coordinate
expression of the PRM1, PRM2, and TNP2 multigene locus in human
testis. DNA Cell Biol 1995;14:155 – 161.
Yang XA, Dong XY, Qiao H, Wang YD, Peng JR, Li Y, Pang XW, Tian C,
Chen WF. Immunohistochemical analysis of the expression of FATE/
BJ-HCC-2 antigen in normal and malignant tissues. Lab Invest 2005;
85:205 – 213.