Article

In search of an improved injection technique

for the clinical application of spermatogonial
stem cell transplantation

Katrien Faes a,*, Tony Lahoutte b,c, Anne Hoorens d, Herman Tournaye a,e,
Ellen Goossens a
a
Biology of the testis (BITE), Vrije Universiteit Brussel (VUB), Laarbeeklaan 103, 1090 Brussels, Belgium
b
In vivo Cellular and Molecular Imaging laboratory, Vrije Universiteit Brussel (VUB), 1090 Brussels, Belgium
c
Nuclear Medicine Department, Universitair Ziekenhuis Brussel, Vrije Universiteit Brussel (VUB), 1090 Brussels, Belgium
d
Department of Pathology, Universitair Ziekenhuis Gent, 9000 Gent, Belgium
e
Centre for Reproductive Medicine, Universitair Ziekenhuis Brussel, Laarbeeklaan 101, 1090 Brussels, Belgium

Katrien Faes started her PhD in the research group Biology of the Testis at the Vrije Universiteit, Brussel in 2011
and successfully defended in May 2016. During this PhD she focused on the short-term preservation of human
testicular tissue and the translation of the SSCT towards a clinical application.

KEY MESSAGE
Donor cells were observed in the seminiferous tubules after infusion using hydrostatic pressure or an infu-
sion pump, but further optimisation of the infusion method is required.

A B S T R A C T

When fertility is impaired by anticancer treatment, spermatogonial stem cell transplantation (SSCT) could be used as a fertility restoration technique
later on in life. Previously, we have demonstrated that a testicular cell suspension could be injected into a human cadaver testis, however, leakage to
the interstitium was observed. In this study, injection of mouse testicular cells at an injection height of 50 cm (hydrostatic pressure) or via an auto-
mated injection pump (1400 μl, 2600 μl and 3000 μl) was evaluated. Significant difference in the filled radioactive volume was reached between the group
in which 1400 μl was injected with an infusion pump and the groups in which 2600 μl (P = 0.019) or 3000 μl (P = 0.010) was injected. In all experimental
groups green fluorescent protein positive (GFP+) cells were observed in the seminiferous tubules. In conclusion, a lower injection height did not resolve
the leakage of the injected cells to the interstitium. Using the infusion pump resulted in more efficient filling of the seminiferous tubules with lower
interexperimental variability. Although leakage to the interstitium was still observed, with further optimisation, the use of an infusion pump for clinical
application is advantageous.
© 2016 Reproductive Healthcare Ltd. Published by Elsevier Ltd. All rights reserved.

* Corresponding author.
E-mail address: kfaes@vub.ac.be (K Faes).
http://dx.doi.org/10.1016/j.rbmo.2016.12.007
1472-6483/© 2016 Reproductive Healthcare Ltd. Published by Elsevier Ltd. All rights reserved.
292 REPRODUCTIVE BIOMEDICINE ONLINE 34 (2017) 291–297
Introduction
With an increasing chance to survive cancer, awareness on the
quality of life after being cured is growing. Since cancer treatment
can cause spermatogonial stem cell (SSC) loss which may induce
sterility, fertility preservation is becoming of greater interest. While
adult men have the option to store a semen sample before
starting their treatment, prepubertal boys cannot benefit from this
option.
Techniques have been developed in various animal models
to restore fertility. These include spermatogonial stem cell
transplantation (SSCT) and testicular tissue grafting (Brinster and
Avarbock, 1994; Brinster and Zimmermann, 1994; Van Saen et al.,
2009). Worldwide, more and more fertility centres start to bank
testicular tissue of prepubertal boys who are at risk of losing their
fertility because of gonadotoxicity (Sadri-Ardekani and Atala, 2014;
Struijk et al., 2013). In total, more than 260 testicular biopsies (data
available up to December 2012) have been cryopreserved in Europe
and it is only a matter of time until the first patients return to the
clinic in the hope that their fertility can be restored (Picton et al.,
2015).
Unfortunately, although (xenogenic) SSCT has been proven
successful in rodents (Brinster and Zimmermann, 1994; Clouthier
et al., 1996; Nagano and Brinster, 1998; Nagano et al., 2001;
Ogawa et al., 1999) this technique cannot be implemented as such
in animals with larger testes because of the very resistant lamina
propria and the coiling of the seminiferous tubules. In rat, bovine,
monkey and human testes, microinjection into the seminiferous
tubules or the efferent ducts may be feasible but requires micro-
surgery. Yet, rete testis injection under ultrasound guidance is
successful (Schlatt et al., 1999) and has been implemented in differ-
ent animal models (Honaramooz et al., 2002, 2003; Izadyar et al.,
2003; Kaul et al., 2010; Kim et al., 2008; Mikkola et al., 2006). An
advantage of this injection method is that invasive surgery can be
avoided. Although studies are scarce, some attempts have been
made to inject a dye or a cell suspension into human testes via
ultrasound guidance and hydrostatic pressure (Brook et al., 2001;
Faes et al., 2013; Ning et al., 2012; Radford et al., 1999; Schlatt
et al., 1999).
In our previous study, we could fill 20% of the seminiferous tubules
but cells were also detected in the interstitium of the injected ca-
daveric human testes. This leakage could have been caused by a too
high hydrostatic pressure during injection. To resolve this issue, a
lower injection height of 50 cm (instead of 75 cm) above the testis was
tested.
Previous experiments in goats and pigs have shown that hydro-
static pressure provides a flow rate of 0.5 − 1 ml/min (Honaramooz
et al., 2003). The lack of controlling the flow rate causes variation
amongst the different experiments, which may influence the success
rate of the transplantation. Therefore, in this study, we also tried an
infusion pump used to perform organ perfusions. The infusion pump
can inject 0.001 μl/h to 2120 ml/h, depending on the syringe type. As
the flow rate is machine controlled, variation is only caused by the
recipients and the handling. We have chosen a flow rate of 0.5 ml/min
to mimic the lowest hydrostatic pressure. Since the low pressure
might allow the possibility to inject more of the testicular cell sus-
pension, higher injection volumes were also tested as injecting a higher
amount of SSC may increase the chance of successful fertility
restoration.
Materials and methods
Ethical approval
Approval for this study was given by the ethical committee of the UZ
Brussel (approval number 2011–278) on 22 December 2012 and the
Animal Care and Use Committee of the Vrije Universiteit Brussel (ap-
proval number 2012–216–4) on 24 September 2012.
Tissue source
Cadaver testes (n = 20) were obtained from autopsied bodies at the de-
partment of pathology of the Universitair Ziekenhuis (UZ) Brussel. The
cadaver testes were kept in physiological saline and refrigerated until
use. The time elapsed between cardiac arrest and time of infusion, also
known as the ischemic time, was on average 31 ± 17 h (Table 1).
Injection of cells in human testis
Donor testes were obtained from mice bred in the animal centre of
the Vrije Universiteit Brussel according to the following breeding
model: male C57BL inbred (Iffracredo, Brussels, Belgium) x female
Sv129 inbred green fluorescent protein positive (GFP+) (gift from
Whithead institute for Biomedical research, Cambridge, MA, USA).
In the latter strain GFP is controlled by the β-actin promoter and is
expressed in all cell types. In this study, adult male GFP+ F1 hybrids
were used (n = 13). The mice were killed by cervical neck disloca-
tion. After disinfection with Hibitane® (P685040; SSL Healthcare,
Anderlecht, Belgium), the testes were removed via an incision in the
abdominal cavity. After decapsulation, the testes were enzymati-
cally digested to a cell suspension and labelled with 99mTc-HMPAO
according to the previously described method (Faes et al., 2013). To
avoid cell clumping, 0.7 mg/ml deoxyribonuclease I (DNase I;
04716728001; Roche Life Sciences, Belgium) was added to the cell
suspension after digestion. Cell viability after digestion was 94 ± 3%.
In order to trace the cell suspension during ultrasound injection, the
cell suspension was mixed in a 1:1 ratio with microbubbles,
perfluoropropane-filled albumin microspheres (Optison; Mallinckrodt
Medical, Hennef, Germany). Due to production problems of the
microbubbles during the course of the experiments, visualisation by
ultrasound was executed by adding ExEm Foam (ExEm Foam kit, de
Smit medical, UK) to the cell suspension. In a preliminary experi-
ment the viability of the cells and visualisation capacity during
ultrasound injection was assessed. Viability of the cells remained
87 ± 3% after 16 h exposure to the ExEm Foam kit. Because of its
viscosity, ExEm Foam was diluted by factor 10 in milliQ water before
addition in a 1:1 ratio to the cell suspension. Preliminary experi-
ments with ExEm Foam showed that the change of microbubbles to
ExEm Foam did not affect the outcome of the study.
In a first set of experiments, cadaver testes (n = 6) were injected
with a 23G needle, since smaller needle sizes were not able to pen-
etrate the tunica (Ning et al., 2012), via the rete testis under ultrasound
guidance using 1400 μl of 99mTc labelled-GFP+ testicular cells (10 million
cells/ml), based on previous research (Faes et al., 2013). In that study,
leakage occurred into the interstitial space presumably due to a too
high hydrostatic pressure. Therefore, to reduce this leakage, the in-
jection height (height difference between the water reservoir and testis)
was lowered from 75 cm to 50 cm, based on the successful injec-
tion in goat testes (Honaramooz et al., 2003). The cell suspension was
 REPRODUCTIVE BIOMEDICINE ONLINE 34 (2017) 291–297 293

added to the tubing which was connected to the syringe. Between the
cell suspension and the water, an air gap was left. The infusion was
stopped as soon as the air gap reached the needle. The transfusion
lasted approximately 1 − 2 min.
In the experiments with the infusion pump (Supplemental Figure S1,
Model 540060, TSE Systems GmbH, Germany), the cell suspension
(1400 μl [n = 6], 2600 μl [n = 4] or 3000 μl [n = 4]) was aspirated by
the pump before injection. Injection was performed at a flow rate of
0.5 ml/min and was stopped when the entire cell suspension had
entered the testis.
A negligible rest volume of 10 μl remained in the tip of the needle.
Analysis
After infusion and a waiting period of 30 min, the testes were scanned
by Single-photon emission computed tomography (SPECT) imaging to
verify the presence of the radioactive cell suspension inside the iso-
lated human testis. SPECT images were acquired and reconstructed
as previously described (Faes et al., 2013). After SPECT imaging, the
testes were divided in 4 − 5 mm thick slices, fixed in acidified formal
alcohol (AFA; 10056710; Labonord), embedded in paraffin and cut in 5 μm
thick sections. Per testis slice, three sections with intervals of 100 μm
were evaluated by GFP immunostaining (Faes et al., 2013). The per-
centage of seminiferous tubules containing GFP+ cells was calculated
(dividing the number of seminiferous tubules containing GFP+ cells/
total number of evaluated seminiferous tubules) (Figure 1A) as well as
the percentage of positivity in the interstitium (the number of sections
with GFP+ cells in the interstitium/total number of evaluated sections)
(Figure 1B, C). Since this only states the absence or presence of
GFP+ cells, the amount of GFP+ cells throughout the interstitium
was estimated per section and categorised as less than 10%, 10 − 50%,
50 − 80% and 80 − 100% of interstitial area filled with GFP+ cells.
To evaluate how far the GFP+ cells were able to spread throughout
the testis after infusion, the scope was evaluated as previously de-
scribed (Faes et al., 2013). The depth of the positive slide at the deepest
level was subtracted from the depth of the slide at the lowest level, and
divided by the total testis size. This gives an indication how far the in-
jected cell suspension was able to spread into the infused testes.
Statistics
Data are expressed as mean ± standard deviation (SD). All data were
analysed by the Mann-Whitney U-test (GraphPad Prism, version 5,
GraphPad Software Inc., USA). Statistical significance was set at P < 0.05.
Results
Donor cell infusion under hydrostatic pressure at 50 cm
of height
When cadaver testes were injected under hydrostatic pressure (50 cm),
a relatively good coverage was seen by SPECT, which was confirmed
by the amount of radioactive labelled cells (3365.3 ± 1282.5 mm3) and
the expansion coefficient (2.4 ± 0.9; Table 1; Figure 2A). The percent-
age of radioactivity bound to the cell suspension is listed in Table 2.

Table 1 – Human donor information and SPECT imaging data.

Infusion Experiment Ultrasound Age Chemotherapy Total Histology Volume Expansion
method visualisation (y) ischemia filled (mm3) coefficient
time (h)
1400 μl − 50 cm 1 microbubbles 82 No 17 Sclerotic 1501.8 1.1
height 2 microbubbles 2664.1 1.9
3 microbubbles 81 No 6 Spermatogenesis 3055.1 2.2
4 microbubbles 3508.1 2.5
5 ExEm Foam 82 Unknown 12 Spermatogenesis 4278.3 3.1
6 ExEm Foam 89 No 56 Spermatogenesis 5184.4 3.7
Mean ± SD 83 ± 3 23 ± 23 3365.3 ± 1282.5 2.4 ± 0.9
1400 μl − pump 1 ExEm Foam 68 Yes 24 Spermatogenesis 4318.0 3.1
2 ExEm Foam 89 No 56 Spermatogenesis, some 5890.4 4.2
sclerotic tub
3 ExEm Foam 73 Yes 43 Sclerotic 4360.9 3.1
4 ExEm Foam 4059.5 2.9
5 ExEm Foam 87 No 25 Spermatogenesis 4493.0 3.2
6 ExEm Foam 4526.8 3.2
Mean ± SD 80 ± 7 37 ± 15 4608.1 ± 649.7(a.b) 3.3 ± 0.5 (a.c)
2600 μl − pump 1 ExEm Foam 78 Yes 16 Spermatogenesis, some 7063.3 2.7
2 sclerotic tub 7568.9 2.9
3 ExEm Foam 74 Unknown 48 Spermatogenesis 5780.9 2.2
4 ExEm Foam 89 No 44 Sclerotic 7218.6 2.8
Mean ± SD 80 ± 6 36 ± 17 6907.9 ± 780.5(a) 2.7 ± 0.3(a)
3000 μl − pump 1 ExEm Foam 74 Unknown 48 Spermatogenesis 6823.8 2.3
2 ExEm Foam 82 Unknown 48 Spermatogenesis /sclerotic 6706.8 2.2
3 ExEm Foam 7767.1 2.6
4 ExEm Foam 89 No 44 Sclerotic 8227.9 2.7
Mean ± SD 82 ± 5 47 ± 2 7381.4 ± 737.4(b) 2.5 ± 0.2(c)
Note: Total ischemia time is the time elapsed between cardiac death and the infusion experiment.
Values in the same column with different superscript letters are significantly different. (a) P = 0.019, (b) P = 0.010, (c) P = 0.005 by Mann Whitney U-test.
SD = standard deviation; SPECT = single-photon emission computed tomography.
294 REPRODUCTIVE BIOMEDICINE ONLINE 34 (2017) 291–297

Figure 1 – Histology and immunohistochemistry stainings. A) Presence of GFP+ cells (brown cells) in the seminiferous tubules (Scale
bar = 100 μm). B) Presence of GFP+ cells in the interstitium (brown cells, indicated by black arrows, Scale bar = 100 μm)). GFP+ mouse
testicular tissue was used as positive control (PC; Scale bar = 200 μm). GFP+ mouse testicular tissue was used as negative control (NC;
Scale bar = 200 μm). C-D) Comparison of GFP+ after hydrostatic injection of a 1400 μl cells/ml cell suspension at 50 cm above the testis and
after injection of a 1400 μl, 2600 μl and 3000 μl cells/ml suspension with the infusion pump. C) Proportion of interstitium filled with GFP+
cells. No significant difference was observed. D) Overall scope of GFP+ cells found in the whole injected testes. The proportion of
interstitium containing GFP+ cells differed significantly (* P = 0.020 by Mann Witney U-test) between the 50 cm height and 1400 μl pump
groups. All data presented as mean ± SD. GFP = green fluorescent protein.

Figure 2 – Single-photon emission computed tomography (SPECT) analysis. SPECT images acquired after injection of 1400 μl of a 10
million cells/ml suspension under hydrostatic pressure, provided by placing a water reservoir at 50 cm above the testis (A) or after
injection of different volumes (B: 1400 μl; C: 2600 μl; D: 3000 μl) of a 10 million cells/ml suspension with an infusion pump. The pictures
shown are representative for the different conditions tested.
 REPRODUCTIVE BIOMEDICINE ONLINE 34 (2017) 291–297 295

GFP+ cells were observed in less than 10% (59.1 ± 10.7%) or in

Table 2 – Radioactivity bound to the cells.
10 − 50% (9.0 ± 12.8%) of the total interstitial area per section
Infusion Experiment Sup + Pellet Pellet Percentage (Figure 1C).
method (mCi) (mCi)
1400 μl – 50 cm 1 5.3 2.2 41.1 Testing an infusion pump for human SSCT
height 2 5.3 2.1 39.1
3 9.7 4.0 41.1
To reduce variability during injection, an infusion pump was tested
4 10.4 4.1 39.5
5 12.3 3.8 30.6
at a flow rate of 500 μl/min to resemble the hydrostatic pressure.
6 12.5 5.1 40.5 In a first series, 1400 μl of a 10 million cells/ml cell suspension
Mean ± SD 9.3 ± 3.3 3.5 ± 1.2 38.6 ± 4.0 was injected to compare the pump function to the hydrostatic injec-
1400 μl – pump 1 12.1 2.8 22.8 tion at 50 cm height. Compared with the injection under hydrostatic
2 12.8 5.6 43.5 pressure at 50 cm height, no difference was seen in the number of
3 12.0 4.4 36.9
radioactive labelled cells (4608.1 ± 649.7 mm3) or expansion coeffi-
4 12.1 5.0 41.6
cient (3.3 ± 0.5) although less variability was observed when using
5 9.9 4.3 42.7
6 9.7 3.9 40.4 the infusion pump (Table 1; Figure 2B). Histological evaluation was
Mean ± SD 11.4 ± 1.3 4.3 ± 1.0 38.0 ± 7.8(a,b) performed on 23 ± 3 sections and presence of donor cells was shown
2600 μl – pump 1 11.2 7.1 63.2 in 11.5 ± 15.1% of the examined seminiferous tubules (Table 3). In
2 10.7 6.9 64.6 three out of the six injected testes no positive cells were observed
3 10.4 6.0 58.3 in the tubules (Table 3). In 48.8 ± 34.2% of the evaluated sections,
4 11.3 7.7 68.4
GFP+ cells were observed in less than 10% of the interstitial area
Mean ± SD 10.9 ± 0.4 6.9 ± 0.7 63.6 ± 4.2(a),
(Figure 1C).
3000 μl – pump 1 10.0 5.6 56.3
2 11.9 6.9 57.6
3 11.5 6.6 57.3 Increasing the distribution of donor cells after injection with
4 11.0 6.8 61.4 the infusion pump
Mean ± SD 11.1 ± 0.8 6.5 ± 0.6 58.2 ± 2.2(b),
Note: Sup = supernatant; SD = standard deviation. (a,b) P = 0.010 by Mann- To reach a larger distribution, larger injection volumes (2600 μl and
Whitney U-test. 3000 μl) were tested. In comparison to the condition in which 1400 μl
of a 10 million cells/ml cell suspension was injected with the infu-
sion pump significant difference was reached for the higher volumes
For the histological evaluation, 26 ± 3 sections were evaluated per (6907.9 ± 780.5 mm3 [P = 0.019] and 7381.4 ± 737.4 mm3 [P = 0.010],
testis. GFP+ cells were observed in all testes, more specifically, respectively). Significant difference was also observed when com-
4.1 ± 2.6% of the examined seminiferous tubules contained GFP+ cells paring the expansion coefficients, 2.7 ± 0.3 (P = 0.019) and 2.5 ± 0.2
(Table 3). Total percentage of sections with GFP+ cells in the inter- (P = 0.005) for the injected volumes of 2600 μl and 3000 μl, respec-
stitium was 68.8 ± 7.9%. Looking in more detail to the interstitium, tively (Table 1; Figure 2C, D).
In the group where 2600 μl of a 10 million cells/ml cell suspen-
sion was injected, 34 ± 12 sections were analysed for histology.
Presence of donor cells was shown in 1.8 ± 1.6% of the examined
Table 3 – Seminiferous tubules containing GFP+ cells.
tubules (Table 3) and in 72.7 ± 5.2% of the sections, GFP+ cells were
Infusion Experiment No. Total Percent Mean ± SD found in less than 10% of the interstitial area (Figure 1C). In the ex-
method positive number of positive (%)
periments testing the injection of 3000 μl of a 10 million cells/ml cell
tubules tubules tubules
suspension, 25 ± 5 sections were analysed for histology. GFP+ cells
50 cm 1 42 1364 3.1 4.1 ± 2.6 were observed in 7.6 ± 5.5% of the examined tubules (Table 3). In
height 2 45 1018 4.4
68.3 ± 12.2% of the sections GFP+ cells were retrieved in less than
3 17 2044 0.8
10% of the interstitial area of each examined section (Figure 1C). In
4 30 1467 2.0
5 95 1439 6.6 one experiment, no GFP+ cells were observed in the seminiferous
6 162 2159 7.5 tubules (Table 3). The variation in the percentage of seminiferous
1400 μl – 1 0 - 0.0 11.5 ± 15.1 tubules containing GFP+ cells is most likely caused by interpatient
pump 2 124 1587 7.8 differences.
3 1747 5413 32.3
4 1625 5613 29.0
Overall scope of GFP-positivity throughout the testes
5 0 - 0.0
6 0 - 0.0
2600 μl – 1 6 898 0.7 1.8 ± 1.6 The scope of seminiferous tubules containing GFP+ cells after injec-
pump 2 100 2585 3.9 tion of 1400 μl of a 10 million cells/ml suspension by hydrostatic
3 7 2007 0.3 pressure at 50 cm of height was 69.6 ± 8.6% (Figure 1D). No signifi-
4 104 4916 2.1 cant difference was observed when the infusion pump (56.5 ± 36.4%)
3000 μl – 1 0 - 0.0 7.6 ± 5.5
was used. However, significant difference was observed between the
pump 2 382 5221 7.3
scope of GFP+ cells in the interstitium of testes injected with the in-
3 442 3483 12.7
4 513 4911 10.4 fusion pump (1400 μl; 57.6 ± 44.5%) and the injection at 50 cm of height
(90.2 ± 4.9%; P = 0.020). When injecting higher volumes, the scope
GFP = green fluorescent protein; SD = standard deviation. ‘-‘ not counted.
of seminiferous tubules containing GFP+ cells amounted 56.0 ± 21.2
296 REPRODUCTIVE BIOMEDICINE ONLINE 34 (2017) 291–297
and 65.4 ± 43.6% for 2600 μl and 3000 μl respectively (Figure 1D). No
significant differences were found in proportion of interstitial area
containing GFP+ cells (88.5 ± 5.1%; 89.6 ± 7.0% for injections with
2600 μl or 3000 μl, respectively) when comparing the group in which
1400 μl was injected and the groups where 2600 μl or 3000 μl was
injected (Figure 1D).
Discussion
The aim of this study was to reduce leakage into the interstitial com-
partment and to decrease the high interexperimental variation seen
in our previous study during infusion of a testicular cell suspension
into human cadaver testes by altering the injection pressure and ex-
ploring use of the infusion pump (Faes et al., 2013). The implementation
of an infusion pump would also increase the applicability of the SSCT
in the clinic.
Although, when using the infusion pump SPECT did not show in-
creased volumes filled with radioactive cells, the variation between
individual experiments was reduced. Moreover, when using the in-
fusion pump fewer cells leaked into the interstitium, while the
proportion of seminiferous tubules containing GFP+ cells remained
the same. Increasing the injected volume led to a larger volume filled
(SPECT) but did not increase the percentage of GFP+ seminiferous
tubules.
In mice, SSCT was successful after microinjection of cells in the
seminiferous tubules, in the rete testis, or in the efferent duct (Ogawa
et al., 1997). Yet in 1999, Schlatt et al. observed that the larger the
recipient’s testis, the harder it is to infuse it. In their study, infusion
via the rete testis proved to be the best injection route for human
testes, although the success rate was low and only the seminifer-
ous tubules close to the rete testis were filled with the dye. They
concluded that the use of ultrasound-guided injection for a human
SSCT application was plausible as it also avoids invasive surgery
(Schlatt et al., 1999).
The percentage of seminiferous tubules containing injected dye
or cell suspension after injection varied from 10 to 50% (Brook et al.,
2001; Honaramooz et al., 2002; Kaul et al., 2010) although in most
transplantation studies using larger testes, real histological proof was
lacking (Brook et al., 2001) and leakage was observed into the inter-
stitium (Schlatt et al., 1999). However, the distribution of the injected
dye or cells throughout the whole testis has never been quantified
before (Hermann et al., 2012; Honaramooz et al., 2002; Ning et al.,
2012).
Compared with literature, in all the tested experimental condi-
tions a number amount of positive cells was observed in the
seminiferous tubules. Yet, the large distribution throughout the testes
indicates that the cells were less concentrated at one place and were
able to translocate further within the testis. This could be explained
by the addition of the DNase which prevented cell clotting, the in-
troduction of the waiting time before SPECT analysis and the
modifications made in the infusion method. Yet, care should be taken
with the interpretation of the distribution throughout the testis, due
to the strong coiling of the seminiferous tubules in the human testis.
Since the seminiferous tubule is organised in a non-branched loop
of which both ends connect to the rete testis, donor cells could have
travelled from the rete testis through nearly the whole seminifer-
ous tubule and settled when approaching the end of the loop.
Therefore, although found in tubules ‘close to the rete testis’, these
cells have translocated over a longer distance than cells located at
the bottom of the loop of the same seminiferous tubule (‘far from the
rete testis’).
The fact that GFP+ cells were observed in the interstitial com-
partment shows that leakage still occurred. This may result from a
too high pressure during injection, puncture of the rete testis or ex-
isting damage in the testis. Further adjustment of the injection flow
rate may avoid this leakage.
Statistical comparison with our previous results where donor cells
were infused in cadaver testes under a hydrostatic pressure at 75 cm
of height is not possible since besides the altering of the height or
the use of the infusion pump, the addition of DNase and the intro-
duction of the waiting time could have influenced the outcome of the
study.
Previously, we concluded that for research purposes, age of the
acceptor testes should be at least 65 years to resemble gonadotoxic-
treated testes (Faes et al., 2013). Unfortunately, although using this
cut-off in the present study, interpatient differences regarding ongoing
spermatogenesis caused a high standard deviation. As it was not pos-
sible to assess the presence of spermatogenesis at the time of infusion,
this was an unknown factor in the performed experiments.
As shown by Hermann et al. the number of cells that can be trans-
planted into busulfan-treated prepubertal and adult rhesus macaque
testes might differ. In their study, around 1000 μl could be intro-
duced into adult testes and only 200 μl into prepubertal testes (Hermann
et al., 2012). This can be explained by the fact that the testicular volume
of prepubertal testes is smaller. Since the optimal timeframe of re-
introducing SSC to the patient is not known yet, this should be addressed
in a relevant human model like non-human primates. Although in
murine SSCT no morphological or major (epi)genetic changes oc-
curred (Goossens et al., 2009, 2010, 2011; Wu et al., 2012), the safety
should also be addressed in higher animal models.
The first clinical trial for in-vivo transplantations has been re-
ported out in 1999 (Radford et al., 1999). In this study frozen/thawed
cell suspensions were transplanted in five adult men after success-
fully completing the cancer treatment. Unfortunately, the outcome
of this study was never published.
In conclusion, injected cells were observed in the seminiferous
tubules after infusion at a height of 50 cm under hydrostatic pres-
sure or after the use of the infusion pump. However, further
optimisation of the infusion method is required since lowering the in-
jection height to 50 cm or using an infusion pump did not prevent the
leakage to the interstitium. Nevertheless, the use of the infusion pump
showed less variability in the amount of radioactive cells in the testis.
Since the infusion pump provides a more standardized con-
trolled injection, it should be optimised further, preferentially in large
living animals, before introduction in the clinic is possible.
Acknowledgement
The authors would like to thank Mr F Van Haelst, Mr P Hilven and
Miss C Peleman for technical assistance. This study was supported
by grants from Ferring and grants from the Vrije Universiteit Brussel.
Appendix: Supplementary material
Supplementary data to this article can be found online at
doi:10.1016/j.rbmo.2016.12.007.
 REPRODUCTIVE BIOMEDICINE ONLINE 34 (2017) 291–297
A R T I C L E I N F O
Article history:
Received 26 August 2016
Received in revised form 7 December 2016
Accepted 8 December 2016
Declaration: The authors report no
financial or commercial conflicts of
interest.
Keywords:
Fertility restoration
Human
Infusion pump
Seminiferous tubule
SSC transplantation
Testis
REFERENCES
Brinster, R., Avarbock, M., 1994. Germline transmission of donor
haplotype following spermatogonial transplantation. Proc Natl Acad
Sci U S A 91, 11303–11307.
Brinster, R., Zimmermann, J., 1994. Spermatogenesis following male
germ-cell transplantation. Proc Natl Acad Sci U S A 91, 11298–11302.
Brook, P., Radford, J., Shalet, S., Joyce, A., Gosden, R., 2001. Isolation of
germ cells from human testicular tissue for low temperature
storage and autotransplantation. Fertil. Steril. 75, 269–274.
Clouthier, D., Avarbock, M., Maika, S., Hammer, R., Brinster, R., 1996.
Rat spermatogenesis in mouse testis. Nature 381, 418–421.
Faes, K., Tournaye, H., Goethals, L., Lahoutte, T., Hoorens, A., Goossens,
E., 2013. Testicular cell transplantation into the human testis. Fertil.
Steril. 100, 981–988.
Goossens, E., De Rycke, M., Haentjens, P., Tournaye, H., 2009. DNA
methylation patterns of spermatozoa and two generations of
offspring obtained after murine spermatogonial stem cell
transplantation. Hum. Reprod. 24, 2255–2263.
Goossens, E., de Vos, P., Tournaye, H., 2010. Array comparative genomic
hybridization does not show genetic alterations in spermatozoa and
offspring generated after spermatogonial stem cell transplantation
in the mouse. Hum. Reprod. 25, 1836–1842.
Goossens, E., Bilgec, T., Van Saen, D., Tournaye, H., 2011. Mouse germ
cells go through typical epigenetic modifications after intratesticular
tissue grafting. Hum. Reprod. 26, 3388–3400.
Hermann, B., Sukhwani, M., Winkler, F., Pascarella, J., Peters, K., Valli,
H., Rodriguez, M., Ezzelarab, M., Daargo, G., Peterson, K.,
Masterson, K., Ramsey, C., Ward, T., Lienesch, M., Volk, A., Cooper,
D., Thomson, A., Kiss, J., Penedo, M., Schatten, G., Mitalipov, S.,
Orwig, K., 2012. Spermatogonial stem cell transplantation into
rhesus testes regenerates spermatogenesis producing functional
sperm. Cell Stem Cell 11, 715–726.
Honaramooz, A., Megee, S., Dobrinski, I., 2002. Germ cell
transplantation in pigs. Biol. Reprod. 66, 21–28.
Honaramooz, A., Behboodi, E., Blash, S., Megee, S., Dobrinski, I., 2003.
Germ cell transplantation in goats. Mol. Reprod. Dev. 64, 422–428.
297
Izadyar, F., Den Ouden, K., Stout, T., Stout, J., Coret, J., Lankveld, D.,
Spoormaker, T., Colenbrander, B., Oldenbroek, J., Van der Ploeg, K.,
Woelders, H., Kal, H., De Rooij, D., 2003. Autologous and
homologous transplantation of bovine spermatogonial stem cells.
Reproduction 126, 765–774.
Kaul, G., Kaur, J., Rafeeqi, T., 2010. Ultrasound guided transplantation
of enriched and cryopreserved spermatogonial cell suspension in
goats. Reprod Domest Anim 45, 249–254.
Kim, Y., Turner, D., Nelson, J., Dobrinski, I., McEntee, M., Travis, A.,
2008. Production of donor-derived sperm after spermatogonial stem
cell transplantation in the dog. Reproduction 136, 823–831.
Mikkola, M., Sironen, A., Kopp, C., Taponen, J., Sukura, A., Vilkki, J.,
Katila, T., Anderson, M., 2006. Transplantation of normal boar
testicular cells resulted in complete focal spermatogenesis in a
boar by the immotile short-tail sperm defect. Reprod Domest Anim
41, 124–128.
Nagano, M., Brinster, R., 1998. Spermatogonial transplantation and
reconstitution of donor cell spermatogenesis in recipient mice.
APMIS 106, 47–55.
Nagano, M., McCarrey, J., Brinster, R., 2001. Primate
spermatogonial stem cells colonise mouse testes. Biol. Reprod. 64,
1409–1416.
Ning, L., Meng, J., Goossens, E., Lahoutte, T., Marichal, M., Tournaye,
H., 2012. In search of an efficient injection technique for future
clinical application of spermatogonial stem cell transplantation:
infusion of contrast dyes in isolated cadaveric human testes. Fertil.
Steril. 98, 1443–1448.
Ogawa, T., Aréchaga, J., Avarbock, M., Brinster, R., 1997.
Transplantation of testis germinal cells into mouse seminiferous
tubules. Int. J. Dev. Biol. 41, 111–122.
Ogawa, T., Dobrinski, I., Brinster, R., 1999. Recipient preparation is
critical for spermatogonial transplantation in the rat. Tissue Cell 31,
461–472.
Picton, H., Wyns, C., Anderson, R., Goossens, E., Jahnukainen, K.,
Kliesch, S., Mitchell, R., Pennings, G., Rives, N., Tournaye, H., van
Pelt, A., Eichenlaub-Ritter, U., Schlatt, S., 2015. A European
perspective on testicular tissue cryopreservation for fertility
preservation in prepubertal and adolescent boys. Hum. Reprod. 30,
2463–2475.
Radford, J., Shalet, S., Lieberman, B., 1999. Fertility after treatment for
cancer. Questions remain over ways of preserving ovarian and
testicular tissue. BMJ 319, 935–936.
Sadri-Ardekani, H., Atala, A., 2014. Testicular tissue cryopreservation
and spermatogonial stem cell transplantation to restore fertility:
from bench to bedside. Stem Cell Res. Ther. 5, 68.
Schlatt, S., Rosiepen, G., Weinbauer, G., Rolf, C., Brook, P., Nieschlag,
E., 1999. Germ cell transfer into rat, bovine, monkey and human
testes. Hum. Reprod. 14, 144–150.
Struijk, R., Mulder, C., van der Veen, F., van Pelt, A., Repping, S., 2013.
Restoring fertility in sterile childhood cancer survivors by
autotransplantating spermatogonial stem cells: are we there yet?
Biomed Res. Int. 2013, 903142.
Van Saen, D., Goossens, E., De Block, G., Tournaye, H., 2009.
Regeneration of spermatogenesis by grafting testicular tissue or
injecting testicular cells into the testis of sterile mice: a comparative
study. Fertil. Steril. 91, 2264–2272.
Wu, X., Goodyear, S., Abramowitz, M., Tobias, J., Avarbock, M., Brinster,
R., 2012. Fertile offspring derived from mouse spermatogonial stem
cells cryopreserved for more than 14 years. Hum. Reprod. 27, 1249–
1259.