Physiology of Testicular Function

Abstract

The testes produce the male gametes and the male sexual hormones (androgens). The term spermatogenesis describes and includes all the processes involved in the production of gametes, whereas steroidogenesis refers to the enzymatic reactions leading to the production of male steroid hormones. Spermatogenesis and steroido-genesis
take place in two compartments morphologically and functionally distinguishable from each other. These are the tubular compartment,
consisting of the seminiferous tubules (tubuli seminiferi) and the interstitial compartment (interstitium) between the seminiferous tubules (Figs. 2.1 and 2.2). Although anatomically separate, both compartments are closely connected
with each other. For quantitatively and qualitatively normal production of sperm the integrity of both compartments is necessary.
The function of the testis and thereby also the function of its compartments are governed by the hypothalamus and the pituitary
gland (endocrine regulation). These endocrine effects are mediated and modulated at the testicular level by local control mechanisms (paracrine and autocrine factors).

Full text

E. Nieschlag et al. (eds.), Andrology,
DOI: 10.1007/978-3-540-78355-8_2, © Springer-Verlag Berlin Heidelberg 2010
11
Physiology of Testicular Function
Gerhard F. Weinbauer, Craig Marc Luetjens,
Manuela Simoni, and Eberhard Nieschlag
2
2.1 Functional Organization
of the Testis
The testes produce the male gametes and the male sex-
ual hormones (androgens). The term spermatogenesis
describes and includes all the processes involved in the
production of gametes, whereas steroidogenesis refers
to the enzymatic reactions leading to the production of
male steroid hormones. Spermatogenesis and steroido-
genesis take place in two compartments morphologi-
cally and functionally distinguishable from each other.
These are the tubular compartment, consisting of the
seminiferous tubules (tubuli seminiferi) and the inter-
stitial compartment (interstitium) between the seminif-
erous tubules (Figs. 2.1 and 2.2). Although anatomically
separate, both compartments are closely connected with
each other. For quantitatively and qualitatively normal
production of sperm the integrity of both compartments
is necessary. The function of the testis and thereby also
the function of its compartments are governed by the
hypothalamus and the pituitary gland (endocrine regu-
lation). These endocrine effects are mediated and mod-
ulated at the testicular level by local control mechanisms
(paracrine and autocrine factors).
2.1.1 Interstitial Compartment
The most important cells of this compartment are the
Leydig cells. These cells are the source of testicular
testosterone and of insulin-like factor 3 (INSL3). Aside
from Leydig cells, the interstitial compartment also
contains immune cells, blood and lymph vessels,
nerves, fi broblasts and loose connective tissue. In
experimental animals this compartment comprises
G. F. Weinbauer (*)
Research and Safety Assessment, Covance
Laboratories GmbH, Kesselfeld 29,
D-48169 Münster,Germany
e-mail: gerhard.weinbauer@covance.com
Contents
2.1 Functional Organization of the Testis ..................... 11
2.1.1 Interstitial Compartment ................................. 11
2.1.2 Tubular Compartment ..................................... 14
2.2 Hormonal Control of Testicular Function .............. 21
2.2.1 Functional Organization
of the Hypothalamo-Pituitary System ............ 21
2.2.2 The Kisspeptin-GPR54 System ...................... 22
2.2.3 GnRH .............................................................. 24
2.2.4 Gonadotropins ................................................. 28
2.2.5 Endocrine Regulation and Relative
Importance of LH and FSH
for Spermatogenesis ........................................ 31
2.2.6 Local Regulation of Testicular Function ......... 33
2.3 Testicular Descent ..................................................... 36
2.4 Vascularization, Temperature Regulation
and Spermatogenesis ................................................ 37
2.5 Immunology of the Testis.......................................... 37
2.6 Testicular Androgens ................................................ 39
2.6.1 Synthesis of Androgens .................................. 40
2.6.2 Testosterone Transport in Blood ..................... 42
2.6.3 Extratesticular Metabolism
of Testosterone ................................................ 43
2.6.4 Mechanism of Androgen Action ..................... 44
2.6.5 Biological Actions of Androgens .................... 49
2.6.6 Androgen Secretion and Sexual
Differentiation ................................................. 52
References ........................................................................... 54

12 G. F. Weinbauer et al.
about 2.6% of the total testicular volume. In the human
testis the interstitial compartment represents about
12–15% of the total testicular volume, 10–20% of
which is occupied by Leydig cells. Human testes con-
tain approximately 200 × 106 Leydig cells.
2.1.1.1 Leydig Cells
These cells were fi rst described in 1850 by Franz Leydig
(1821–1908). Leydig cells produce and secrete the most
important male sexual hormone, testosterone. From the
developmental, morphological and functional view-
point different types of cells can be distinguished: stem
Leydig cells as founder cell, progenitor Leydig cells as
a committed stem cell, fetal Leydig cells as a terminally
differentiated cell in the fetus, and adult Leydig cells as
the terminally differentiated Leydig cell (Ge and Hardy
2007). Fetal Leydig cells become neonatal Leydig cells
at birth and degenerate thereafter or regress into imma-
ture Leydig cells (Prince 2007). Fetal Leydig cells pro-
duce testosterone. Immature Leydig cells that mainly
produce androstane-3α, 17β-diol instead of testoster-
one have also been described.
Adult Leydig cells are rich in smooth endoplasmic
reticulum and mitochondria with tubular cristae. These
physiological characteristics are typical for steroid-
producing cells and are very similar to those found in
other steroidogenic cells, such as those in the adrenal
gland and in the ovary. Other important cytoplasmic
components are lipofuscin granules, the fi nal product
of endocytosis and lyosomal degradation, and lipid
Fig. 2.1 Section of an entire
human testis cut transversally.
The preparation also includes
parts of the efferent ducts and
the epididymis. The lobular
architecture of the testis is
evident (Courtesy of Prof. Dr.
A.F. Holstein, Institute of
Anatomy, University of
Hamburg)

2 Physiology of Testicular Function 13
Fig. 2.2 (a) Schematic representation of the architecture of the
human seminiferous epithelium. Note that the tubular wall is
composed of several layers of peritubular cells (PT) and a basal
lamina (BL). RB = Residual Body, LS = Late/elongating and
elongated spermatids, ES = Early/round spermatids, P =
Spermatocytes, Ad = A-dark-type spermatogonia (testicular
stem cells), Ap = A pale-type spermatogonia; B = B-type sper-
matogonia, SC = Sertoli cells, JC = junctional complexes consti-
tuting the blood-testis barrier built by interconnected Sertoli
cells. Modifi ed from Ross (1985)
(b) Depicts the kinetics of the human seminiferous epithelium.
Ap spermatogonia are the progenitor stem cells that enter the
spermatogenic cycle (color-coded). All descendents of this pro-
genitor cell represent a single clone of germ cells. Ad = A-dark-
type spermatogonia, B = B-type spermatogonia, Pl = prelepototene
spermatocyte; L = leptotene spermatocyte; Z = zygotene sper-
matocyte; D = diplotere spermatocyte. It takes 4–4.6 cycles
(“generation” on the y-axis, denoted as Start and End) until a
sperm (Sd) has developed from a progenitor cell (Sa, Sb, Sc) 2°
= 2nd meitotic division. (Modifi ed from Amann 2008)
a
b
Ad SC Ap Basal membrane BPT Sertoli-cell

14 G. F. Weinbauer et al.
droplets, in which the preliminary stages of testoster-
one synthesis take place. Special formations, called
Reinke’s crystals, are often found in the adult Leydig
cells. These are probably subunits of globular proteins
whose functional meaning is not known. The prolifera-
tion rate of the Leydig cells in the adult testis is rather
low and is infl uenced by LH. The ontogeny of Leydig
cells is not entirely clear and mesonephros, neural crest
and coelomic sources have been involved. In the adult
testis, Leydig cells develop from perivascular and peri-
tubular mesenchymal-like cells and the differentiation
of these cells into Leydig cells is induced by LH but
also by growth factors and differentiation factors
derived from Sertoli cells.
2.1.1.2 Macrophages. Lymphocytes
and Nerve Fibers
Besides Leydig cells, the interstitial compartment also
contains cells belonging to the immune system: mac-
rophages and lymphocytes. For every 10–50 Leydig
cells one macrophage is to be found. The macrophages
probably infl uence the function of the Leydig cells, in
particular their proliferation, differentiation and ste-
roid production, through the secretion of cytokines.
Macrophages secrete stimulators and inhibitors of ste-
roidogenesis. Proinfl ammatory cytokines, reactive
oxygen species, nitric oxide and prostaglandins can
inhibit Leydig cell function (Hales 2007). There is evi-
dence for an involvement of neurotransmitters and
related signalling factors during regulation of Leydig
cell function (Mayerhofer et al. 1999). The immuno-
logical meaning of these cells for testicular physiology
will be discussed under Sect. 2.5.
2.1.2 Tubular Compartment
Spermatogenesis takes place in the tubular compart-
ment. This compartment represents about 60–80% of
the total testicular volume. It contains the germ cells
and two different types of somatic cells, the peritubu-
lar cells and the Sertoli cells. The testis is divided by
septa of connective tissue into about 250–300 lobules
(Fig. 2.1), each one containing 1–3 highly convoluted
seminiferous tubules. Overall, the human testis con-
tains about 600 seminiferous tubules. The length of
individual seminiferous tubules is about 30–80 cm.
Considering an average number of about 600 seminif-
erous tubules per testis and an average length of the
tubuli seminiferi of about 60 cm each, the total length
of the tubuli seminiferi is about 360 m per testis, i.e.,
720 m of seminiferous epithelium per man.
2.1.2.1 Peritubular Cells
The seminiferous tubules are covered by a lamina pro-
pria, which consists of a basal membrane, a layer of
collagen and the peritubular cells (myofi broblasts).
These cells are stratifi ed around the tubulus and form
up to concentrical layers that are separated by collagen
layers (Fig. 2.1). These characteristics differentiate the
human testicle from the majority of the other mam-
mals, whose seminiferous tubules are surrounded only
by 2–4 layers of myofi broblasts. Peritubular cells pro-
duce several factors that are involved in cellular con-
tractility: panactin, desmin, gelsolin, smooth muscle
myosin and actin (Holstein et al. 1996). These cells
also secrete extracellular matrix and factors typically
expressed by connective tissue cells: collagen, lami-
nin, vimentin, fi bronectin, growth factors, fi broblast
protein and adhesion molecules (Albrecht et al. 2006;
Schell et al. 2008). The latter work established a human
peritubular cell culture system demonstrating secre-
tion of nerve growth factor and pro-infl ammatory mol-
ecules, e.g., Il-1b and cyclooxygenase-2 under the
infl uence of TNF-a (Schell et al. 2008). Myofi broblasts
are poorly differentiated myocytes with the capacity of
spontaneous contraction. Mature sperm are transported
towards the exit of the seminiferous tubules by con-
traction of these cells and several regulators of cell
contractions are reported, e.g., oxytocin, oxytocin-like
substances, prostaglandins, androgenic steroids,
endothelins, endothe lin converting enzymes and
endothelin receptors. Peritubular contractility is medi-
ated by endothelin and this effect is modulated by the
relaxant peptide adrenomedullin produced by Sertoli
cells (Romano et al. 2005). Mice with selective peritu-
bular cell androgen receptor defi ciency revealed
defects in contractility-related genes, e.g., endothe-
lin-1 and endothelin receptor A and B, adrenomedullin
receptor and vasopressin receptor 1a (Zhang et al.
2006).
Disturbances of testicular function and decreased or
absent spermatogenic activity are associated with a

2 Physiology of Testicular Function 15
thickening of the layer of collagen fi bres and of the
material present between the peritubular cells. When
this is the case, the tubular wall becomes fi brotic or –
based on the histological appearance – hyalinized. The
decrease of testicular volume involves folding of the
wall along the length of the tubuli seminiferi, thereby
causing an enlargement of the tubular diameter. This
becomes particularly evident when fl uid is injected
into regressed seminiferous tubules. Tubular diameter
increases and tubular wall thickness decreases (Schlatt
et al. 1999). An interaction between testicular mast
cells and peritubular cells leading to fi brotic changes
of the seminiferous tubular wall has been suggested
(Albrecht et al. 2006). Peritubular and interstitial fi bro-
sis incidence correlated progressively with spermato-
genic damage in testis from vasectomized men (Raleigh
et al. 2004).
2.1.2.2 Sertoli Cells
Sertoli cells are somatic cells located within the germi-
nal epithelium. In adulthood these cells are mitotically
inactive. They are named after Enrico Sertoli (1842–
1910), the Italian scientist who fi rst described these
cells in 1865 and, due to their prominent cytoplasmatic
projections and ramifi cations called them “cellulae
ramifi cate”. These cells are located on the basal mem-
brane and extend to the lumen of the tubulus semi-
niferus and, in a broad sense, can be considered as the
supporting structure of the germinal epithelium.
Along the cell body, extending over the entire height of
the germinal epithelium, all morphological and physi-
ological differentiation and maturation of the germinal
cell up to the mature sperm take place. Special ecto-
plasmic structures sustain alignment and orientation of
the sperm during differentiation. About 35–40% of the
volume of the germinal epithelium is represented by
Sertoli cells. The intact testis with complete spermato-
genesis contains 800–1200 × 106 Sertoli cells
(Zhengwei et al. 1998a) or approximately 25 × 106
Sertoli cells per gram testis (Raleigh et al. 2004).
Sertoli cells synthesize and secrete a large variety of
factors: proteins, cytokines, growth factors, opioids,
steroids, prostaglandins, modulators of cell division
etc. The morphology of Sertoli cells is strictly related
to their various physiological functions. Cytoplasm
contains endoplasmic reticulum both of the smooth
(steroid synthesis) and rough type (protein synthesis),
a prominent Golgi apparatus (elaboration and trans-
port of secretory products), lysosomal granules
(phagocytosis) as well as microtubuli and intermedi-
ate fi laments (adapation of the cell shape during the
different phases of germ cell maturation). It is gener-
ally assumed that Sertoli cells coordinate the spermato-
genic process topographically and functionally. On the
other hand, more recent data support the contention
that germ cells control Sertoli cell functions. At least
the time pattern of germ cell transitions and develop-
ment during the spermatogenic cycle seem to be auton-
omous as suggested from heterologous germ cell
transplantation studies (Nagano et al. 2001). One sper-
matogenic cycle lasts about 8 days in mice and 12–13
days in rats. Notably, the cycle duration of rat germ
cells transplanted into mouse testis remained 12–13
days whereas that of the host germ cells was main-
tained at 8 days (Franca et al. 1998).
Another important function of Sertoli cells is that
they are responsible for fi nal testicular volume and
sperm production in the adult. Each individual Sertoli
cell is in morphological and functional contact with a
defi ned number of sperm. The number of sperm per
Sertoli cell depends on the species. In men we observe
about 10 germ cells or 1.5 spermatozoa per each Sertoli
cell (Zhengwei et al. 1998a). In comparison, every
macaque monkey Sertoli cell is associated with 22
germ cells and 2.7 sperm (Zhengwei et al. 1997,
1998b). This suggests that within a certain species a
higher number of Sertoli cells results in a greater pro-
duction of sperm and testis size, assuming that all the
Sertoli cells are functioning normally. In contrast, as
determined by fl ow cytometry, testicular cell numbers
were very similar across several primate species, sug-
gesting that testis size is the main determinant of total
germ cell output (Luetjens et al. 2005).
Stereological investigations suggest that the num-
ber of Sertoli cells in men increases until the 15th year
of life. In the prepubertal cynomolgus monkey and the
rhesus monkey, Sertoli cells exhibit little mitotic activ-
ity, whereas some proliferative activity of A-type sper-
matogonia occurs in the quiescent testis. Sertoli cell
proliferation is markedly activated when exposed to
gonadotropin activity (Plant et al. 2005; Schlatt et al.
1995). Both Sertoli cell number and expression of
markers of cell division are stimulated by these hor-
mones. The division of Sertoli cells ends when the fi rst
germ cells undergo meiotic division and Sertoli cells
have built tight junctions between each other, the

16 G. F. Weinbauer et al.
so-called blood-testis-barrier (see Sect. 2.5). Lack of
connexin-43, a predominant gap-junction protein, pre-
vents Sertoli cell maturation associated with continued
division of Sertoli cells and spermatogenic arrest
beyond spermatogonial development (Brehm et al.
2007; Sridharan et al. 2007). Expression of Sertoli
cells markers such as transferrin, androgen-binding
protein and junctional proteins such as N-cadherin,
connexin-43, gelsolin, laminin-γ3, occludin, testin,
nectin, zyxin and vinculin is androgen-dependent
(Zhang et al. 2006). It appears that several of these
components are involved in establishing the blood-
testis-barrier but also in the release of sperm and sub-
sequent remodelling of the Sertoli cell-germ cell
junctions (Yan et al. 2008). In the rat, the experimental
prolongation of the division phase of Sertoli cells, pro-
duced for example by a deprivation of thyroid hor-
mones, results in an increase of testicular weight and
sperm production by about 80%. On the other hand,
the decrease of Sertoli cell numbers such as that pro-
duced by an antimitotic substance leads to a reduction
of testicular volume and sperm production. Patients
with Laron dwarfi sm suffer from a disturbance of thy-
roid function and growth hormone/IGF-I defi ciency,
and often have testicles larger than normal.
Through the production and secretion of tubular
fl uid Sertoli cells create and maintain the patency of
the tubulus lumen. More than 90% of Sertoli cell fl uid
is secreted in the tubular lumen. Special structural ele-
ments of the blood-testis barrier prevent reabsorption
of the secreted fl uid, resulting in pressure that main-
tains the patency of the lumen. Sperm are transported
in the tubular fl uid, the composition of which is known
in detail only in the rat (Setchell 1999). Unlike blood,
the tubular fl uid contains a higher concentration of
potassium ions and a lower concentration of sodium
ions. Other constituents are bicarbonate, magnesium
and chloride ions, inositol, glucose, carnitine, glycero-
phosphorylcholine, amino acids and several proteins.
Therefore, the germ cells are immersed in a fl uid of
unique composition.
The basolateral aspect of neighboring Sertoli cells
comprises membrane specializations forming a band
sealing the cells to each other and obliterating the
intracellular space (occluding tight junctions). The
physiological function of the blood-testis barrier has
been proven in experiments showing that dyes or lan-
thanum applied outside the barrier could diffuse only
up to the tight junctions without reaching the lumen of
the seminiferous tubules. The closure of the blood-
testis barrier coincides with the beginning of the fi rst
meiosis in the germinal cells (preleptotene, zygotene)
and with the arrest of proliferation of Sertoli cells.
Through the blood-testis-barrier the seminiferous epi-
thelium is divided into two regions which are anatomi-
cally and functionally completely different from each
other. Early germ cells are located in the basal region
and the later stages of maturing germ cells in the adlu-
minal region. During their development germ cells
are displaced from the basal to the adluminal compart-
ment. This is accomplished by a synchronized dissolu-
tion and reassembly of the tight junctions above and
below the migrating germ cells.
Two important functions are postulated for the
blood-testis-barrier: the physical isolation of haploid
and thereby antigenic germ cells to prevent recogni-
tion by the immune system (prevention of autoimmune
orchitis, see Sect. 2.5) and the preparation of a special
milieu for the meiotic process and sperm development.
In certain seasonal breeders the opening and closure of
the barrier depends much more on the activity of the
Sertoli cells than on the developmental phase of the
germinal epithelium. The constitution of the blood-
testis-barrier and its selectivity in excluding certain
molecules means that the cells localized in the adlumi-
nal compartment have no direct access to metabolites
deriving from the periphery or from the interstitium.
Therefore, these cells are completely dependent on
Sertoli cells for their maintenance. This “nourishing
function” could be exercised through different mecha-
nisms: selective transport and transcytosis as well as
synthesis and vectorial secretion.
2.1.2.3 Germinal Cells
Spermatogenesis starts with the division of stem cells
and ends with the formation of mature sperm (Figs. 2.3
and 2.4). The various germ cells are arranged in typical
cellular associations within the seminiferous tubules
known as spermatogenic stages (Fig. 2.5) and the
entire spermatogenic process can be divided into four
phases:
1. Mitotic proliferation and differentiation of diploid
germ cells (spermatogonia) (spermatogoniogenesis)
2. Meiotic division of tetraploid germ cells (spermato-
cytes) resulting in haploid germ cells (spermatids)
3. Transformation of spermatids into testicular sperm
(spermiogenesis)
4. Release of sperm from the germinal epithelium into
the tubular lumen (spermiation).

2 Physiology of Testicular Function 17
Spermatogonia lie at the base of the seminiferous
epithelium and are classifi ed as type A and type B
spermatogonia. Two types of A spermatogonia can be
distinguished, originally from a cytological and now
also from a physiological point of view: the Ad (dark)
spermatogonia and the Ap (pale) spermatogonia. The
Ad spermatogonia do not show proliferating activity
under normal circumstances and are believed to divide
only rarely (Ehmcke and Schlatt 2006). These sper-
matogonia are considered to represent testicular stem
cells (Ehmcke et al. 2006). These germ cells, however,
undergo mitosis when the overall spermatogonial popu-
lation is drastically reduced, for example due to radia-
tion (de Rooij 1998). In contrast, the Ap spermatogonia
divide, renew themselves and differentiate into two
B spermatogonia. Detailed studies in non-human pri-
mates led to a revised model for spermatogonial expan-
sion in men (Ehmcke and Schlatt 2006): only Ap
spermatogonia divide and give rise to Ap spermatogonia
(to replenish this cell pool) as well as to B-type sper-
matogonia for further development. In contrast to the
earlier model of Clermont, Ad spermatogonia are not
the source of Ap spermatogonia during regular sper-
matogenic cycles. From B spermatogonia the prelepto-
tene spermatocytes are derived directly before the
beginning of the meiotic division. The latter germ cells
commence DNA synthesis. Mother and daughter cells
remain in close contact with each other through
intercellular bridges (Alastalo et al. 1998). This “clonal”
Spermatogoniogenesis
Spermiogenesis Spermatozoa Spermiation
B
Ad
Ap PL LZ
EP
MP
LP
II
Sa1/2
Spermatids
Sb1 Sb2 Sc Sd1
Sd2
RB
Meiosis
Fig. 2.3 Schematic
representation of all germ
cell types that occur in the
human seminiferous
epithelium. Ap spermatogo-
nia enter the spermatogenic
process (arrow on the cell
indicates direction of germ
cell development). Ad
spermatogonia are believed
to constitute the testicular
stem cells. Ad = A-dark
spermatogonium, Ap =
A-pale spermatogonium, B
= B spermatogonium, Pl =
preleptotene spermatocytes,
L = leptotene spermato-
cytes, EP = early pachytene
spermatocytes, MP = mid
pachytene spermatocytes,
LP = late pachytene
spermatocytes, II = 2nd
meiotic division, RB =
residual body, Sa1 – Sd2 =
developmental stages of
spermatid maturation
SC 1 SC 1
B
SC 2 SC 2
SC 1
ES
RS
ES
RS
SC 2
ES
RS
ES
RS
SC 2
SC1
B
Ap
Ap
Ad
Ad
Fig. 2.4 Schematic representation of the proliferative kinetics of
human gametogenesis. For the sake of clarity, complete develop-
ment of only one spermatogonium is shown. The human testis con-
tains about 1 billion sperm and releases around 25,000 sperm every
minute (Amann 2008). One Ap spermatogonium can be the pro-
genitor for 16 elongated spermatids. Since the human seminiferous
epithelium contains only one generation of B-type spermatogonia,
the fi nal germ cell number produced is lower than in species with
multiple spermatogonial divisions. Ad = A-dark spermatogonium
(testicular stem cells, divides rarely), Ap = A-pale spermatogonium
(self-renewing and progenitor cell for spermatogenesis), B = B
spermatogonium, SC1 = primary spermatocyte, SC2 = secondary
spermatocyte, RS = round spermatid, ES = elongated spermatid

18 G. F. Weinbauer et al.
mode of germ cell development – also confi rmed for
primates (Ehmcke et al. 2005) – is possibly the basis and
at the same time probably the prerequisite for the coor-
dinated maturation of gametes in the seminiferous
epithelium.
Tetraploid germ cells are known as spermatocytes
and go through the different phases of the meiotic divi-
sion. The pachytene phase is characterized by intensive
RNA synthesis. Haploid germ cells, the spermatids,
result from the meiotic division. The meiotic process is
a critical event in gametogenesis, during which recom-
bination of genetic material, reduction of chromosome
number and development of spermatids have to be
accomplished. Secondary spermatocytes are derived
from the fi rst meiotic divison. These germ cells con-
tain a haploid chromosomal set in duplicate form.
During the second meiotic division spermatocytes are
divided into the haploid spermatids. The prophase of
the fi rst meiosis lasts 1–3 weeks, whereas the other
phases of the fi rst meiosis and the entire second meio-
sis are concluded within 1–2 days.
Spermatids are derived from the second meiotic
division and are round mitotically inactive cells which
undergo a remarkable and complicated transformation
leading to the fi nal production of differentiated elon-
gated spermatids and sperm. These processes include
Spermiation
Sd1
Sb1
Sa2Sa1
Sb2 Sc Sc
II
LPLPMPMPEPEP
B B PL L L Z
ApApApApApAp
Ad Ad Ad Ad Ad Ad
VIVIVIIIIII
RBSd2
Fig. 2.5 Representation of
the specifi c stages of
spermatogenesis of the
human testis using the
six-stage system. A tubular
cross-section contains typical
germ cell associations that
are denoted as stages of
spermatogenesis. The six
stages (I–VI) in the human
last altogether 16 days. Since
a spermatogonium has to pass
through minimally four cell
layers, the complete duration
of spermatogenesis in men is
at least 64 days. The
complete duration of the
human spermatogenic
process is still not entirely
clear (Amann 2008).
Ad = A-dark spermatogonium
(testicular stem cells, divides
rarely), Ap = A-pale
spermatogonium (self-renewing
and progenitor cell for
spermatogenesis), B = B
spermatogonium, Pl =
preleptotene spermatocytes,
L = leptotene spermatocytes,
EP = early pachytene
spermatocytes, MP = mid
pachytene spermatocytes,
LP = late pachytene
spermatocytes, II = 2nd
meiotic division,
RB = residual body,
Sa1–Sd2 = developmental
stages of spermatid
maturation

2 Physiology of Testicular Function 19
condensation and structural shaping of the cell nucleus,
the formation of a fl agellum and the expulsion of a
large part of cytoplasm. The overall process is called
spermiogenesis and, from a qualitative point of view,
is identical in all species. It is useful to divide spermio-
genesis into four phases: Golgi, cap, acrosomal and
maturation phases.
During the Golgi phase acrosomal bubbles and
craniocaudal symmetry appear. In the cap phase the
spermatids become elongated and the acrosome
develops, covering the cranial half to two-thirds of
the spermatid. During the fertilization process
enzymes are released by the acrosome, allowing the
sperm to penetrate the egg (see Chap. 3).
In the acrosomal phase the cell nucleus becomes
further condensed and elongation of the cell continues.
During condensation the majority of histones are lost
and gene transcription stops. Nuclear chromatin is now
extremely condensed, implying that the proteins nec-
essary for spermiogenesis have to be transcribed before
this timepoint and justifying the fi nding of RNA spe-
cies with very long half-life and RNA binding proteins.
This is the case for transition proteins and protamines.
The mRNA translational control mechanisms are just
being unravelled and RNA-binding proteins seem to
play an important role. The fl agellum is now mature.
The principal event during the maturation phase of
the spermatids is the extrusion of the rest of the cyto-
plasm as the so-called residual body. Residual bodies
are phagocytosed by Sertoli cells and have a regulatory
role. Elongated spermatids and their residual bodies
infl uence the secretory function of Sertoli cells (pro-
duction of tubular fl uid, inhibin, androgen-binding
protein and interleukin-1 and 6). In parallel with deg-
radation of the residual bodies, a new spermatogenic
cycle begins.
The release of sperm into the tubular lumen is des-
ignated as spermiation. This event is infl uenced by
plasminogen activators and possibly also by thimet oli-
gopeptidases. This process can be particularly affected
by hormonal modifi cations, temperature and toxins.
The reasons for this sensitivity are, however, not yet
known. Sperm that are not released are phagocytosed
by Sertoli cells. Round and elongated spermatids
already contain all the information necessary for fertil-
ization; since introduction of intracytoplasmatic injec-
tion of testicular sperm and even round spermatids it
has become possible to induce pregnancies success-
fully (see Chap. 3 for details).
2.1.2.4 Kinetics of Spermatogenesis
The complex process of division and differentiation of
germ cells follows a precise pattern. All germ cells
pass through several stages characterized by particular
cellular associations. Recognizing that acrosome
development is stage-dependent was crucial for the
understanding of germ cell maturation. The number of
stages of spermatogenesis depends on morphological
criteria. For the rat 14 stages (I–XIV) are used and 12
stages (I–XII) in macaques, For men, originally 12
spermatid maturation steps were described but a six-
stage (I–VI) approach is currently used. More recently,
the six-stage system has also been applied to Old World
and New World primate species (Wistuba et al. 2003).
The succession of all stages along time is called the
spermatogenic cycle.
The duration of the spermatogenic cycle depends
on the animal species and lasts between 8–17 days
in mammals. One human spermatogenic cycle
requires 16 days.
For the development and differentiaton of an Ap
spermatogonium into a mature sperm at least four
spermatogenic cycles are necessary. It can be deduced
that the overall duration of spermatogenesis is cal-
culated as around 50 days in the rat, 37–43 days in
different monkey species and at least 64 days in man.
It must be pointed out, however, that a recent review
recommends 74 days by including time for sper-
matogonial renewal (Amann 2008). Investigations car-
ried out in the 1960s led to the conclusion that the
duration of spermatogenesis is genetically determined,
does not vary throughout life and cannot be infl uenced
experimentally. However, many indirect experimental
fi ndings oppose this hypothesis. For example, the fi rst
spermatogenic cycle during puberty proceeds faster
than in the adult age. It has also been demonstrated in
the rat that the duration of germ cell maturation can
actually be manipulated by exogenous factors. In con-
trast, endocrine factors do not alter the duration of
spermatogenic cycles (Aslam et al. 1999).
The spermatogenetic stages appear well orches-
trated, not only in time but also in space. In the rat,
serial transversal sections through the seminiferous
tubules show that stage I is always followed by stage
II, stage III always by stage IV and so on. This is
known as the spermatogenic wave. In contrast, in the

20 G. F. Weinbauer et al.
entire human testis and in parts or whole testis of vari-
ous monkey species, tubular sections show different
stages simultaneously (Fig. 2.6). While this was ini-
tially considered to be an irregular arrangement, quan-
titative analysis of the germ cell population suggested
a helical topography of spermatogenic stages with sev-
eral helices being spaced apart with spermatogonia at
their basis and elongated spermatids at their apical
part (Schulze and Rehder 1984; Zannini et al. 1999) ).
Other investigations of human spermatogenesis con-
fi rmed the principle of helical patterns but not the
presence of a complete spermatogenic wave, i.e., the
complete succession of all stages (Johnson et al. 1996).
At the most, 2–4 consecutive stages could be found on
serial sections. Since the topographical distribution of
the stages could be reproduced by assigning random
numbers, the arrangement of the human spermato-
genic stages along the seminiferous tubule might be at
random. On the other hand, germ cell transplantation
revealed that one spermatogenic stage represents a
single clone of germ cells (Nagano et al. 2001).
Therefore, variation in clonal size could lead to the
appearance of several stages per cross-section (Wistuba
et al. 2003 for further discussion) and that species dif-
ferences with regard to number of spermatogenic
stages are related – at least in part – to clonal size. A
comparative and quantitative analysis of the incidence
of tubules with one or more spermatogenic stages in
17 primate species yielded that in men, great apes and
New World monkeys, multi-stage tubules are more
common, whereas in prosimians and Old World mon-
keys single-stage tubules predominate (Luetjens et al.
2005; Fig. 2.6).
Human germ cell production results in compara-
tively low sperm numbers per Sertoli cell (see Sect.
2.1.2.2). When expressed in millions of sperm per g/
testis in 24 h, the rat has values of 10–24, non-human
primates values of 4–5 and men values of 3–7. Since
earlier work also suggested that about 50% of germ
cells are lost during the meiotic divisions human sper-
matogenesis has been considered ineffi cient. However,
studies using contemporary stereological approaches
failed to detect meiotic germ cell losses in primates
including men (Zhengwei et al. 1997, 1998a, b). More
stages / tubule
multi-stages [%]
0
1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4
20
40
60
80
100 Prosimian monkeys
New World monkeys
Old World monkeys
Great apes / Humans
a1 Otolemur spec.
a2 Microcebus murinus
b1 Cebus apella
b2 Saimiri scireus
b3 Saguinus oedipus
b4 Saguinus fuscicollis
b5 Callithrix jacchus
c1 Papio hamadryas
c2 Mandrillus sphinx
c3 Macaca nigra
c4 Macaca fascicularis
c5 Macaca tibetana
c6 Cercopithecus aethiop
s
d1 Pan paniscus
d2 Pan troglodytes
d3 Homo sapiens
d4 Pongo pygmaeus
a1
a2
c4
c5
c6
c1
c2
b1
b3 b2
d4
b4 b5
d3 d2
d1
c3
Fig. 2.6 Frequency (%) of seminiferous tubules containing
more than one spermatogenic stage versus the number of stages
per tubular cross-section across the primate order. a1–a2: pros-
imians, b1–b5: New World monkeys, c1–c6: Old World mon-
keys, d1, d2 and d4: great apes, d3: men. Note the clustering of
multi-stage distribution and the increased number of stages in
New World monkeys, great apes and humans. The incidence of
multi-stage versus single-stage tubules was not related to germ
cell production (Luetjens et al. 2005; Fig. 2.7)

2 Physiology of Testicular Function 21
recent work using fl ow cytometric quantitation of tes-
ticlar cell numbers including meiotic cells and sperma-
tids showed that human germ cell yields are comparable
to other primates (Luetjens et al. 2005; Fig. 2.7). Hence
it is obvious that human spermatogenesis is more effi -
cient than assumed earlier. Differences in germ cell
number per cell or tissue unit are rather related to the
number of spermatogonial divisions (Ehmcke et al.
2005) with men considered to have only a single gen-
eration of B-type spermatogonia (Fig. 2.4), whereas
macaques can have four such generations.
Primate sperm production is controlled through the
number of spermatogonia entering meiosis.
2.1.2.5 Apoptosis and Spermatogenesis
Programmed cell death (apoptosis) comprises a coor-
dinated sequence of signalling cascades leading to
cell suicide. Unlike necrosis, this form of cell death
occurs under physiological conditions (spontaneous
apoptosis) but can also be induced by exposure to
toxicants, disturbances of the endocrine milieu, etc. In
the human testis spermatogonia, spermatocytes and
spermatids undergoing apoptosis have been detected
and ethnic differences in the incidence of testicular
apoptosis have been suggested. Blockade of apoptotic
events in the mouse model leads to accumulation of
spermatogonia and infertility underpinning the need
for germ cell death via apoptosis as a physiological
process. Endocrine imbalance or heat treatment
induced testicular apoptosis via the intrinsic and
extrinsic pathways in non-human primates (Jia et al.
2007). In a non-human primate model, differentiation
of Ap spermatogonia into B-type spermatogonia was
found to be gonadotropin-dependent (Marshall et al.
2005). Recent human data suggest that gonadotropins
may act as cell survival factors for spermatogonia
rather than as stimulators of cell proliferation
(Ruwanpura et al. 2008).
2.2 Hormonal Control of Testicular
Function
The endocrine regulation of testicular function, i.e.,
the production of sperm and of androgens is well
investigated. Understanding the hormonal interactions
has important clinical consequences, presented in the
following paragraphs. Figure 2.8 offers an overview of
the systems involved, of the endocrine factors and of
their physiological effects.
2.2.1 Functional Organization
of the Hypothalamo-Pituitary
System
The gonadotropins luteinizing hormone (LH) and
follicle-stimulating hormone (FSH) are produced and
secreted by the gonadotropic cells of the anterior pitu-
itary. Their designation is derived from the function
Callithrix jacchus
Macaca fascicularis
Papio hamadryas
Homo sapiens
0.8
0.6
0.4
0.2
0.8
0.6
0.4
0.2
Efficiency index
Meiosis index
Fig. 2.7 Spermatogenic effi ciency index (mean ± SEM) and
meiosis indices for New World monkeys (Callithrix jacchus =
marmoset, n = 4), Old World monkeys (Macaca fascicularis =
cynomolgus monkey, n = 5; Papio hamadryas = Hamadryas
baboon, n = 6) and man (Homo sapiens, n = 9) based upon fl ow
cytometric analyses of testicular tissue. The effi ciency index is
defi ned as the number of elongated cells divided by total cell
number. The meiosis index is defi ned as the number of haploid
cells divided by total cell number. Note that effi ciency and meio-
sis indices are comparable between human and other primates
(Based on Wistuba et al. 2003 and Luetjens et al. 2005)

22 G. F. Weinbauer et al.
exerted in females. In males, they control steroidogenesis
and gametogenesis in the testis. Pituitary gonadotropes
are the central structure controlling gonadal function
and in turn, are regulated by the hypothalamic gonado-
tropin-releasing hormone (GnRH). Since GnRH secre-
tion is pulsatile, gonadotropin release also occurs in
discrete peaks, more evident in the case of LH, due to
its shorter half-life in circulation compared to FSH. In
turn, GnRH secretion depends on the activation of the
GPR54 receptor, located on the surface of the GnRH
neurons and stimulated by the peptide kisspeptin.The
pituitary function is also under the control of gonadal
steroids and peptides that infl uence its activity both
directly and through the hypothalamus (Fig. 2.8). Due
to their very strict anatomical and functional connec-
tion, hypothalamus and pituitary gland have to be con-
sidered as an unique functional unit.
The hypothalamus is the rostral extension of the
brain stem reticular formation. It contains, among oth-
ers, the cellular bodies of neurons that project their
axon terminals toward the median eminence (ME), a
specialized region of the fl oor of the third ventricle
from which the pituitary stalk originates. The hypo-
thalamus is classically subdivided into three longitu-
dinal zones, periventricular, medial and lateral, the
latter functioning as the connecting area between lim-
bic and brain stem regions, whereas the former two
contain most of the nuclei controlling neuroendocrine
and visceral functions. The ME is the ventral bulge of
the hypothalamus and is the site where the axon termi-
nals of the neurosecretory neurons make contact with
the capillary plexus, giving rise to the hypophyseal
portal circulation. The nerve terminals form buttons on
the capillaries and release the neurohormones into the
portal blood by diffusion through the basal membrane.
The ME is outside the blood-brain barrier and thereby
freely accessible to the regulatory infl uences of hor-
mones and substances present in the systemic circula-
tion and mediating the release of neurohormones in
portal blood. The blood supply of the ME is provided
by the superior hypophyseal arteries. The long portal
hypophyseal vessels originate from the confl uence of
capillary loops which supply the anterior pituitary
gland with the highest blood fl ow of any organ in the
body. In humans, the perikarya of neurons stained pos-
itive for GnRH are especially found in the ventral part
of the mediobasal hypothalamus, between the third
ventricle and the ME, scattered throughout the periven-
tricular infundibular region.
The pituitary gland lies in the sella turcica, beneath
hypothalamus and optic chiasm, covered with the
sellar diaphragm. Thus, pituitary tumors can result in
visual impairment by exerting pressure on the optical
nerves. Gonadotropic cells are localized in the adeno-
hypophysis, the most ventral part of the gland, of
ectodermic origin from Rathke’s pouch. The adeno-
hypophysis consists of the anterior lobe (or pars dista-
lis, the anatomically and functionally most important
part), the pars intermedia and pars tuberalis.
The pars distalis is of pivotal importance for pitu-
itary function. Gonadotropin-producing cells consti-
tute approximately 15% of the adenohypophyseal cell
population, are scattered in the posteromedial portion
of the pars distalis and are basophilic and PAS-positive.
Although the secretion of LH and FSH can be partially
dissociated under certain circumstances, the same cell
type is believed to secrete both gonadotropins. About
80% of the gonadotropic cells in men contain both LH
and FSH. The cells have a very well developed RER, a
large Golgi complex and are rich in secretory granules.
In normal men, the pituitary contains approximately
700 IU of LH and 200 IU of FSH. Following gonadec-
tomy or in primary hypogonadism the cells become
vacuolated and large (castration cells). Finally, pitu-
itary gonadotropes are often found in close connection
with prolactin cells, suggesting a paracrine interaction
between the two cell types.
2.2.2 The Kisspeptin-GPR54 System
GnRH secretion is under the control of the kisspeptin-
GPR54 system. Kisspeptin is the product of the KISS1
gene, located on chromosome 1q32.1. The name of the
KISS1 gene derives from the chocolate “kisses” of
Hershey, Pennsylvania, the city in which the gene was
identifi ed. KISS1 was originally described as a human
tumor suppressor gene for its ability to inhibit the
growth of melanoma and breast cancer metastasis.
Later on it was shown that kisspeptin (also known as
metastin) is the natural ligand of the orphan receptor
GPR54 and has an important role in initiating GnRH
secretion at puberty.
The product of the KISS1 gene is a 145-amino-acid
peptide, which is cleaved into a 54-amino-acid peptide
known as kisspeptin-54. Shorter peptides (kisspeptin
10, −13, and −14), sharing a common C-terminal,

2 Physiology of Testicular Function 23
Cortex
Inhibin
Follistatin
Activin
FSH
Pituitary
LH
Testosterone
Estradiol
Dihydro-
testosterone
Testo-
sterone
Testosterone
Testis
and
epididymis
Leydig cells
Hypothalamus
Spermatogonia
Target organs
Sertolicells
Primary
spermatocyte
Secondary
spermatocyte
Spermatide
Sperm
germ-
cell
of the
tubulus
semini-
ferus
GnRH
Fig. 2.8 Hormonal regulation of the testicular function and
effects of androgens. Key hormones are luteinizing hormone
(LH) and follicle-stimulating hormone (FSH), synthesized and
secreted under hypothalamic control of gonadotropin-releasing
hormone (GnRH). Leydig cells are located between the seminif-
erous tubules and synthesize and secrete testosterone under the
control of LH. Testosterone stimulates the maturation of germ
cells in seminiferous tubules. FSH acts directly on the seminifer-
ous tubules. In the germinal epithelium only Sertoli cells possess
receptors for testosterone and FSH. It is therefore believed that
the trophic effects of testosterone/FSH on gametogenesis are
mediated via somatic Sertoli cells. The testis and the hypo-
thalamo-pituitary system communicate through steroids and pro-
tein hormones. Testosterone inhibits the secretion of GnRH and
gonadotropins. Inhibin B and follistatin suppress selectively the
release of FSH from the pituitary gland, while activin stimulates
this process. Beside the effects on gametogenesis, testosterone
plays an important role in hair growth, bone metabolism, muscle
mass and distribution, secondary sexual characteristics and func-
tion of the male reproductive organs. (Nieschlag et al. 2008)

24 G. F. Weinbauer et al.
RF-amidated motif with kisspeptin-54, are probably
degradation products (Popa et al. 2008). Kisspeptin-
expressing neurons are located in the anteroventral
periventricular nucleus (AVPV), in the periventricular
nucleus, in the anterodorsal preoptic nucleus and in the
the arcuate nucleus (ARC). Outside the nervous sys-
tem, the KISS1 gene is expressed in placenta, testis,
pancreas, liver and intestine.
When injected icv or iv to rodents and primates,
kisspeptin stimulates LH secretion, an effect mediated
by the interaction with its receptor, GPR54, located on
the surface of the GnRH-secreting neurons. This
receptor was fi rst identifi ed as an orphan G protein
coupled receptor in the rat in 1999. The GPR54 gene
is located on chromosome 19p13.3. In 2003 it was dis-
covered that loss-of-function mutations of GPR54 in
the human cause failure to progress through puberty
and hypogonadotropic hypogonadism (De Roux et al.
2003; Seminara et al. 2003). Therefore, the kisspeptin-
GPR54 system is essential to initiate gonadotropin
secretion at puberty and to maintain normal androg-
enization in adulthood. In fact, kisspeptin neurons
located in ARC and AVPV send projections to the
medial preoptica area, a region rich in GnRH cell bod-
ies, providing the anatomical evidence of a direct rela-
tionship between kisspeptin fi bers and GnRH neurons
which, in turn, express GPR54. The indispensable role
of the kisppeptin-GPR54 system for gonadotropin
secretion is proven also by the hypogonadotropic
hypogonadal phenotype of mice bearing targeted null
mutations of the kiss1 or gpr54 gene (Seminara et al.
2003; d’Anglemont de Tassigny et al. 2007).
GPR54 signals through a Gq-type of G protein.
Experimentally, kisspeptin stimulates phosphati-
dylinositol (PI) turnover, calcium mobilization and
arachidonic acid release in GPR54 – expressing cells
and induces phosphorylation of mitogen-activated pro-
tein (MAP) kinases. Continuous infusion of kisspeptin
results in rapid increase in LH secretion after 2 h, fol-
lowed by a decrease to the basal levels by 12 h of infu-
sion due to desensitization of GPR54, since GnRH is
still able to elicit LH secretion under these conditions.
This suggests that endogenous, pulsatile kisspeptin
release is physiologically responsible for pulsatile
GnRH and LH secretion.
Kisspeptin is sensitive to steroid levels within the cir-
culation and is the mediator of the negative and positive
(in the female) feedback regulation of gonadotropin
secretion. In fact, although androgens, estrogens and
progesterone suppress gonadotropin secretion through
androgen receptor (AR)-, estrogen receptor (ER)α-, and
progesterone receptor (PR)-dependent mechanisms,
respectively, none of these sex steroids affect GnRH
secretion by direct action on GnRH neurons. On the
contrary, kisspeptin neurons in the ARC are direct tar-
gets of sex steroids in all species and should be viewed
as the site of the negative feedback control of GnRH
production. In addition, kisspeptin produced by the
AVPV, a sexually dimorphic nucleus rich in steroid-sen-
sitive neurons in the female, mediates the positive feed-
back effects of estrogen on GnRH secretion (Popa et al.
2008). Finally, kisspeptin neurones seem to be involved
in the regulation of the reproductive axis by metabolic
signals sensing the energy balance of the organism, e.g.,
leptin. Reproductive hormones are inhibited during star-
vation and kisspeptin mediates some of leptin’s effects
on reproduction. According to the current model, leptin
and perhaps other adiposity and satiety factors stimulate
KISS1 expression, which results in stimulation of GnRH
release. When levels of adiposity and satiety factors
decrease or when such factors are not detected, the
expression of KISS1 (and presumably its secretion)
decreases, thus reducing excitatory input to GnRH neu-
rons. Among the other metabolic factors infl uencing
gonadal function, the growth hormone secretagogue
ghrelin should be considered as a possible modulator of
kisspeptin neurones (Tena-Sempere 2008). The regula-
tion of GnRH and gonadotropin secretion by kisspeptin
is shown in Fig. 2.9.
2.2.3 GnRH
2.2.3.1 Structure of GnRH
Two forms of GnRH, termed GnRH-I (or GnRH) and
GnRH-II, encoded by separate genes have been identi-
fi ed. The two forms are structurally very similar but
show a signifi cantly different tissue distribution and
regulation of gene expression (Cheng and Leung
2005). GnRH-I, the peptide involved in gonadotropin
regulation, is a decapeptide produced in the GnRH
neurons of the hypothalamus. Unique among neurons
producing hypothalamic neurohormones, they origi-
nate from olfactory neurones and during embryonic
development migrate toward the basal forebrain along
branches of the terminal and vomeronasal nerves,

2 Physiology of Testicular Function 25
across the nasal septum. This event is regulated by a
number of factors that infl uence the migration of dif-
ferent portions of the GnRH neuronal population at
different steps along the route and the formation of the
olfactory bulb (Tobet and Schwarting 2006). The
importance of such factors is demonstrated by muta-
tions in the respective coding gene in patients with
Kallmann syndrome. In about 10% of patients with
Kallmann syndrome and anosmia due to a hypoplasia
of the bulbus olfactorius, mutations or deletions of the
KAL1 gene on the X chromosome were detected. This
gene was the fi rst implicated in Kallmann syndrome
and encodes for anosmin-1 which is produced in the
bulbus and in other tissues, and which is transiently
expressed as an extracellular matrix and basal mem-
brane protein during organogenesis and interacts with
heparan sulphate. Other genes implicated in GnRH
neuron migration and Kallmann syndrome are those
encoding the fi broblast growth factor receptor 1
(FGFR1) and its ligand fi broblast growth factor 8
(FGF8), as well as prokineticin 2 (PK2) and its recep-
tor (PKR2) (Falardeau et al. 2008; Kim et al. 2008).
In primates, the main locations of GnRH neurons are
the medio-basal hypothalamus and the arcuate nucleus,
but they are found also in the anterior hypothalamus,
preoptic area, septum and other parts of the forebrain.
GnRH neurons are synaptically connected with termi-
nals stained positive for pro-opiomelanocortin-related
peptides and enzymes involved in the metabolism of
catecholamines and g-aminobutyric acid (GABA).
Furthermore, GnRH-positive neurons of the nucleus
arcuatus are connected to neuropeptide Y (NPY)
neurons in the preoptical area and in the eminentia
mediana. All these substances are known to infl uence
GnRH secretion (Evans 1999; Li et al. 1999).
The gene encoding GnRH is localized at the chro-
mosomal site 8p21-p11.2. GnRH is produced by suc-
cessive cleavage stages from a longer precursor, called
preproGnRH, transported along the axons to the ME
and there released into portal blood. Phylogenetically
GnRH is a rather ancient hormone, highly conserved
among different species with 80% sequence identity
between mammals and fi sh. In the precursor with a
length of 92 amino acids, GnRH is preceded by a sig-
nal peptide consisting of 24 amino acids, and followed
by a stretch of 56 amino acids forming the GnRH-
associated peptide (GAP). PreproGnRH is processed
in the rough endoplasmic reticulum and in the Golgi
complex, the fi rst step being the removal of the signal
peptide and the cyclization of the aminoterminal Gln
residue to pyroGlu. At the junction between GnRH and
GAP a Gly-Lys-Arg sequence provides a processing
signal important for the cleavage of GAP and
C-terminal amidation of the last Pro residue. Mature
GnRH is therefore a single chain decapeptide cyclized
at the N-terminus and amidated at the C-terminus and
assumes a folded conformation as the result of a ß-II
type bend involving the central Tyr-Gly-Leu-Arg resi-
dues that brings the N- and C termini in close proxim-
ity (Millar et al. 2008).
GnRH has a very short half-life (<10 min) and is
mostly retained and degraded in the pituitary gland
immediately after secretion by several peptidase sys-
tems. Deciphering the GnRH sequence earned Andrew
GPR54
GnRH neuron
kisspeptin
arcuate nucleus
steroids?
leptin?
other signals?
GnRH
GnRH receptor
pituitary gonadotrope
portal pituitary
circulation
LH/FSH
AVPN
+
-
Fig. 2.9 Current model of
regulation of gonadotropin
secretion. Kisspeptin produced
by the arcuate nucleus and in
the anteroventral periventricu-
lar nucleus (AVPV) stimulates
GnRH release by acting on the
G protein–coupled receptor
GPR54. GnRH release results
in increased gonadotropin
secretion from the pituitary
gland, which stimulates LH
and FSH production.
Peripheral sex steroids (e.g.,
androgens, estrogens and
progesterone) as well as
metabolic signals (e.g., leptin)
regulate Kiss1 expression and
signaling to GnRH neurons

26 G. F. Weinbauer et al.
Schally the 1977 Nobel prize, and enabled the devel-
opment of analogs with agonistic or antagonistic prop-
erties. As the generation of synthetic analogs of GnRH
has shown, the amino acids in position 6–10 are impor-
tant for high affi nity binding of the neuropeptide,
whereas positions 1–3 are critical for biological activ-
ity and positions 5–6 and 9–10 are involved in enzy-
matic degradation. The discovery of the amino acid
sequence of GnRH permitted the design of GnRH ana-
logs exerting agonistic or antagonistic action relative
to the endogenous GnRH.
2.2.3.2 Secretion of GnRH
GnRH is released into the portal blood in discrete
pulses. Although this event cannot be directly demon-
strated in vivo, all the experimental data accumulated
up to now demonstrate that each LH peak is induced
by a GnRH pulse. It is the frequency of GnRH pulses
and the amplitude of its secretory episodes that deter-
mine the type of LH and FSH secretion from the pitu-
itary gland (Fig. 2.10). GnRH is the sole releasing
factor for both gonadotropins, but modulating its fre-
quency results in preferential release of LH or FSH
(Hayes and Crowley 1998).
The pulsatile nature of GnRH secretion is partly an
intrinsic characteristic of the GnRH neurons, since
isolated immortalized GnRH neurons have a sponta-
neous pulsatile secretory activity in vitro. However, in
vivo GnRH is under the control of the kisspeptin/
GPR54 system, which mediates the effects of the
peripheral steroids on GnRH secretion and is involved
in the control of GnRH pulsatility. The pulse genera-
tor is under the continuous tonic inhibition of periph-
eral steroids and, e.g., gonadectomy results in an
immediate increase of frequency and amplitude of
gonadotropin secretion. Thus, in the absence of ste-
roids the pulse generator becomes free-running (Lopez
et al. 1998).
In man, the major hormone controlling GnRH
secretion is testosterone, which inhibits gonadotropin
secretion via negative feedback both at the hypotha-
lamic and pituitary level (Fig. 2.8). Testosterone can
act as such or after metabolism to DHT or estradiol.
The effects of testosterone and its metabolites appear
to vary depending on the experimental model but, in
general, we can assume that both T and DHT act
mainly at the hypothalamic level by decreasing the fre-
quency of GnRH pulsatility, whereas estrogens depress
gonadotropin secretion by reducing the amplitude of
LH and FSH peaks at the pituitary level (Hayes and
Crowley 1998). Progesterone inhibits gonadotropin
release at least in part via arcuate nucleus dopaminer-
gic and NPY neurons (Dufourny et al. 2005). The neg-
ative feedback action of androgens and progestins is
most important for the development of a male fertility
control regimen (Chap. 29).
Fig. 2.10 Importance of the
pulsatile pattern of GnRH
secretion for gonadotropin
secretion and testicular
function. Unphysiologically
high GnRH pulse frequencies
or continuous administration
of GnRH inhibit gonadotro-
pin secretion and testicular
function (red). Similarly,
blockade of GnRH receptors
by means of GnRH analogs
results in suppression of
testicular function
Nonphysiological patternPhysiological pattern
GnRH
Pulsatile
or
GnRH
continuous
GnRH
GnRH analogues
(receptor blockade)
LH
Pulsatile
GnRH
LH
Gonadotropic
cells
Gonadotropin
secretion
TESTIS

2 Physiology of Testicular Function 27
Among the neurotransmitters and neuromodulators
that might infl uence GnRH secretion, the noradrener-
gic system and NPY show stimulatory activity, whereas
interleukin-1, dopaminergic, serotoninergic and
GABAergic systems are inhibitory. Opioid peptides
seem to modulate the negative feedback of gonadal
steroids. Finally, leptin has been shown to stimulate
gonadotropin secretion. Leptin is produced by the fat
cells of the body and infl uences the interactions con-
cerning the control of body mass and gonadotropin
secretion. This effect is probably mediated by the
hypothalamus via NPY-, POMC- and especially kiss-
peptin-containing neurons with numerous receptors
for leptin (Popa et al. 2008).
2.2.3.3 Mechanism of GnRH Action
GnRH acts through interaction with a specifi c recep-
tor. The GnRH receptor belongs to a family of
G-coupled receptors linked by the typical seven-mem-
brane domain structure. This group also encompasses
the receptors for LH, FSH and TSH. With 328 amino
acids, it is the smallest G protein-coupled receptor
known up to now and possesses a rather short extra-
cellular domain (Fig. 2.11). The intracellular
C-terminus is practically absent and the signal trans-
duction is probably carried out by the intracytoplas-
mic loops connecting the seven membrane spanning
segments, especially the third one which is unusually
long. The receptor contains two glycosylation sites,
projecting into the extracellular space, of uncertain
function.
The conformation of the binding site is unknown.
The gene encoding the human GnRH receptor includes
three exons separated by two introns. The 5′ fl anking
region contains multiple TATA transcription initiation
sites and several cis-acting regulatory sequences
which confer responsiveness to cAMP, glucocorti-
coids, progesterone, thyroxin, PEA-3, AP-1, AP-2 and
Pit-1-sensitive sequences. The GnRH receptor is spe-
cifi cally expressed in the gonadotropic cells within the
pituitary. An orphan receptor, steroidogenic factor
1(SF-1) is involved in the expression of human GnRH
(Ngan et al. 1999). Transcription factors such as SF-1,
Pit-1 and Pro-Pit-1 are generally needed for the devel-
opment and maturation of the hypothalamo-hypophy-
seal-gonadal axis. SF-1-defi cient mice and patients
bearing a mutation in the Pro-Pit-1 gene exhibit
pronounced alterations of gonadotropin secretion.
Recently, a second GnRH receptor gene was identifi ed
in non-human primates (type II GnRH receptor) which
is structurally and functionally distinct from the clas-
sical, type I receptor. The GnRH type II receptor,
however, is not functional in the human and many
other species and its role is yet unknown (Millar 2005).
Both GnRH-I and GnRH-II were shown to signal
through the GnRH type I receptor (ligand-selective-
signaling). While GnRH-I regulates gonadotropins,
GnRH-II appears to be a neuromodulator and stimu-
lates sexual behavior.
Fig. 2.11 Schematic representation of the receptors for GnRH
(upper panel), LH (middle panel) and FSH (bottom panel).
These receptors belong to the family of G protein-coupled recep-
tors. They possess an extracellular domain, a transmembrane
domain with seven membrane spanning segments and an intrac-
ellular domain. The GnRH receptor is characterized by a very
short extracellular domain and practically no intracellular
domain. On the contrary, the extracellular domain is very large
in the gonadotropin receptors, where it plays a crucial role in
hormone binding. The intracellular domain is important for sig-
nal transduction
-NH2
III IVIII VIV VII
GnRH receptor
cell
membrane
-COOH
–NH2
III IVIII VIV VII
LH receptor
cell
membrane
– COOH
–N H2
III IVIII VIV VII
FSH receptor
cell
membrane
-COOH

28 G. F. Weinbauer et al.
Following GnRH-receptor interaction, a hormone
receptor complex is formed. This results in the inter-
action with Gq protein, hydrolysis of phosphoinositide
and production of diacylglycerol and inositol trisphos-
phate, which leads to calcium mobilization from the
intracellular stores and infl ux of extracellular calcium
into the cell. Diacylglycerol and calcium then activate
protein kinase C (PKC), inducing protein phosphory-
lation and further activation of calcium channels. The
increase in intracellular calcium results in prompt
gonadotropin release by exocytosis and, with time,
more sustained gonadotropin synthesis and secretion.
Thereafter, the hormone-receptor complex is internal-
ized by endocytosis and undergoes degradation in
lysosomes.
GnRH is capable of modulating number and
activity of its own receptors and the effects depend
on the secretory pattern and dose of neurohormone.
The receptor expression is higher when GnRH is
given in a pulsatile manner and the withdrawal of
GnRH during the interpulse intervals leads to increase
of GnRH binding sites just before the next pulse
occurs (self priming). Conversely, continuous expo-
sure to GnRH results in an initial rise in response fol-
lowed by desensitization. This property of the GnRH
receptor is exploited in therapy with GnRH agonists
that, owing to their prolonged and sustained stimula-
tory activity, cause slow receptor desensitization and
decrease of gonadotropin secretion. The molecular
mechanism of receptor desensitization is not com-
pletely understood. Owing to the lack of an intracel-
lular domain, GnRH agonists cannot induce
phosphorylation and rapid desensitization of the
receptor.
2.2.4 Gonadotropins
2.2.4.1 Structure of Gonadotropins
LH and FSH are glycoprotein hormones secreted by
the pituitary gland that control development, matura-
tion and function of the gonad. Like the related thyroid
stimulating hormone (TSH) and human chorionic
gonadotropin (hCG), they consist of two polypeptide
chains, α and β, bearing carbohydrate moieties
N-linked to asparagine (Asn) residues. The α subunit
is common to all members of the glycoprotein
hormone family, whereas the β subunit, although struc-
turally very similar, differs in each hormone and con-
fers specifi city of action.
The subunits are encoded by separate genes local-
ized on different chromosomes but structurally related.
The gene encoding for the α subunit is composed of
four exons and three introns, whereas the β genes con-
sist of three exons separated by two introns. The FSH
β gene is located on chromosome 11 and differs from
the other glycoprotein hormone β subunit genes in
possessing a rather long 3’ untranslated region proba-
bly involved in RNA stability. The LH β gene belongs
to an extraordinary complex cluster of genes also
including at least seven nonallelic hCGβ-like genes
arranged in tandem on chromosome 19. The regulation
of gene expression of LH and FSH has been exten-
sively studied in experimental animals, expecially
rodents, and involves a complex interplay between
hypothalamic GnRH and gonadal steroids and pep-
tides acting at the hypothalamic and pituitary level
(Burger et al. 2004).
The common α subunit contains two glycosylation
sites, at position 52 and 78. The glycosylation sites of
FSH ß are 7 and 24, whereas LH is glycosylated only
at position 30. In mammals the α subunit is also pro-
duced by the placenta and, conversely, the pituitary
gland has been shown to contain and secrete trace
amounts of hCG. α and β subunits are non-covalently
linked and the probable tertiary structure of pituitary
gonadotropins can be approximated and deduced by
analogy with its cognate hCG, whose crystal structure
has been resolved.
LH and hCGβ subunits are structurally very similar
and, in fact, LH and hCG act on the same receptor. A
peculiar feature of hCGβ is a carboxyl-terminal exten-
sion containing four O-linked sugar residues that
remarkably reduces the rate of metabolism and
increases the half-life of the hormone. This peculiarity
of hCGß has been recently exploited for the produc-
tion of a synthetic gonadotropin hybrid containing a
similar C-terminal extension in the ß subunit which
resulted in a conspicuous increase of the gonadotropin
half-life. A prolongation of the half-life of hCG was
achieved by producing a chimera containing fusioned
α and β chains (Boime and Ben-Menahem 1999),
while a synthetic chimeric gonadotropin containing a
β subunit derived from both hCGβ and FSHβ displays
the biological properties of both gonadotropins (Garone
et al. 2006).

2 Physiology of Testicular Function 29
The oligosaccharide structure consists of a central
mannose core, bound to an Asn residue through two
residues of N-acetyl-glucosamine, and terminal exten-
sions of tetrasaccharide branches, bi- or triantennary,
terminating with sialic acid (FSH) or sulfate (LH)
residues. These carbohydrate structures can be more
or less extended in length and are rich in sugar termi-
nals, constituting the molecular basis of the gonado-
tropin heterogeneity evident after chromatographic
separation.
Having a different terminal glycosylation, LH and
FSH also have a different half-life. LH is rich in
N-acetyl-glucosamine sulfate and is quickly removed
from the circulation after interaction with specifi c
liver receptors that recognize sulfate terminals. This
rapid removal of sulfate LH from the blood results in
rapid clearance of a relevant amount of the LH dis-
charged in each secretory episode and “amplifi es” the
pulsatile features of LH in circulation. Conversely,
FSH is predominantly sialylated and thereby protected
from immediate capture and metabolism in the liver.
As a result, LH and FSH half-lives are about 20 min
and 2 h, respectively. Therefore, although both gonad-
otropins are secreted simultaneously from the pitu-
itary gland following a GnRH pulse, LH appears to be
highly pulsatile and FSH much less so (Moyle and
Campbell 1995).
The importance of sugar residues on gonadotropin
activity has been investigated in vitro using glyco-
sylation-defi cient hormones. It was found that gly-
cosylation is not critical for receptor binding but is
important for receptor activation
The use of recombinant variants of hCG and FSH
defective in sialic acid or truncated at the mannose
ramifi cation revealed that glycosylation at position α
52 is necessary for the steroidogenic and cAMP
response. Isoforms completely devoid of carbohy-
drates cannot be secreted by the producing cells and
behave as competitive antagonists of the wild type.
Overall, the current view is that glycosylation is funda-
mental for gonadotropin secretion and bioactivity, and
strongly infl uences the half-life in circulation and in
vivo biopotency.
It was found recently that two polymorphic variants
of LH are present in the normal population. One of
them has two amino acid exchanges in positions 8 and
15 of the β chain leading to a second glycosylation site
in position 13. Approximately 12% of Europeans pro-
duce this allelic variant. Under in vitro conditions, this
variant displays increased bioactivity and shortened
half-life. A difference in immunoactivity can be
detected when certain monoclonal antibodies are used
(Huhtaniemi et al. 1999). Polymorphic variants of the
FSHβ gene exist as well and are associated with serum
FSH levels in men (Grigorova et al. 2008). Such a
variant was found to be associated with a signifi cant
reduction in free testosterone and testes volume, but
on the contrary, in an increase of semen volume, sex
hormone-binding globulin, serum testosterone and
estradiol.
2.2.4.2 Secretion of Gonadotropins
After the synthetic process is completed, LH and FSH
are stored in different secretion granules, ready to be
released upon stimulation with GnRH. A portion of
molecules, however, is not stored in secretory gran-
ules, i.e., does not enter the regulated pathway of secre-
tion and is, instead, constitutively secreted. FSH
especially follows the latter route. Storage in separate
granules and the natural propensity to follow one of
the two secretory pathways are the main reasons why
the same GnRH stimulus can, under certain conditions,
preferentially release one of the two gonadotropins.
Low GnRH pulse frequency causes preferential release
of FSH probably due to differential expression of the
FSH receptor (Ferris and Shupnik 2006).
LH and FSH are measurable in the pituitary gland
as early as the 10th week of gestation and during the
12th week in peripheral blood. In fetal life and in
infancy FSH is predominant over LH and the FSH/LH
ratio is higher in females than in males. The relative
abundance of the two gonadotropins changes during
development.
It is noteworthy that before birth both male and
female fetuses grow in an environment extraordi-
narily rich in potent, maternal estrogens.
It is testosterone that determines the initial phase of
testicular migration and the development of male
external genitalia. Testosterone is already produced by
the fetal testicle during the 10th week of gestation,
under the stimulation of fetal LH and maternal hCG.

30 G. F. Weinbauer et al.
The role of maternal hCG in this crucial phase of
gonadal development is suggested by the fact that a
mutation of the LH ß chain leading to a biologically
inactive gonadotropin is associated with normal sexual
differentiation (Huhtaniemi et al. 1999). Conversely,
inactivating mutations of the LH receptor produce a
clinical syndrome resembling complete androgen
insensitivity, with a phenotype of female external geni-
talia (Themmen et al. 1998).
During infancy gonadotropins in serum are very
low. The pulsatile secretion of gonadotropins becomes
evident at the time of puberty, when LH and FSH pulses
in serum are detected fi rst during sleep, at night, and
then progressively also during the day. Before puberty,
gonadotropin levels are very low and GnRH secretion
appears to be extremely limited, even in the presence of
negligible steroid production by the gonads. High sen-
sitivity of the hypothalamus to negative steroid feed-
back is believed to suppress GnRH production before
puberty, but certainly other factors such as body mass,
leptin and signals from the central nervous system are
important to maintain the hypothalamo-pituitary-
gonadal axis silent before the programmed time.
The steroid regulation of gonadotropin gene expres-
sion, synthesis and secretion is rather complex and
shows many facets depending on the experimental
model. In general, however, it is currently accepted that
gonadal steroids exert their negative control on gonado-
tropins mainly at the hypothalamic level, depressing the
release of GnRH most probably via the kisspeptin/
GPR54 system. The steroid effect at the pituitary level
is more complex, but there is considerable evidence that
estrogens inhibit GnRH-stimulated gonadotropin syn-
thesis and secretion at this level. In rodents, testosterone
has a specifi c stimulatory effect on FSH gene expres-
sion, synthesis and secretion directly at the pituitary
level. In primates, however, the effects of testosterone
are always inhibitory. Testosterone is the main testicular
product suppressing FSH and LH secretion in men.
FSH secretion is, obviously, also under the control of
some other factor(s) related to the effi ciency of spermato-
genesis, since oligoazoospermia is often accompanied by
selective increase of serum FSH in the presence of nor-
mal testosterone levels. New assays for inhibin B per-
mitted analysis of the relationship between FSH and
inhibin secretion in man. A pronounced inverse correla-
tion was established between serum concentrations of
inhibin B and serum levels of FSH, testis size and sperm
numbers. Clearly, inhibin B is the physiologically
relevant form of inhibin in men (von Eckardstein et al.
1999). It appears at present that the serum levels of
inhibin B directly refl ect the integrity of the germinal epi-
thelium and of the Sertoli cells (Boepple et al. 2008).
2.2.4.3 Mechanism of Action
of Gonadotropins
LH and FSH exert their function via specifi c recep-
tors (Simoni et al. 1997). The gonadotropin receptors
also belong to the family of the G protein-coupled recep-
tors and are characterized by a very large extracellular
domain to which the hormone binds specifi cally, the
usual membrane-spanning domain including 7 hydro-
phobic segments connected to each other through three
extracellular and three intracellular tracts, and an intrac-
ellular carboxyterminal domain (Fig. 2.11).
The genes for LH and FSH receptors are localized on
chromosome 2 and consist of 11 and 10 exons, respec-
tively. The last exon encodes a small portion of the
extracellular domain, the entire transmembrane domain
and the intracellular C-terminus. The extracellular
domain contains the high-affi nity hormone binding site
and is rich in leucine repeats. The 5′-fl anking region of
the two genes contains no conventional promoter and
has multiple transcription start sites. In addition, the
human LH receptor was recently shown to contain a
cryptic exon (denominated exon 6A) which plays a very
important role in the intracellular processing of the
mature receptor protein and can be mutated in rare cases
of LH resistance (Kossack et al. 2008). Several alterna-
tively spliced transcripts of LH and FSH receptors have
been described, lacking one or more exons, but presently
it is not known whether these RNA isoforms are trans-
lated into proteins of any physiological function. Single
nucleotide polymorphisms give rise to allelic variants
with different biological activity in vitro and/or in vivo
(Gromoll and Simoni 2005; Piersma et al. 2007).
The mature receptor proteins are glycosylated at
several points, a process that does not seem to be
involved in receptor activation and signal transduction
but probably necessary for receptor folding and trans-
port to the cell membrane. Recently, the partially
deglycosylated complex of human FSH bound to the
extracellular hormone-binding domain of its receptor
was crystallized, showing that binding specifi city is
mediated by key interaction sites involving both the
common alpha- and hormone-specifi c beta-subunits.

2 Physiology of Testicular Function 31
On binding, FSH undergoes a concerted conforma-
tional change that affects protruding loops implicated
in receptor activation. The FSH-FSHR complexes
form dimers in the crystal important for transmem-
brane signal transduction (Fan and Hendrikson 2005)
resulting in activation of the G protein, cAMP produc-
tion and activation of protein kinase A. Gonadotropins
act mainly through stimulation of intracellular cAMP.
More recently it has been shown that LH and FSH can
also induce an increase of Ca++ infl ux in target cells,
but the physiological importance of this mechanism is
still unknown. cAMP remains, therefore, the main sig-
nal transducer and calcium could possibly act as a sig-
nal amplifi cation or modulating mechanism. Following
the hormone-receptor interaction there is an increase
in cAMP concentrations and subsequent activation of
protein kinaseses which, in turn, phosphorylate exist-
ing proteins such as enzymes, structural and transport
proteins and transcriptional activators. Activating and
inactivating mutations of the gonadotropin receptors
have been identifi ed. The biological consequences of
these mutations are described below and in Chap. 13.
2.2.5 Endocrine Regulation and Relative
Importance of LH and FSH
for Spermatogenesis
The primary functions of the testis, androgen produc-
tion and gamete development, are regulated by the
brain, e.g., hypothalamus and hypophysis via GnRH
and gonadotropins. Importantly, the hypothalamo-
hypophyseal circuit is subject to negative feedback
regulation mediated by testicular factors (Fig. 2.8).
Testosterone inhibits the secretion of LH and FSH. For
FSH, the protein hormone inhibin B plays an impor-
tant role (Boepple et al. 2008).
For the interpretation of hormonal regulation and
hormonal effects on spermatogenesis, the following
terminology should be remembered:
1. Initiation: First complete cycle of spermatogenesis
during puberty
2. Maintenance: Hormonal requirements of intact
spermatogenesis in the adult
3. Reinitiation: Hormonal requirements for the restim-
ulation of gametogenesis after transitory interruption
4. Qualitatively normal spermatogenesis: All germ
cells are present although in subnormal numbers
5. Quantitatively normal spermatogenesis: All germ
cells are present in normal numbers
Considerable efforts were undertaken to unravel the
relative importance of LH/testosterone and FSH for
qualitative and quantitative initiation, maintenance and
reinitiation of spermatogenesis (Fig. 2.12). It is gener-
ally assumed that either testosterone and FSH alone
are able to initiate, maintain and reinitiate spermato-
genesis but only to a qualitative extent (Weinbauer
et al. 2004). In order to achieve quantitative effects on
germ cell production and sperm numbers, at least
under physiological conditions, both LH and FSH
activities are needed.
These assertions are based upon controlled studies
in non-human primate models and volunteers, and on
studies in patient populations and case reports. The lat-
ter has provided interesting information but can also
confound interpretation of fi ndings owing to the vari-
able endocrine and medical history of the patients.
Complete spermatogenesis is seen in the vicinity of
testosterone-producing Leydig cell tumors and in
patients with activating mutations of the LH receptor,
suggesting that pharmacologically high local
Fig. 2.12 Testicular histology of a 5.2-year old boy with an acti-
vating mutation of the LH receptor. Note complete spermato-
genesis (arrows). (Courtesy of Prof. Dr. W. Rabl, Pediatric Clinic
of the Technical University Munich)

32 G. F. Weinbauer et al.
testosterone concentrations induce sperm formation
(Fig. 2.13). The aim of treatment is to obtain suffi -
ciently high intratesticular testosterone concentrations,
which are crucial. This is normally pursued clinically
by giving hCG, which contains high LH activity,
together with FSH. On the other hand, patients bearing
a defective FSH β subunit, presented with azoospermia
(Lindstedt et al. 1998; Phillip et al. 1998). One of these
patients was normally virilized, suggesting the need of
FSH for complete initiation of spermatogenesis in man.
Conversely, patients with Pasqualini syndrome, a dis-
order with selective LH defi ciency, can have complete
spermatogenesis, indicating the ability of FSH to initi-
ate the entire male germ cell development cascade.
Exogenous provision of supranormal doses of tes-
tosterone or of gestagenic compounds suppresses
gonadotropin secretion through the negative feedback
mechanism and leads to a drastic decrease of sperm
numbers in the ejaculate. In primates – unlike in
rodents – it is essential that complete suppression of
FSH secretion is achieved despite inhibition of LH
secretion (Narula et al. 2002; Weinbauer et al. 2001).
In the latter study, albeit LH bioactivity had been com-
pletely eliminated, a slight and transient rebound of
FSH secretion provoked an escape of spermatogenic
suppression. In gonadotropin-suppressed men, either
FSH or LH maintained spermatogenesis (Matthiesson
et al. 2006). The importance of FSH is also evident
Fig. 2.13 Sites of action of testosterone and FSH on the sper-
matogenic process in primates. (?) denotes unresolved ques-
tions. Ap spermatogonia enter the spermatogenic process (arrow
on the cell indicates direction of germ cell development). Ad
spermatogonia are believed to constitute the testicular stem
cells. The majority of available data indicate that testosterone
and FSH act on spermatogenesis via increasing numbers of type
A-pale spermatogonia followed by an increase of subsequent
germ cell populations. These endocrine factors might act via
stimulation of proliferation and/or prevention of cell death.
Meiotic transitions appear to be independent of testosterone/
FSH action. FSH has been reported to play a role in chromatin
condensation during spermiogenesis. Whether testosterone is
needed for spermiogenesis has not been studied yet. Spermiation
is affected by gonadotropin suffi ciency but it is currently unclear
whether this is related to diminished actions of testosterone or
FSH or both. Ad = A-dark spermatogonium (testicular stem
cells, divides rarely), Ap = A-pale spermatogonium (self-renew-
ing and progenitor cell for spermatogenesis), B = B spermatogo-
nium, Pl = preleptotene spermatocytes, L = leptotene
spermatocytes, EP = early pachytene spermatocytes, MP = mid
pachytene spermatocytes, LP = late pachytene spermatocytes, II
= 2nd meiotic division, RB = residual body, Sa1–Sd2 = develop-
mental stages of spermatid maturation
Spermatogoniogenesis
Spermiogenesis Spermatozoa Spermiation
Ad
Ap BPL LZ
EP
MP
LP
II
Sa1/2
Spermatids
Sb1 Sb2 Sc Sd1
Sd2
RB
Testosterone
FSH
Testosterone
FSH
Testosterone ? / FSH ?
Testosterone ? / FSH
Meiosis
?
FSH ?

2 Physiology of Testicular Function 33
from a hypophysectomized patient in whom an acti-
vating mutation of the FSH receptor coexisted with
normal spermatogenesis in the absence of LH (Gromoll
et al. 1996). Conversely, inactivating mutations of
FSH action do not necessarily lead to a complete block
of spermatogenesis (Huhtaniemi 1996). Although
either hormone on its own has the potential to elicit
the entire spermatogenic process, this is not always
the case in patients receiving androgen/hCG therapies.
In case of failure of hCG, however, the addition of
FSH has been shown to permit completion of sper-
matogenesis in hypogonadotropic men with azoo-
spermia (Bouloux et al. 2003).
In certain animal species, e.g., Djungarian ham-
sters, FSH is the only hormone responsible for sper-
matogenesis, while LH and testosterone stimulate the
development of androgen-dependent organs and sexual
behavior. Conversely, in primates, both gonadotropins
are necessary for spermatogenesis. The biological
meaning of this dual regulation system is not clear
yet (Weinbauer et al. 2004).
From a clinical viewpoint it is concluded that the
synergistic action of LH/testosterone and FSH is
necessary for the initiation, maintenance and also
for reinitiation of normal spermatogenesis.
2.2.6 Local Regulation of Testicular
Function
As described above, the regulation of testicular func-
tion is primarily controlled by central structures. The
complexity of the testicular cell types and architecture
also mandates a variety of local control and regulatory
mechanisms. The categories of local interactions and
communication can be classifi ed as paracrine, refer-
ring to factors acting – mainly by diffusion – between
neighboring cells; autocrine, referring to factors which
are released from the cell and work back on the same
cell and intracrine, referring to factors and substances
which never leave the cell and whose site of production
and action is the same cell. The term “paracrinology”
has inadvertently been used earlier to characterize all
types testicular cell interactions which seem better
described by “local interaction” (Weinbauer and
Wessels 1999). In addition, the interplay between the
different testicular compartments are also subsumed
under local interactions.
Many local factors have been identifi ed and for sev-
eral of them, gene-targetting in mice either confi rmed
or challenged their pivotal role for somatic and germ
cell development. This approach in mice has been
strengthened by the ability of conditional and cell-spe-
cifi c gene targetting. In contrast and for obvious exper-
imental and ethical reasons, the identifi cation of
essential local factors for human testicular function
has been limited.
It is evident that the endocrine mechanisms play
the central role in the regulation of testicular function
and factors produced locally are important for the
modulation of hormone activity and local factors
could thus be seen as mediators of hormone action
and intra-/intercellular communication. From this
point of view both gametogenesis and endocrine func-
tion of the testis are under local control. An example
for this might be the earlier report of stage-specifi c
expression of androgen receptor in the human testis
(Suarez-Quian et al. 1999). While Sertoli cells were
viewed as coordinators and regulators of germ cell
development and maturation for a long time, these
cells are now believed to be infl uenced by germ cell
products that can infl uence the secretory activity of
Sertoli cells. Hence, Sertoli cells are under the local
control of germ cells having varying requirements for
metabolic substances depending on the spermatogenic
cycle phase (Franca et al. 1998).
A plethora of factors with potentially local tes-
ticular activity has accumulated, e.g., growth fac-
tors, stem cell factors, immunological factors,
opioids, oxytocin and vasopressin, peritubular
cell modifying substance, renin and angiotensin,
GHRH, CRH, ACTH, GnRH, calmodulin, ceru-
loplasmin, transport proteins, glycoproteins,
plasminogen activator, metalloproteases, dynor-
phin, PACAP, etc. Moreover, it can be reasonably
assumed that other, still unidentifi ed protein factors
mediate the communication between interstitial and
tubular compartments, between Sertoli cells and
germ cells and between germ cells.
2.2.6.1 Steroid Hormones
Testosterone is the main secretory product of the testis,
along with 5α-dihydrotestosterone (DHT), androster-
one, androstenedione, 17-hydroxyprogesterone, pro-
gesterone and pregnenolone. The role of androsterone,
progesterone and 17-hydroxyprogesterone in the testis

34 G. F. Weinbauer et al.
is unknown but progesterone receptors have been
found in some peritubular cells and on spermatozoa
(Luetjens et al. 2006; Modi et al. 2007). Using a deriv-
ative of progesterone, norethisterone enanthate, no
direct effects on testicular/epididymidal function were
found (Junaidi et al. 2005).
For testosterone, a classic endocrine factor, compel-
ling evidence is available as a pivotal local regulator of
spermatogenesis. Rodent data demonstrated that selec-
tive elimination of Leydig cells, interruption of testicu-
lar testosterone transport and specifi c Sertoli cell
androgen receptor knockout models provoked profound
alterations of germ cell maturation (Takaimya et al.
1998). Selective peritubular cell androgen receptor
knockout mice exhibited specifi c Sertoli cell and peri-
tubular cell defects (Zhang et al. 2006). Spermatogenesis
was present in boys with testosterone-producing Leydig
cell tumors but only in seminiferous tubules adjacent to
the tumor and not in tumor-free areas. Similarly, acti-
vating mutations of the LH receptor prematurely
induced qualitatively normal spermatogenesis.
In fertile men testicular testosterone concentrations
exceed that of SHBG/ABP by about 200-fold (Jarow
et al. 2001), indicating a substantial surplus of testoster-
one in the testis. Relative to serum, testicular testoster-
one concentrations were >80-fold higher (Coviello et al.
2005). Testosterone is metabolized to DHT by testicular
5α-reductase activity and to estradiol by testicular aro-
matase activity. To what extent these metabolic activi-
ties are essential for spermatogenesis besides testosterone
itself is not entirely clear. Treatment of volunteers with
the 5α-reductase-inhibitor fi nasteride did not alter sper-
matogenesis (Kinniburgh et al. 2001; Overstreet et al.
1999), whereas more recently, a mild decrease in semen
parameters was reported for fi nasteride and dutasteride
(Amory et al. 2007b). Addition of dutasteride to a con-
traceptive steroid regimen (testosterone and levonorg-
estrel) did not augment suppression of spermatogenesis
(Matthiesson et al. 2005a). With regard to estradiol, aro-
matase activity and estrogen receptor-β are present in
human Sertoli cells and germ cells (Carreau et al. 2006
for review; Berensztein et al. 2006). Administration of
aromatase inhibitor in a non-human primate model with
seasonal reproduction (bonnet monkey) resulted in
impairment of spermiogenesis and altered sperm chro-
matin condensation (Shetty et al. 1997, 1998). The clini-
cal evidence for a causal role of estrogens for testicular
functions is ambiguous since isolated cases with estro-
gen receptor defi ciency or aromatase defi ciency did not
reveal a consistent picture, with some patients being ill
and having cryptorchidism (O’Donnell et al. 2001;
Maffei et al. 2004).
Although it is established beyond doubt that testos-
terone is an essential local regulator of spermatogene-
sis, it has been surprisingly diffi cult to demonstrate a
clear-cut relationship between testicular testosterone
concentrations and germ cell production. In non-
human primates, no correlation between testicular
androgen levels and germ cell production/spermato-
zoal number was observed (Narula et al. 2002;
Weinbauer et al. 2004 for review). Similarly, contra-
ceptive studies in volunteers failed to demonstrate a
correlation between intratesticular steroids and germ
cell numbers (Matthiesson et al. 2005b).
In the non-human primate testosterone induces the
formation of smooth muscle actin in the peritubular
cells during prepubertal testicular maturation (Schlatt
et al. 1993). Peritubular cells express the androgen
receptor. The testosterone effect is signifi cantly rein-
forced by FSH. Since FSH receptors are found only in
Sertoli cells it follows that FSH infl uences androgen
action indirectly through factors arising in the Sertoli
cell. This indicates that as an endocrine factor FSH can
also induce the formation of physiologically relevant,
locally acting factors in the primate testis. Interestingly,
recombinant FSH stimulates testosterone production
in men (Levalle et al. 1998) and in patients with selec-
tive FSH defi ciency (Lofrano-Porto et al. 2007), lend-
ing further support to the importance of local
interactions between Sertoli cells, Leydig cells and
peritubular cells in connection with the actions of
androgens and gonadotropins. In-vitro studies using
monkey Sertoli cells exposed to testosterone and FSH,
revealed that estradiol was produced in the presence of
testosterone but not of FSH (Devi et al. 2006). This is
surprising since based upon rodent data, Sertoli cell
aromatase activity is regulated by FSH.
2.2.6.2 Insulin-like Factor 3
Insulin-like factor 3 (INSL3) is a relaxin-like proteo-
hormone produced by Leydig cells (Foresta et al.
2004) and signals through a G-coupled receptor
(LGR8) that is expressed in Leydig cells and meiotic/
Testosterone acts both as an endocrine and local
(paracrine and autocrine) factor within the testis.

2 Physiology of Testicular Function 35
postmeiotic testicular human germ cells but not in
peritubular and Sertoli cells (Anand-Ivell et al. 2006).
Compelling evidence is available to indicate that
INSL3 is a marker of Leydig cell differentiation and of
entry into male puberty (Ferlin et al. 2006; Wikström
et al. 2006). Levels of INSL3 are infl uenced by hCG/
LH but this effect appears uncoupled from the ste-
roidogenic effects of LH on testosterone synthesis
(Bay et al. 2005, 2006). Since receptors for INSL3 are
present on advanced germ cells it is tempting to assume
a local role for INSL3 during spermatogenesis. A ret-
rospective analysis of several male contraceptive stud-
ies suggested that circulating INSL3 levels were lower
in azoospermic subjects compared to non-azoospermic
subjects and a weak correlation between INSL3 levels
and sperm number was reported (Amory et al. 2007a).
More data is needed to reach a fi rm conclusion on the
relevance of INSL3 as a local regulator of human
spermatogenesis.
2.2.6.3 Growth Factors
Growth factors bind to surface receptors and induce
cell-specifi c differentiation events via specifi c signal
transduction cascades. Among those factors participat-
ing in the local regulation of spermatogenesis are
transforming growth factor (TGF)-α and -β, inhibin
and activin, nerve growth factor (NGF), insulin-like
growth factor I (IGF-I), fi broblast growth factor
(FGF) and epidermal growth factor (EGF).
Inhibin and activin have been detected not only in
Sertoli cells, but also in Leydig cells of primates.
Inhibins and activins are structurally related proteins.
The heterodimer inhibin consists of an α subunit and a
βA or βB subunit whereas activins are homodimers
(βAβA or βBβB). Generally, activins are considered to
stimulate spermatogonial proliferation whereas inhib-
ins exert inhibitory actions. Of considerable clinical
interest are recent discoveries that serum concentra-
tions of inhibin are correlated with spermatogenic
activity, testis size and sperm production. This growth
factor can actually be used as an endocrine indicator of
local spermatogenic defects (Boepple et al. 2008;
Meachem et al. 2001).
In vitro studies have suggested that the local func-
tion of inhibin and activin could be a modulation of
steroidogenic activity in Leydig cells. Activins inhibit
or stimulate Leydig cell steroidogenesis in a species-
dependent manner. Generally, IGF-I and TGF-α exert
a stimulatory activity in the testis, while TGF-β acts as
an inhibitor. In the rat, the development of Leydig cells
is sustained by an interplay between TGF-α and TGF-β
and LH activity is modulated by IGF-I. In the human
Leydig cell the steroidogenic activity is also stimulated
by EGF. This growth factor directly infl uences sper-
matogenesis: IGF-I concentrations are positively
related to the number of pachytene spermatocytes. In
man, IGF-I shows the highest expression in these sper-
matocytes and stimulates DNA synthesis in mitotic
germ cells. Administration of IGF-I to patients
increased testis size but this was also associated with
increased gonadotropin secretion and testosterone
(Laron and Klinger 1998). An important role of NGF
for the structural organization of the human seminifer-
ous tubules is postulated, since the culture of seminif-
erous tubules can be successful only in the presence of
NGF. NGF has been localized in peritubular cells by
immunocytochemistry. In the rat, NGF is an important
regulator of meiotic division. Fibroblast growth factor
has been involved in mitosis and Sertoli-germ cell
interactions.
2.2.6.4 Immune System Factors
Cells of the immune system and the blood-testis-bar-
rier provide a special environment for the develop-
ment of otherwise antigenic germ cells (see Sect. 2.5).
It appears, however, that immune factors may also
play a more direct role during testicular steroidogen-
esis and gametogenesis and might have a relationship
to male infertility (Albrecht et al. 2005; Fijak and
Meinhardt 2006; Hedger 2002). These factors involve
leucocyte, macrophage and mast cell products. For
example, cytokines such as interferon, tumor necrosis
factor (TNF), interleukins, leukemia inhibing factor
(LIF), stem cell factor (SCF), macrophage migration
inhibiting factor (MIF) that bind to cell surface recep-
tors and provoke cell proliferation and differentiation.
TNF and LIF are suspected to play a role in Sertoli
cell-germ cell interactions and in the autocrine control
of Sertoli cell proliferation. MIF is produced specifi -
cally by Leydig cells and is found in Sertoli cells,
basal germ cells and peritubular cells following elimi-
nation of Leydig cells, indicating that compensatory
mechanisms maintain testicular MIF production.
Unlike interleukins, SCF and its receptor (c-kit) are

36 G. F. Weinbauer et al.
clearly essential local factors that govern germ cell
migration during ontogenesis and spermatogonial dif-
ferentiation in the adult testis. SCF is synthesized and
secreted by Sertoli cells whereas the receptor is
expressed on spermatogonial surfaces. The SCF/c-kit
system is important for spermatogonial differentiation
and development.
2.3 Testicular Descent
The incidence of positional anomalies of the testis is
over 3% and ranges among the most common congeni-
tal defects (reviewed in Virtanen et al. 2007a). These
defects are associated with spermatogenic disturbances
such as fewer spermatogonial stem cells at birth com-
pared with normal boys (Virtanen et al. 2007b) and
increased risk of testicular tumor development.
Testicular descent is multifactorial with two distinct
phases. First a descending phase from the lower kid-
ney pole to the pelvic cavity (transabdominal phase
of descent) controlled by the swelling of the guber-
naculum. The shortening of the gubernacular cord and
the outgrowth of the gubernacular bulb controlled by
the genitofemoral nerve is independent of androgens.
The gubernaculum deposits extracellular matrix, rich
in glycosaminoglycans and hyaluronic acid, and forms
a cone-like structure at the caudal end of the gonad,
anchoring the developing testis close to the inguinal
region during fetal growth. The second phase, the
descent into the scrotum (inguino-scrotal phase of
descent) is controlled by androgen action (Shono
2007). In the 26th gestational week the gubernaculum
begins to grow through the inguinal canal and reaches
the scrotum by gestational week 35, pulling the testis
in its path before the gubernaculum shrinks to a fi brous
remnant The intra-abdominal pressure and the shrink-
age of the gubernaculum may force the testis through
the inguinal canal. At birth, the testes reach at the
bottom of the scrotum and in 97% of boys testicu-
lar descent is completed within another 12 weeks
(Fig. 2.14).
The physiological and endocrine mechanisms that
govern testicular descent are not known in detail.
Several factors besides androgen such as calcitonin
gene-related peptide, epidermal growth factor (EGF)
and fi broblast growth factor (FGF) family are candi-
dates for inducing the testicular descent (Nightingale
et al. 2008). Hoxa-10 and insulin-like factor 3 (INSL3)
are potential regulators of these phases as suggested by
the fact that in gene knockout mice maldescended tes-
tes remain located in the abdominal cavity. INSL3 pro-
duced by the Leydig cells together with androgen
induces the gubernaculum growth and is therefore
needed in the early phase (Emmen et al. 2000). INSL3
knockout mice have their testes high in the abdominal
cavity. In Hoxa-10 knockout mice, testes remain close
to the scrotum after the initial descending phase.
Estrogens or environmental endocrine disruptors have
also been suspected to induce a down-regulated INSL3
expression and thus disturb testicular descent (Toppari
et al. 2006). Genetic analysis in men revealed several
functionally deleterious mutations in both INSL3 and
Kidney (K)
Testis (T)
Gubernaculum (G)
Bladder (B)
Wolffian duct (W)
K
W
B
G
W
T
K
B
G
T
Fig. 2.14 Testicular descent.
At an early stage the gonad
precursor organ is located
next to the kidney (K).
During the fi rst testicular
descendent phase between
week 8 (left image) and week
17 (right image), the testis is
anchored by the swollen
gubernaculum close to the
inguinal region. The bud-like
growth of the gubernaculum
is regulated by INSL-3 and
androgens

2 Physiology of Testicular Function 37
its receptor GREAT/LGR8 gene. Although some muta-
tions were found only in patients with maldescended
testes, the causative link between the presence of muta-
tions in INSL3 or GREAT/LGR8 and the undescended
testes remains to be demonstrated (Feng et al. 2006;
Yamazawa et al. 2007).
2.4 Vascularization, Temperature
Regulation and Spermatogenesis
Vascularization of the testis has two main roles: trans-
port and mobilization of endocrine factors and metab-
olites, as well as regulation of testicular temperature.
The arterial supply of the testicular parenchyma fol-
lows the lobular division of the seminiferous tubules.
Each lobule is supplied by one artery from which seg-
mental arteries originate at a distance of about 300 µm
from each other, supplying blood to the lateral regions
of the lobuli (Ergün 1994a, b). Segmental arteries and
capillaries become branched between the Leydig cells
and fi nally give rise to the venous system.
In men, testicular temperature is about 3–4°C below
core body temperature and about 1.5–2.5°C above the
temperature of scrotal skin. For the maintenance of a
physiologically lower temperature the testis relies on
two thermoregulatory systems. Heat can be transferred
to the external environment through the scrotal skin, as
the scrotal skin is very thin, possesses hardly any subcu-
taneous fat tissue and has a very large surface.
The second regulatory system is the pampiniform
plexus. In this system, the convoluted testicular artery
is surrounded by several veins coiling around the artery
several times. Arterial blood arriving at the testis is
thereby cooled down by the surrounding venous blood.
The usual explanation for the pampiniform plexus here
is to effi ciently maintain the optimal temperature
which is below body temperature. Recently, a new
theory has been put forward hypothesizing that the
process of spermatogenesis results in a large amount
of heat which has to be regulated. Testes are located in
the scrotum in order to maintain lower than body tem-
peratures (Skandhan and Rajahariprasad 2007). Some
mammals’ testes remain functional inside the body,
e.g., elephants, but these animals lack sweat glands
and are closely related to aquatic ancestors which have
to compensate for the chilling effects of the heat-con-
ducting environment, namely water. Human scrotal
skin is devoid of subcutaneous fat and the presence of
high sweat gland density enables heat transmission.
Upon exposure to cold temperatures, the scrotal sur-
face is minimized by contraction for preventing tem-
perature loss and cremaster muscles retract the testes
closer to the abdomen for temperature maintenance.
In case of varicoceles, defi ned as abnormally
dilated scrotal veins, scrotal temperature is increased
(Lerchl et al. 1993). Approximately 15% of normal
men have a varicocele, with 40% of them having fer-
tility problems. Venous refl ux and testicular tempera-
ture elevation appear to play an important role in
testicular dysfunction, although the exact pathophysi-
ologic mechanisms involved are not yet completely
understood. An increase of testicular temperature
results in damage of the spermatogenic function of the
testis (Jung and Schuppe 2007). If testicular tempera-
ture is increased in adults, reversible spermatogenic
damage can be induced. Most importantly, however,
substantial increase of temperature must be achieved.
Scrotal temperature elevations by 0.8–1°C over a
period of 52 weeks in healthy volunteers had no
adverse effect on the number and quality of spermato-
zoa (Wang et al. 1997).
Treatment with heat leads to sustained activation of
both mitogen-activated protein kinase MAPK13 and
MAPK14. Activation of MAPK13 and MAPK14 is
accompanied by an increase in B-cell leukemia/lym-
phoma 2 (BCL2) levels in both cytosolic and mito-
chondrial fractions of testicular lysates, leading to
apoptosis mediated by cytochrome c and DIABLO
release (Jia et al. 2007). Heat also results in inactiva-
tion of Bcl2, an anti-apoptotic protein, through phos-
phorylation at serine 70, thereby favoring the apoptotic
pathway. Both the protein kinases and the anti-
apoptotic factor Bcl2 help balance the death pathway
in male germ cells. However, although testicular tem-
perature is lower than body temperature in the majority
of mammals, this difference is not obligatory in every
species.
2.5 Immunology of the Testis
Gonocytes migrate to the testis even during prenatal
development, but spermatogonia begin to differentiate
into spermatozoa only at puberty after the immune
system has matured and a systemic self-tolerance is

38 G. F. Weinbauer et al.
developed. As the spermatogonia proliferate and dif-
ferentiate into spermatocytes, many new surface and
intracellular proteins are expressed which are unique
in the body, especially for the immune system. These
proteins are also novel antigens which have to be toler-
ated by the immune system. At the same time, adjacent
Sertoli cells form complex networks of specialized
tight junctions (zonula occludens) that cause isolation
of the tubular contents from the blood vascular com-
partment. This barrier contains several integral mem-
brane proteins, comprising various components such
as junctional adhesion molecules, claudin family pro-
teins, and occludin. The blood-testis-barrier provides
a separation between immune cells and haploid germ
cells with antigenic properties. This barrier was thought
to provide the basis of the systemic tolerance to newly
developing auto-antigens. Transplantation of sper-
matogonia into testis tubuli can restore spermatogene-
sis, even across species borders (Schlatt et al. 1999),
demonstrating that the seminiferous tubules are an
immunprivileged site.
Notwithstanding this, auto-antigens are also
expressed in cells located outside the inter-Sertoli
cell tight junctions, e.g., in the basal compartment of
the seminiferous epithelium and are recognized as
foreign (Pöllänen and Cooper 1994) ). Experiments
with allo- and xenografts placed directly into the tes-
ticular interstitial space can lack signs of degenera-
tion caused by a graft-versus-host reaction, indicating
absent or reduced immune reactions (Gores et al.
2003; Isaac et al. 2005). However, the testis is capa-
ble of a normal infl ammatory response, demonstrated
by its effective reaction to viral and bacterial infec-
tions, especially in the rete testis where testicular
sperm enter the efferent ducts of the epididymis
(Naito and Itoh 2008). This transition zone has a spe-
cial arrangement of Sertoli-like cells, forming a valve
with many macrophages patrolling the tissue. The
blood-testis-barrier is terminated in this area; subse-
quently, spermatozoa are no longer protected, whereas
later on, an epididymidal-blood-barrier is present
(Dube et al. 2007). This status is confi rmed by the
observation that certain forms of autoimmune orchi-
tis are fi rst manifested in the rete testis. Studies dem-
onstrated that the blood-testis-barrier at this transition
point is incomplete against humoral antibodies and
also intravenously injected horseradish peroxidase.
The immune privilege of the testis is not only due to
the blood-testis-barrier, but also depends on a specifi c
intratesticular regulation of the immune system func-
tion (Fig. 2.15). The predominant interstitial cell type
is the Leydig cell. In transplantation studies, rats pre-
treated with estrogen to suppress Leydig cell testos-
terone production promptly rejected intratesticular
allografts, indicating that high intratesticular testos-
terone concentrations seem to play an important role
in the maintenance of testicular immune privilege.
However, it remains unknown how testosterone may
mediate an anti-infl ammatory effect. In fact, the inter-
stitium contains a variety of immunocompetent cells
e.g., leukocytes, macrophages, monocytes, dendritic
cells, T and B lymphocytes, and mast cells.
Testicular macrophages are found as early as week
7 of gestation and probably originate from hematopoi-
etic precursor cells that migrate to the testis.
Macrophages proliferate in the testis during postnatal
life, probably under pituitary control, since hCG is
Fig. 2.15 Immunological compartments of the testis. Sertoli
cells (S) traverse the testicular tubules, keeping in close contact
with the germ cells. Together with the peritubular cells, they
form the seminiferous epithelium. The blood–testis barrier (tight
junctions) is built by tight junctions between neighboring S,
dividing the seminiferous tubules into a basal and adluminal
compartment. The interstitial space contains the Leydig cells (L)
and the immune cells such as macrophages (MP), dendritic cells
(D), mast cells (M), and T cells as well as blood vessels (BV)
with migrating leukocytes

2 Physiology of Testicular Function 39
able to increase the mitotic index of testicular mac-
rophages in rats. In the adult, human testis mac-
rophages represent about 25% of all interstitial cells.
Morphologically and biochemically they are similar
to macrophages resident in other tissues. Testicular
macrophages have a reduced capacity to excrete some
cytokines such as IL-1β and TNF-α compared to mac-
rophages from other tissues (Hayes and Crowley
1998). Furthermore, lipopolysaccharides, resembling
the surface of bacteria, given to immature and mature
mice resulted in enhanced levels of the testicular
cytokine IL-6 and constitutively elevated the produc-
tion of other anti-infl ammatory mediators (Elhija et al.
2005; Isaac et al. 2005). Interestingly, the expression
of pro-infl ammatory cytokines such as IL-1β and
TNF-α by testicular macrophages demonstrates the
testicular capability of an infl ammatory response. Up
to now two macrophage types have been distinguished
in the adult testis that differ in the expression of mark-
ers and infl ammatory mediators. In the rat, the ED2+
expressing macrophages do not participate in promot-
ing infl ammatory processes. They may take part in
maintaining the immune privilege as an immunoregu-
latory team player. However, the ED1+ ED2− mac-
rophages are involved in testicular infl ammatory
responses. During acute and chronic infl ammation the
infl ux of ED1+ monocytes change the equilibrium of
the macrophage population. The number of mononu-
clear cells increases in case of testicular disease.
In about 5% of testicular biopsies from infertile
men lymphoid cells surround the tubules with mark-
edly higher spermatogenic damage. Similarly, mono-
nuclear infi ltrates are often associated with carcinoma
in situ. Seminomas generally show a very conspicuous
infi ltration of immunocompetent cells. Diseases such
as mumps can be complicated by a severe testicular
infl ammation in 35% of cases.
The contribution of mast cells to the immune system
has often been underestimated, but recently the com-
plexity of these cells was shown and their involvement
in the innate and adaptive immune system (Gilfi llan
and Tkaczyk 2006; Stelekati et al. 2007). Mast cells
can release factors that act as mediators and as such are
capable of infl uencing disease induction and progres-
sion. In the brain these cells can change vascular per-
meability through factor release, thereby opening the
blood-brain-barrier, allowing the entry of activated T
lymphocytes and increased infl ammatory cell traffi c. In
men the testicular mast cell presence changes with age,
with numbers increasing slightly during infancy,
decreasing during childhood, and increasing again at
puberty. Their major released product, a serine protease
tryptase, is a mitogen for fi broblasts, enhancing the
synthesis of collagen, resulting in fi brosis, thickening,
and hyalinization of the tubuli walls.
Histological features of spermatogenic pathologies
are associated with increased numbers of mast cells
and are often found in men with infertility problems
(Apa et al. 2002; Sezer et al. 2005). Mast cell inhibi-
tors are benefi cial in the treatment of idiopathic oligo-
zoospermia and oligoasthenozoospermia too and can
prevent mast cell activation (Cayan et al. 2002; Hibi
et al. 2002). Human peritubular cells express receptors
for the mast cell products histamine and tryptase
(Albrecht et al. 2006). This may be helpful in the main-
tenance of an immunosuppressive phenotype not only
in the testis.
2.6 Testicular Androgens
Androgens are essential for the development and
function of testes, maturation of secondary sexual
characteristics, masculinization of the bone-muscle
apparatus, libido, and stimulation of spermatogenesis.
Physiological effects of androgens depend on differ-
ent factors such as number of androgen molecules,
distribution of androgens and their metabolites inside
the cell, interaction with the receptors, polyglutamine
number of the amino acid sequence in the androgen
receptor and receptor activation (Palazzolo et al.
2008). In turn, androgen concentrations in the blood
depend on the synthesis rate, balanced by metabolic
conversion and excretion. Androgens also exert rapid
non-genomic effects contributing to the physiological
actions (Lösel et al. 2003). Whereas genomic effects
take hours or days to produce their actions, rapid ste-
roid effects are activated within seconds or minutes.
These non-genomic actions are not removed by inhi-
bition of transcription or translation and are often acti-
vated by membrane-impermeant steroid conjugates.
However, rapid pathways of androgen action can
modulate transcriptional activity of androgen recep-
tors or other transcription factors (Rahman and
Christian 2007).
In men, testosterone is by far the most important
and abundant androgen in blood. More than 95% of

40 G. F. Weinbauer et al.
the existing androgens derive from the testis, which
synthesizes about 6–7 mg testosterone per day. Besides
the testes the remaining contribution to androgen pro-
duction derives mainly from the adrenals. The site of
androgen production in the testis is the Leydig cell.
Both synthesis and secretion are under regulation of
pituitary LH and local factors (Lei et al. 2001; Sriraman
et al. 2005).
Since Leydig cells cannot store androgens, de
novo biosynthesis takes place continuously. The start-
ing point for androgen synthesis is cholesterol, a fun-
damental substance of metabolism, with the typical
steroid ring conformation energetically compatible
with the transformation into androgens. Unlike most
cells that use cholesterol primarily for membrane
synthesis, Leydig cells have additional requirements
for cholesterol, because it is the essential precursor
for all steroid hormones. LH as the central regulatory
factor controls both steroidogenesis and Leydig cell
cholesterol homeostasis in vivo. Cholesterol can
either be incorporated by the cell through receptor-
mediated endocytosis from low-density lipoproteins
(LDL), or can be synthesized de novo within the
Leydig cell starting from acetyl-coenzyme A. In addi-
tion testosterone signaling regulates lipid homeosta-
sis in Leydig cells. It also affects the synthesis of
steroids and modulates the expression of genes
involved in de novo cholesterol synthesis (Eacker
et al. 2008). Cholesterol is stored in cytoplasmic lipid
droplets. The number of lipid droplets is inversely
related to the rate of androgen synthesis in the Leydig
cell, i.e., a high synthesis rate leads to a low content
of lipid droplets and vice versa.
2.6.1 Synthesis of Androgens
Androgen synthesis requires the conversion of choles-
terol to testosterone (Fig. 2.16). This transformation
goes through fi ve different enzymatic steps in which
the side chain of cholesterol is shortened through oxi-
dation from 27C to 19C. The steroidal A-ring assumes
a keto confi guration at position 3. The starting point
for the transformation of cholesterol into testosterone
is the shortening of the side chain through C 22 and C
20 hydroxylases, followed by cleavage of the bond
between C 20 and C 22, leading to production of preg-
nenolone. The steps following pregnenolone formation
occur in the endoplasmic reticulum either through the
∆4 or through the ∆5 pathway.
The designation ∆4 or ∆5 refers to the localization of
the double bond in the steroid. The D5 pathway is pre-
dominant over the ∆4 in human steroid synthesis. Along
the D4 pathway, pregnenolone is dehydrated to proges-
terone, a key biological substance. The ∆4 pathway pro-
ceeds to the intermediate 17α-hydroxyprogesterone. If
the side chain is removed at this stage, the intermediate
androstene-3, 17-dione is produced, which, through
further reduction at position C 17, is then transformed
into testosterone. In the ∆5 synthesis pathway, testos-
terone synthesis occurs through the intermediates
17-hydroxypregnenolone and dehydroepiandrosterone.
Cholesterol is the starting point for biosynthesis of
steroids, oxysterols and bile acids. After cholesterol,
an insoluble molecule is de novo synthesized or taken
up via the LDL receptor into the cell. Cholesterol for
steroidogenesis is stored in an ester form in lipid drop-
lets, which are hydrolyzed by LH activation of choles-
terol ester hydrolase. It has to be transported within the
cell to the mitochondria where it is imported into the
cristae of the mitochondria.
The discovery of the Steroidogenic Acute
Regulatory protein (StAR) and related proteins con-
taining StAR-related lipid transfer domains have
helped much to understand this limiting step of testos-
terone synthesis. StAR mRNA expression is triggered
by endocrine stimuli and is rapidly and widely distrib-
uted in steroidogenic tissues including the adrenals
and corpora lutea. StAR moves cholesterol from the
outer to the inner mitochondrial membrane, but acts
exclusively on the outer membrane. The precise mech-
anism by which StAR’s action stimulates the infl ux of
cholesterol remains unclear, but when StAR connects
to cholesterol it performs a conformational change
that opens a cholesterol-binding pocket (Miller 2007).
After a phosphorylation StAR interacts with voltage-
dependent anion channel 1 (VDAC1) on the outer
membrane, which processes the phospho-StAR to a
smaller intermediate. If VDAC1 is lacking, phospho-
StAR is degraded by cysteine proteases preventing the
mitochondrial membrane transport (Bose et al. 2008).
The physiological importance of StAR is highlighted
by the phenotype of patients with an inactivating
mutation of the StAR gene. These patients suffer from
life-threatening congenital adrenal hyperplasia as they
are unable to produce the necessary amounts of
steroids.

2 Physiology of Testicular Function 41
Fig. 2.16 (a) Steroid biosynthesis in the Leydig cell. Steroid biosynthesis is induced
by LH through activation of adenylyl cyclase. Starting material is cholesterol or ace-
tate. The C atoms of cholesterol are numbered in order to follow better the different
enzymatic modifi cations and their localization. StAR (Steroidogenic acute regulatory
protein) plays a key role during steroidogenesis. StAR is localized to the inner mito-
chondrial membrane and governs cholesterol transport. (b) Cholesterol uptake and
transport to the inner mitochondrial membrane. (c) Normally human steroidic cells
take up circulating low-density lipoproteins (LDL) through receptor-mediated endo-
cytosis, directing the cholesterol to endosomes. It may be esterfi ed by acyl-CoA cho-
lesterol transferase and stored in lipid droplets as cholesterol esters. Free cholesterol,
set free by the hormone-sensitive lipase, is probably bound by StarD4 for trans-
cytoplasmic transport to the mitochondrial membrane. StAR is responsible for the
rapid movement of cholesterol from the outer mitochondrial membrane to the inner
mitochondrial membrane, where it is converted by P450scc to pregnenolone

42 G. F. Weinbauer et al.
At the inner mitochondrial membrane site cyto-
chrome P450ssc (ssc = side chain cleavage) catalyzes
the conversion of cholesterol into pregnenolone. The
enzyme cytochrome P450ssc is responsible for the
different enzymatic reactions leading to the produc-
tion of pregnenolone. Like other steroid synthetic
enzymes, it belongs to the group of monooxygenases,
containing a prosthetic hemogroup as in hemoglobin
and is localized on the internal membrane of mito-
chondria. This reaction consists of three consecutive
monooxygenations requiring two electrons to activate
molecular oxygen; a 22-hydroxylation, 20-hydroxyla-
tion and the cleavage of the C 20–C 22 bond, yielding
pregnenolone and isocaproic aldehyde. Pregnenolone
diffuses across the mitochondrial membranes and is
transformed into testosterone through the enzyme
cytochrome P450 C 17, also belonging to the group of
monooxygenases and located in the endoplasmic
reticulum. The overall enzymatic system, however, is
not capable of transforming every molecule of preg-
nenolone into testosterone so that several intermedi-
ates are produced.
Testosterone is the main secretory product of the
testis, along with 5a-dihydrotestosterone (DHT),
androsterone, androstenedione, 17-hydroxyprogester-
one, progesterone and pregnenolone. The transforma-
tion of testosterone into DHT takes place principally in
the target organs, e.g., prostate. Androstenedione is
important as a precursor for the production of extrates-
ticular estrogens. Biologically active estradiol can be
produced as a result of extratesticular aromatization of
androstenedione to estrone that is subsequently reduced
to estradiol in peripheral tissues. Only a very small por-
tion of the testosterone produced is stored in the testis
and the androgen is mainly secreted in blood.
Testosterone concentrations in the testicular lym-
phatic circulation and in the venous blood are very
similar, but there are essential differences in the fl ow
rate and velocity of both systems. Therefore, transport
of testosterone in the general blood circulation occurs
mainly through the spermatic vein. Androgens diffuse
into interstitial fl uid and then enter testicular capillar-
ies or enter capillaries directly from Leydig cells that
are in direct contact with the testicular microvascula-
ture. The mechanism for testosterone transport from
the Leydig cell into the blood or lymph is not com-
pletely known. Probably lipophilic steroids distributed
within cells or small cell groups are released through
passive diffusion.
2.6.2 Testosterone Transport in Blood
During transport in plasma, testosterone is mainly
bound to albumin or to sex hormone binding globu-
lin (SHBG) which is produced by hepatocytes. A pro-
tein, the androgen binding protein (ABP), with similar
steroid-binding characteristics was found to be pro-
duced in the testis. SHBG is a β-globulin consisting of
different protein subunits. In rats it is expressed in
Sertoli cells and is secreted preferentially into the
seminferous tubules, migrates into the caput epididymis
where it is internalized by epithelial cells regulating
androgen-dependent mechanisms of sperm matura-
tion. Testicular SHBG isoforms are found in sperm
and released from these during the capacitation reac-
tion. Plasma SHBG is about 95 kDa in molecular
weight, 30% of which is represented by carbohydrate,
and possesses one androgen binding site per molecule.
Human testicular SHBG transcripts are expressed in
the germ cells and contain an alternative exon 1 sequence,
appearing to encode an SHBG isoform that is 4–5 kDa
smaller than plasma SHBG. The testosterone binding
capacity is also much lower compared to the plasma
SHBG (Selva et al. 2005).
In normal men, only 2% of total testosterone circu-
lates freely in blood, while 44% is bound to SHBG and
54% to albumin. The binding affi nity of testosterone to
albumin is about 100 times lower compared to SHBG.
However, since albumin concentration is much higher
than that of SHBG, the binding capacity of both pro-
teins for testosterone is about the same. The ratio of
testosterone bound to SHBG over free SHBG is pro-
portional to SHBG concentration. A direct measure-
ment of free testosterone is impractical in routine
practice, so that several equations are used to estimate
the free testosterone concentration in serum.
The main dissociation of testosterone from binding
proteins takes place in capillaries. The interaction of
binding proteins with the endothelial glycocalyx leads
to a structural modifi cation of the hormonal binding site
and thereby to a change in affi nity. As a result testoster-
one is set free and can diffuse freely into the target cell
or binds together with SHBG to megalin, a cell importer
protein (Hammes et al. 2005). Megalin is expressed in
sex-steroid target tissues and is a member of the low
density lipoprotein receptor superfamily of endocytic
proteins. In the serum 98–99.5% of the sex steroids are
protein-bound and endocytosis is quantitatively more
relevant for tissue delivery of biologically active steroid

2 Physiology of Testicular Function 43
hormones than free diffusion. Until now several differ-
ent ways have been described how steroids can enter
the target cells and it is still under debate which of these
are the most relevant pathways to take up all kinds of
steroids.
SHBG not only binds testosterone but also estradiol.
The type of binding is infl uenced by the different SHBG
isoforms, but generally testosterone binds threefold
higher than estradiol to SHBG. For example, it could
be demonstrated that post-translational changes in the
carbohydrate structure of SHBG can lead to different
binding affi nity of the protein to testosterone or estra-
diol. SHBG concentration in serum is under hormonal
regulation and primarily regulated through opposing
actions of sex steroids on hepatocytes, estrogen stimu-
lates and androgen inhibits SHBG production. Other
hormones such as thyroid hormones are also potent
stimulators of SHBG production. SHBG concentration
in men is about one third to one half of the concentra-
tion found in women. In normal, healthy men with an
intact hypothalamo-pituitary-testicular axis, an increase
in plasma concentrations of SHBG leads to an acute
decrease of free testosterone and simultaneous stimula-
tion of testosterone synthesis, persisting until achieve-
ment of normal concentrations. SHBG concentrations
can be elevated in hypogonadal men.
2.6.3 Extratesticular Metabolism
of Testosterone
Testosterone is a precursor of two important hormones:
through 5α-reduction it gives rise to the highly bio-
logically (three to sixfold compared to testosterone)
active hormone 5a-dihydrotestosterone (DHT), and
through aromatization to estradiol. The half-life of
testosterone in plasma is only about 12 min. Estrogens
infl uence testosterone effects by acting either synergis-
tically or antagonistically. Moreover, estrogens have
other specifi c effects which were originally described
to be typical of testosterone. It has been found that
inactivating mutations of the estrogen receptor or aro-
matase, preventing estrogen action on the bones, result
in continuous linear growth and lack of epiphyseal
closure.
Low levels of bioavailable estrogen and testoster-
one are strongly associated with high bone turnover,
low bone mineral density and high risk of osteoporotic
fractures. In aromatase-knockout and estrogen recep-
tor-knockout male mice an association between
impaired glucose tolerance with insulin resistance and
lack of estrogens with elevated testosterone concentra-
tions have been found (Takeda et al. 2003). It also
seems that aromatase defi ciency in men is associated
with the occurrence of insulin resistance and diabetes
mellitus type 2 during high-dose testosterone treat-
ment (Maffei et al. 2004). The impairment of the estro-
gen to testosterone ratio is thought to be responsible
for the development of impaired glucose tolerance and
insulin resistance in aromatase defi ciency patients
receiving testosterone replacement therapy. An imbal-
ance in the ratio of estrogen to androgen tissue levels is
also postulated as a major cause in the development of
gynecomastia. Furthermore recent investigations have
shown a local neuroprotective effect of newly aroma-
tized estradiol on the brain.
Reduction of testosterone to DHT occurs in the endo-
plasmic reticulum through the enzyme 5α-reductase
which is in