THE BLOOD-TESTIS BARRIER MIGHT BE DEVELOPED IN TESTICULAR ORGAN CULTURES DERIVED FROM TWO DIFFERENT MOUSE STRAINS

THE BLOOD-TESTIS BARRIER MIGHT BE DEVELOPED IN

TESTICULAR ORGAN CULTURES DERIVED FROM TWO

DIFFERENT MOUSE STRAINS

L.J. Zijlstra

Supervisors: prof. dr. A.M.M. van Pelt and dr. C.L. Mulder

Reproductive biology laboratory

22 OKTOBER 2020

1 Summary

Spermatogenesis is an intricate process in which spermatogonia differentiate into mature sperm. The process is dependent on multiple factors, including the presence of a functional blood-testis barrier (BTB). Prepubertal boys who are at risk irreversible fertility loss, for instance due to

gonadotoxic treatment, currently do not have options on fertility preservation. While postpubertal men can cryopreserve their sperm cells, prepubertal boys cannot since they do not possess sperm cells yet. Prepubertal boys do possess spermatogonia, the cells that have the capacity to

differentiate into mature sperm cells. In vitro development of sperm cells from spermatogonia in testicular organ culture is a promising approach for fertility preservation in postpubertal boys who are at risk of losing their fertility before puberty. Full spermatogenesis is not yet achieved in human tissue using this procedure. For mouse, a successful method to grow mature sperm from spermatogonia in vitro was recently described by making use of gas-liquid interphase organotypic culture. Interestingly, the success of these cultures in mouse appears to be dependent on the genetic background of the donor mice. More specifically, while full spermatogenesis could be achieved in vitro using testicular tissue from neonatal C57BL6/J mice, spermatogenesis could not be realized in tissue derived from neonatal B6D2F2 mice even though they share 75% of the same genetic background. The reason why full spermatogenesis is not achieved in B6D2F2 mice and in human is not yet clear. If it is investigated why full spermatogenesis is not achieved in B6D2F2 mice, this will probably help investigate why the procedure fails in human. The production of the BTB in testicular organ culture derived from B6D2F2 and C57B6/J mice and from human is not yet clear. Therefore in this study, it is aimed to investigate the development of the BTB in testicular organ cultures derived from C57BL6/J and B6D2F2 mice. To do so, cultured testicular fragments derived from B6D2F2 and C57B6/J mice were immunohistologically stained with haematoxylin and antibodies against zona occludens one (ZO-1). Antibodies against ZO-1 indicate the BTB since the protein is one of the constituents of the BTB. It was found that ZO-1 was expressed after 1 week of culturing until at least 5 weeks of culturing in both strains, suggesting that the BTB is developed equally at testicular organ cultures derived from both strains. Therefore, there is no evidence for the involvement of aberrant development of the BTB in the failure to achieve full spermatogenesis in testicular organ culture derived from B6D2F2 mice. However, the presence of one of the

constituents of the BTB is no guarantee for the presence of a functional BTB. Further research is needed to determine the role of the development of the BTB in the failure to achieve full spermatogenesis in testicular organ culture derived from some mice species and from human.

Introduction

Spermatogenesis is an intricate process in which spermatogonia differentiate into mature sperm. A proportion of spermatogonia termed spermatogonial stem cells (SSCs) have stem cell-capacity and thereby are crucial for spermatogenesis. SSCs may divide to maintain the SSC pool, or divide and differentiate to eventually form more mature types of germ cells, including differentiating spermatogonia, spermatocytes, haploid round spermatids and elongated spermatids. The process of spermatogenesis is supported by somatic cell populations, including adjacent Sertoli cells, and Leydig cells that produce testosterone and reside in the interstitium. Sertoli cells are involved in the development of the blood-testis barrier (BTB), which is crucial for initiation of spermatogenesis during puberty (Chen et al., 2012). The BTB is a structure that is constituted of multiple junction types, including tight junctions, that reside between Sertoli cells that reside near the basement membrane. Its function is magnified by action filaments that reside on the Sertoli-Sertoli cell interphase at the apical site of the seminiferous epithelium.

2 Spermatogonia are vulnerable to depletion as a result of gonadotoxic therapies (Whitehead et al. (1982), Relander et al. (2000) and Jahnukeinen et al. (2011)). Also, hereditary conditions such as Fanconi syndrome and Klinefelter syndrome are linked to prepubertal spermatogonial loss (Goossens

et al., 2020). Moreover, boys who start to initiate gender transition before the start of puberty will

lose fertility before the completion of the first round of spermatogenesis as well. As a result of spermatogonial depletion, men become infertile. To fulfill the wish to have a child later in life, postpubertal males at risk of fertility loss can cryopreserve their sperm cells before starting treatment. However, prepubertal boys do not produce sperm cells yet. In these prepubertal males, immature testicular tissue, that harbors SSCs, can be harvested via testicular biopsy. Multiple strategies to restore fertility using these biopsies are being investigated, including spermatogonial stem cell transplantation (SSCT), testicular grafting and the development of sperm in vitro. None of these procedures are clinically available today because they are not successful in human yet. Still, the number of institutes offering cryopreservation to prepubertal males at risk of spermatogonial depletion is rising (Goossens et al., 2020). Therefore there is a strong need for the development of a procedure that is able to fulfill the wish to have a child using these cryopreserved biopsies. If it is investigated why the procedures called above are not successful in human, this will have great impact on fertility preservation of male patients suffering from prepubertal fertility loss.

In vitro development of sperm cells from spermatogonia in testicular organ culture is a promising

approach for fertility preservation in various patient groups. This is especially the case for male childhood cancer survivors whose tissue or isolated cells cannot be placed back in the body after gonadotoxic treatment because of the risk of cancer cell contamination in the tissue. The latter could result in return of the disease if the tissue would be placed back in the body. For mouse, a successful method to grow mature sperm from spermatogonia in vitro was recently described by making use of gas-liquid interphase organotypic culture (Sato et al., 2011). The gas-liquid interphase method is a culture method which enables optimal balance between access to nutrients and oxygen at the same time. The latter is achieved by placing the testicular biopsy on a thin layer of agarose before culture medium is added to the petri dish in such an amount that de thin layer of agarose is just in contact with the medium. For human, full spermatogenesis has not been achieved using this method due to unknown reasons (Portela et al., 2019, B).

Interestingly, the success of these cultures in mouse appeared to be dependent on the genetic background of the donor mice (Portela et al., 2019, A). More specifically, while full spermatogenesis could be achieved in vitro using testicular tissue from neonatal C57BL6/J mice, spermatogenesis could not be realized in tissue derived from neonatal B6D2F2 mice even though they share 75% of the same genetic background. B6D2F2 mice are a second generation hybrid between C57B6/J and DBA2 mice. The reason for the difference in success between the two strains may lay in differences in the production of the blood-testis barrier (BTB). If it is investigated why full spermatogenesis is not achieved in B6D2F2 mice, this will help to investigate the factors that will also influence in vitro spermatogenesis in human.

In vivo, the BTB is formed during puberty, which is after two weeks postpartum in mouse (Hu et al.,

2015). An intact BTB is crucial for spermatogenesis to proceed correctly. Firstly because the BTB functions as a selective barrier between the apical and basal site of the seminiferous epithelium with its tight junctions, thereby dividing the seminiferous epithelium in an apical and basal compartment . It also provides an immunological barrier, protecting the “foreign” haploid gametes from immunological destruction and causing inflow of selective nutrients only. Secondly, the blood testis barrier is of importance for Sertoli cell polarity within the seminiferous tubules to support the germ cells in their maturation in direction from basal membrane (least matured; spermatogonia) to lumen

3 (most matured; elongated spermatids) (Chen et al., 2012). Correct differentiation of supporting somatic cells is crucial either way.

The presence of the androgen receptor is a marker for Sertoli cell maturation. Earlier research showed that the expression of the androgen receptor (AR) is normal in both strains, implicating that the Sertoli cells actually do mature in both strains using the testicular organ culture method (Portela et al., 2019, A).

de Michele et al. (2018) stated that some of the main proteins of the BTB (claudin-11 and connexin 43) were expressed in testicular organ culture derived from human. The latter suggests the presence of the BTB in the culture, although functionality of the BTB was not tested. In contrast, Rondanino et al. (2017) found major differences in expression of some of the proteins that constitute the BTB and differences in BTB permeability in vitro compared to the situation in vivo in a mouse strain wherein only low yield spermatogenesis is achieved (CD-1). The latter study implicates that incorrect development of the BTB in testicular organ culture could be the reason why full spermatogenesis is not yet achieved in some mouse strains and therefore suggests a role of incorrect development of the BTB in human derived testicular organ culture as well. It is currently unclear whether the BTB is developed completely in both strains mentioned above and in human. Therefore, further research is necessary to evaluate whether and how the BTB is developed in testicular organ culture derived from human. Thus, aberrant development of the BTB in B6D2F2 fragments may underlie the failure to achieve full spermatogenesis in vitro.

Therefore in this study, it is aimed to investigate the development of the BTB in testicular organ culture derived from B6D2F2 and C57B6 mice. Considering the findings of Portela et al. (2019, A), it is hypothesized that it is likely that the BTB is not, or aberrantly developed in B6DF2F fragments compared to C57BL6/J fragments. Aberrantly developed meaning that either the expression pattern of the constituents or the timing of the development of the BTB are different. To answer this question, it is aimed to evaluate the development of the BTB in testicular organ cultures derived from C57BL6/J and B6D2F2 mice. This will be investigated in neonatal testes derived from C75BL6/J and B6DF2F mice which were cultured for variable times. The development of the BTB will be examined in these cultured fragments using immunohistochemical staining for markers involved in the development of the BTB, namely claudin-11 (CLDN11) and zona occludens 1 (ZO-1). CLDN11 is one of the main constituents of the tight junctions that form the BTB. ZO-1 is one of the adaptor proteins that attach integral membrane proteins of the BTB to the actin cytoskeleton of the Sertoli cells (Jiang et al., 2014). ZO-1 and CLDN11 are expected to be stained at the cultured testicular sections derived from C57B6/J mice, and are not or aberrantly visible at the sections derived from B6D2F2 mice.

Materials and methods

Mice

All paraffin embedded mouse material from this project was available prior to this specific study. For antibody optimization, noncultured fragments of B6D2F2-eGFP mice (n=1) of 16 weeks old that were fixed in diluted Bouins fluid, 4% paraformaldehyde or modified Methacarn were used. These mice were adult and have full spermatogenesis, meaning that the yielded images could serve as positive controls for BTB components. For the identification of the BTB in vivo, noncultured testicular sections of B6D2F2 mice (n=1) of respectively 8 days, 2 weeks and 4 weeks postpartum that were fixed in modified Methacarn were used. For the experiments that aimed to investigate differences in the development of the BTB in testicular organ culture between B6D2F2 (n=1) and C57B6/J mice (n=1), mice that were

4 sacrificed at 8 days postpartum before their testicular fragments were either uncultured or cultured 1 week, 2 weeks, 3 weeks, 4 weeks and 5 weeks respectively were used. The fragments of the latter mice were fixed in modified Methacarn.

Testicular organ culture

Testicular organ cultures were performed prior to the initiation of this research. It was performed according to the gas-liquid interphase method described by Portela et al. (2019, B, figure 1). In brief, the tunica albuginea was removed from the dissected testes. Then, the testes were cut into 1-2 mm³ sections. One fragment was fixed right away in order to serve as baseline control. The remaining fragments were cultured on top of an agarose stand (0.35 % agarose (w/v)) containing equal volumes of 0.7% (w/v) agarose, 1% Penicillin-Streptomycin and Minimal Essential Medium alpha (α-MEM, Gibco, Thermo Fisher Scientific, USA) supplemented with 10% KnockOut serum. The α-MEM culture medium was added to the agarose stands to such an extent that the medium just reached the top of the stand. Preprocessed testicular fragments were placed on top of the agarose stand, maximum three fragments per stand, before incubation at 34 °C and 5% CO₂. The incubation time varied between 1 and 6 weeks. After incubation the fragments were prepared for immunohistochemical examination. The fragments were fixed in modified Methacarn.

Figure 1: The process of culturing testicular tissue using the gas-liquid interphase method (Wikimedia.org, adapted and consulted at 10-6-2020). First, the testis is taken from the mouse. Then, the testis is cut into smaller fragments. Thereafter,

the testicular fragments are placed on an agarose grid, which makes it possible to incubate the fragments with the culture medium without soaking the tissue in the medium because the fluid is just touching the top of the grid. The latter allows good balance between oxygen and nutrient access.

5

Antibody optimization

First, the immunohistochemistry protocol was optimized for rabbit polyclonal CLDN11 (ab53041, Abcam, UK) and rabbit polyclonal ZO-1 (61-7300, Invitrogen, US) primary antibodies. In order to do so, earlier described methods of researchers who have been using these antibodies in another setting before (Alves-Lopes et al., 2017 & Kato et al., 2020), and the locally used DAB immunohistochemistry protocol (H008.1, see appendix 1) were compared and integrated. Different fixatives (4% paraformaldehyde, modified Methacarn and diluted Bouin’s solution) and primary antibody concentrations were tested. The added value of antigen retrieval was examined. The optimized protocol is described below under “immunohistochemistry”.

Immunohistochemistry

Preprocessed testicular organ culture paraffin blocks, fixed with modified Methacarn, were cut into 5-µm-thick sections. Thereafter, the sections were deparaffinized in Xylene twice (each for 3 to 5 minutes) and were rehydrated using a descending ethanol series (2 x 100%, 96%, 70%, each for 2 minutes). Antigen retrieval was carried out at half of all slides in 0.01M sodium citrate solution, pH 6. Therefore, the sections were boiled in the microwave (for 10 minutes at maximum watt) before cooling down for three minutes and incubation at 95-100 °C for 10 minutes, in the microwave as well. Endogenous peroxidase activity was blocked by incubation with 0.3% H₂O₂/PBS for 10 to 15 minutes at room temperature. To block non-specific staining, the sections were incubated with superblock (Klinipath, Olen, Belgium) for 1 hour at room temperature in a humid slide box before incubation with rabbit polyclonal CLDN11 (ab53041, Abcam, UK) and rabbit polyclonal ZO-1 (61-7300, Invitrogen, US) primary antibody overnight at 4 °C (for antibody dilutions per experiment see Table 1). Sections that were incubated with normal rabbit serum instead of primary antibodies, at the same dilution as the primary antibodies, served as negative control. Subsequently, the sections were incubated with the Powervision Poly-Hrp-anti Ms/Rb secondary antibody for 1 hour at room temperature in a humid slide box. The sections were stained with the bright DAB substrate kit (Immunologic, the Netherlands) and haematoxylin before dehydration with an ascending ethanol series (70%, 96%, 2x 100%, each for 2 minutes) and incubation with Xylene twice (each for 2 minutes). At last, the sections were embedded in Entellan for permanent mounting of the tissue on the slides. To evaluate the efficacy of the protocol, and to determine which fixative suits best, the quality of the tissue staining was assessed using a bright-field microscope (Olympus DP20) at 40 times magnification.

Table 1: Antibody dilutions per experiment. Every experiment was performed with different primary antibody dilutions

because antibody optimization was performed at the same time.

experiment ZO-1 concentration CLDN11 concentration

Effect of AR, ZO-1 and CLDN11 in adult uncultured

testes 1:400 1:500

Development of the BTB in vivo 1:1000 -

ZO-1 in testicular organ cultures 1:2000 -

6 Results

ZO-1 represented the BTB more clearly with immunohistological staining than CLDN11 in adult mouse testis tissue

To properly detect the BTB, a distinctive immunological staining for a BTB component was required. Antibodies directed against CLDN11 and ZO-1 on adult mouse tissue (B6D2F2-eGFP) were tested, since the BTB should be intact once mice have gone through puberty. Both antibodies stained their epitope correctly (figure 2, appendix 2). For CLDN11, antigen retrieval was needed in order to properly stain the BTB. For this specific antibody, fixatives diluted Bouins fluid and modified

Methacarn gave the most clear results. CLDN11 only stained small spots which made the distinction between background staining actual CLDN11 signal difficult.

ZO-1 solely defined the BTB clearly when fixative modified Methacarn was used. With antigen retrieval, clear and distinctive staining of the BTB was achieved with few background staining. It encircles the seminiferous tubules in a stitch-like manner. However, without antigen retrieval, the BTB was defined clearly even though a lot of background staining was present (appendix 3). Also without antigen retrieval, clear staining of the peritubular cells disturbed the contrast of the BTB staining. ZO-1 defined the BTB more clearly than CLDN11.

Figure 2: Immunohistological identification of the blood-testis barrier with primary antibodies CLDN11 (claudin-11) and zona occludens-1 (ZO-1) in adult uncultured B6D2F2-eGFR mice fragments fixated in modified Methacarn and treated with antigen retrieval. A. CLDN11 primary antibody, fixated with modified Methacarn, with antigen retrieval. B. ZO-1

primary antibody, fixated with modified Methacarn, with antigen retrieval. White arrowheads: blood-testis barrier. With staining using the antibodies against CLDN11, only small brown spots were visible. With staining using the antibodies against ZO-1, clear brown structures that encircle the seminiferous tubules were visible. Therefore, the BTB was easier to detect using antibodies against ZO-1 rather than using antibodies against CLDN11. Negative serum controls can be found in appendix 2 .

In vivo, ZO-1 was present after 2 weeks until at least 4 weeks in B6D2F2-eGFP mice

To properly compare the development of the BTB in vitro and in vivo, the process should be

evaluated in vivo using the same tools to qualify the development first. Therefore, the development of the BTB in vivo was evaluated using immunohistochemistry with antibodies against ZO-1. Serum negative controls were included. ZO-1 was expressed from 2 weeks onwards (figure 3). At the serum negative controls, ZO-1 was not stained.

7

Figure 3: Development of the blood-testis barrier (BTB) in B6D2F2-eGFP mice in vivo, visualized using

immunohistochemistry with primary antibodies against zona occludens 1 (ZO-1). Pictures were taken with the bright field

microscope at 40x magnification. The depicted bar represents 20 µm. The BTB is represented by the brown line that encircles the seminiferous tubules. The BTB was stained at the sections that were treated with primary antibodies against ZO-1, while the structure was not stained at the negative serum controls. The BTB was present after 2 weeks postpartum.

8

ZO-1 is expressed at the natural location in testicular organ culture derived from B6D2F2 and C57B6/J mice after 1 week until at least 5 weeks of culturing

To determine the presence of the BTB in testicular organ culture derived from neonatal B6D2F2 and C57B6/J mice, testicular fragments that were cultured for different periods in time (uncultured, 1 week, 2 weeks, 3 weeks, 4 weeks, 5 weeks) were immunohistologically stained with haematoxylin and antibodies against ZO-1. Negative serum controls were included and uncultured testicular sections of postpubertal mice (4 weeks postpartum) served as positive control. ZO-1 was stained at the positive control and was not at the negative controls. The cultured fractions of both strains expressed ZO-1 from 1 week until 5 weeks at the natural location (figure 4). Negative serum controls can be found in appendix 4.

9

Figure 4: immunohistological examination of the presence of zona occludens 1 (ZO-1) in testicular organ cultures derived from B6D2F2 and C57B6/J mice until 5 weeks of culturing. The depicted bar represents 20 µm.

ZO-1 is marked by the brown structures that encircle the seminiferous tubules. ZO-1 was stained at the positive control. At the cultured sections, ZO-1 was present after one week of culturing until 5 weeks of culturing. Before culturing, the protein was not present. ZO-1 was not stained at the serum negative controls.

10 Discussion

This study showed that ZO-1 is expressed at the natural location in testicular organ cultures derived from B6D2F2 and C57B6/J mice. Also, the timing of the development of the protein was comparable. CLDN11 was not suitable for the detection of the BTB. These outcomes suggest the presence of the BTB in both strains. These outcomes do not confirm the hypothesis that the BTB is not or aberrantly developed in B6D2F2 mice while developed normally in C57B6/J mice.

Since the results of this study suggest the presence of the BTB in both strains, it is unlikely that aberrant development of the BTB causes the failure to achieve full spermatogenesis in testicular organ culture derived from B6D2F2 mice. This could be declared by the fact that spermatogenesis is an intricate process that is influenced by numerous factors. All of these factors potentially being a limiting factor on the success of testicular organ culturing. An organ culture is an unnatural habitat for human tissue, composed based on human knowledge of an intricate human. The natural process being highly complicated, it is plausible to think that researchers do not completely understand all factors of influence on the process of spermatogenesis yet. Such unknown factors of influence would always limit the success of the process of simulation of a natural process. The genetic background determines the ability to adapt to new situations.

At some areas, the BTB appears interrupted. These interruptions could be caused by the way the sections of the tissue were cut. Because the adjacent cells are not all on one line, the BTB passes adjacent cells from the front and behind. At some places, when a cross section is made, the BTB will not be visible because it lays in front or behind the cell that lays on the surface of the section. The latter makes it seem like the BTB is interrupted while in fact it lays in front or behind the cell that lays on the surface of the cut tissue.

Slight differences in staining intensity were observed between different mouse strains and different culturing periods. The extent to which the structure of interest is stained using

immunohistochemistry does not reveal so much about the extent to which the stained structure is present because of the inability to perform the immunohistological staining accurate enough to compare the extent to which the structure is stained on different slides.

The presence of one of the proteins that constitute the BTB does not necessarily proof the presence of an intact and functional BTB. In fact, as described above, with immunohistochemistry one cannot evaluate the expression level of proteins. Moreover, the presence of all the proteins that constitute the BTB would not guarantee an optimally functional BTB either. Minor differences in conformation would not be measurable with immunohistochemistry but could be of high impact on the function of the BTB. However, ZO-1 is an established marker for the BTB (Rondanino et al., 2017, Alves-Lopes et

al., 2017, Kato et al., 2020), demonstrating the relevance of its presence.

ZO-1 was clearly visible with immunohistochemistry while CLDN11 was harder to detect. The latter could be declared by the natural position of the proteins. This difference in visibility makes ZO-1 more suitable to detect the presence of the BTB than CLDN11. Nevertheless, knowing that ZO-1 is expressed, it is interesting to also verify the co-localization of CLDN11 expression or other

components of the BTB in testicular organ cultures from the two different strains.

Earlier research of Hu et al. (2015) stated that the BTB is formed around 2 weeks postpartum in vivo. This is in accordance with the results yielded in this study. Namely, mice were sacrificed at eight days postpartum before incubation at the petri dish. After one week of culturing, the BTB was observed for the first time. After one week of culturing, the mouse would have been 2 weeks old.

11 Earlier research of Rondanino et al. (2018) implicated that there might be a link between BTB

integrity and spermatogenesis efficacy in vitro in CD-1 mice. In these mice, full spermatogenesis was achieved but the spermatogenic yield was low. They found that some of the main proteins of the BTB, including ZO-1 and CLDN11, were expressed normally after 4 weeks of culturing testicular sections of the CD-1 mouse. In contrast, they found that expression of one of the BTB constituents (claudin-3) was drastically decreased. At last they found that the BTB permeability was increased after 4 weeks in vitro compared to the situation in vivo. The findings called above demonstrate that the presence of one of the constituents of the BTB is no prove of the presence of an effective BTB. Thus, more research is needed to accurately evaluate the role of the BTB in the success of testicular organ culturing.

As in mouse, the process of spermatogenesis relies on a functional BTB in human as well (de Michele

et al., 2017). Since this study does not provide evidence for the role of the BTB in the failure to

achieve full spermatogenesis in mice, it does not indicate a role for this sophisticated structure in human either. However, if it is investigated why full spermatogenesis is not achieved in testicular organ culture derived from some mouse strains, this will probably help investigate why the procedure does not work in human. If a successful method to grow human mature sperm from prepubertal testicular biopsies via testicular organ culture is developed, child wish of male childhood cancer survivors can be fulfilled.

Therefore, further research that evaluates the presence of the constituents of the BTB and its expression patterns is necessary to evaluate if aberrant composition of the BTB causes the failure to achieve full spermatogenesis in testicular organ culture derived from B6D2F2 mice. Furthermore, research that evaluates the functionality of the BTB in B6D2F2 and C57B6/J mice is crucial to determine if BTB functionality plays a role in the failure to achieve full spermatogenesis in testicular organ culture derived from the same mouse strain. One of the studies should specifically focus on the role of BTB protein claudin-3, since aberrant expression of this protein is observed at the CD-1 mouse (Rondanino et al., 2017). The proteins that constitute the BTB can be localized using

immunohistochemistry, expression levels of the BTB associated genes proteins can be determined by measuring mRNA levels using quantitative RT-PCR and the permeability of the BTB can be tested using a biotin tracer (Rondanino et al., 2017).

In short, this study suggested that the BTB is formed in testicular organ culture derived from B6D2F2 and C57B6D/J mice based on ZO-1 expression. Therefore, this study provides no prove for aberrant development or constitution of the BTB to be the reason for the failure to achieve full

spermatogenesis in testicular organ culture derived from B6D2F2 mice. However, shortcomings of the instruments used in this study should be considered and further research is necessary to accurately evaluate the role of the BTB in the success of testicular organ culturing.

12 References

- Alves-Lopes JP, Söder O, Stukenborg JB. Testicular organoid generation by a novel in vitro three-layer gradient system. Biomaterials. 2017 Jun;130:76-89.

- Yan Cheng, C., & Mruk, D. D. (2012). The blood-testis barrier and its implications for male contraception. Pharmacological Reviews, 64(1), 16–64.

- Goossens, E., Jahnukainen, K., Mitchell, R., van Pelt, A., Pennings, G., Rives, N., … Stukenborg, J. (2020). Fertility preservation in boys: recent developments and new insights †. Human

Reproduction Open, 2020(3), 1–18.

- https://commons.wikimedia.org/wiki/File:Testis_Organ_Culture_Vs_Cell_Culture.jpg. Consulted and adapted at October 6th 2020.

- Hu, X., Tang, Z., Li, Y., Liu, W., Zhang, S., Wang, B., Tian, Y., Zhao, Y., Ran, H., Liu, W., Feng, G. S., Shuai, J., Wang, H., & Lu, Z. (2015). Deletion of the tyrosine phosphatase Shp2 in Sertoli cells causes infertility in mice. Scientific Reports, 5, 1–15.

- Jahnukainen, K., Heikkinen, R., Henriksson, M., Cooper, T. G., Puukko-Viertomies, L. R., & Mäkitie, O. (2011). Semen quality and fertility in adult long-term survivors of childhood acute lymphoblastic leukemia. Fertility and Sterility, 96(4), 837–842.

- Jiang XH, Bukhari I, Zheng W, Yin S, Wang Z, Cooke HJ, Shi QH. Blood-testis barrier and spermatogenesis: lessons from genetically-modified mice. Asian J Androl 2014;16:572-80. - Kato T, Mizuno K, Nishio H, Moritoki Y, Kamisawa H, Kurokawa S, Nakane A, Maruyama T,

Ando R, Hayashi Y, Yasui T. Disorganization of claudin-11 and dysfunction of the blood-testis barrier during puberty in a cryptorchid rat model. Andrology. 2020 Sep;8(5):1398-1408.

- Komeya, M., Sato, T., & Ogawa, T. (2018). In vitro spermatogenesis: A century-long research journey, still half way around. Reproductive Medicine and Biology, 17(4), 407–420.

- De Michele, F., Poels, J., Giudice, M. G., De Smedt, F., Ambroise, J., Vermeulen, M., Gruson, D., & Wyns, C. (2018). In vitro formation of the blood-testis barrier during long-term organotypic culture of human prepubertal tissue: Comparison with a large cohort of pre/peripubertal boys. Molecular Human Reproduction, 24(5), 271–282.

- Portela, J. M. D., Mulder, C. L., van Daalen, S. K. M., de Winter-Korver, C. M., Stukenborg, J. B., Repping, S., & van Pelt, A. M. M. (2019). Strains matter: Success of murine in vitro spermatogenesis is dependent on genetic background. Developmental Biology, 456(1), 25– 30.

13 - Portela, J. M. D., De Winter-Korver, C. M., Van Daalen, S. K. M., Meißner, A., De Melker, A. A.,

Repping, S., & Van Pelt, A. M. M. (2019). Assessment of fresh and cryopreserved testicular tissues from (pre)pubertal boys during organ culture as a strategy for in vitro

spermatogenesis. Human Reproduction, 34(12), 2443–2455.

- Relander, T., Cavallin-Stahl, E., Garwicz, S., Olsson, A.M., Willen, M. (2000). Gonadal and sexual function in men treated for childhood cancer. Med Pediatr Oncol;35:52–63. - Rondanino, C., Maouche, A., Dumont, L., Oblette, A., & Rives, N. (2017). Establishment,

maintenance and functional integrity of the blood-testis barrier in organotypic cultures of fresh and frozen/thawed prepubertal mouse testes. Molecular Human Reproduction, 23(5), 304–320.

- Sato, T., Katagiri, K., Gohbara, A., Inoue, K., Ogonuki, N., Ogura, A., … Ogawa, T. (2011). In vitro production of functional sperm in cultured neonatal mouse testes. Nature, 471(7339), 504–508.

- Whitehead, E., Shalet, S. M., Morris Jones, P. H., Beardwell, C. G., & Deakin, D. P. (1982). Gonadal function after combination chemotherapy for Hodgkin’s disease in childhood.

14 Appendices

Appendix 1: immunohistochemistry protocol

Optimizing an antibody with immunohistochemistry

Immunochemistry is studied from the aspect of using antibodies to label antigens/epitopes of interest in cells (immunocytochemistry) or tissues (immunohistochemistry). An antigen can consist of multiple epitopes. A polyclonal antibody binds several epitopes of an antigen and a monoclonal antibody will bind one epitope of an antigen. Antigens are usually proteins or polysaccharides. Lipids and nucleic acids are antigenic only when combined with proteins and polysaccharides.

Purpose:

Every newly arrived antibody should first be tested with immunohistochemistry, except when the antibody is belonging to a batch number that has been tested before. The antibody is optimized with immunohistochemistry and not with immunocytochemistry. This because immunohistochemistry not only tells you something about the location of your epitope of interest in the cell (nuclear or

cytoplasmic), but also provides information about the location of the positive cells in the used tissue. This will help you to identify the cell type.

Chemicals:

Day 2 Day 3

• Xylene • PBS

• EtOH • Powervision Poly-Hrp-anti Ms/Rb, anti-RT, secondary

antibody • PBS • Diaminobenzidine (DAB) • Sodiumcitrate buffer (0,01M, pH 6,0) • H2O2 (30%) • Triton X-100 • Heamatoxylin • H2O2 (30%) • EtOH • Tween-20 (20% in PBS) • Xylene • BSA (10% in PBS) • Entellan

• Blocking buffer: superblock • Neg. Controle serum/IgG

• Primary antibody (monoclonal/polyclonal)

PS: Warning! Diaminobenzidine (DAB) is a CMR chemical! Work always in a fume hood and wear gloves!

Materials:

Day 2 Day 3

• Incubation/washing jars (glass) • Incubation/washing jars (glass)

• Aluminium foil • Aluminium foil

• Plastic glass holder • Cover slides

• Measuring jar (2 l) • Slide box

• Tissues • Parafilm

15

Equipment:

Day 2 Day 3

• Shaking table • Timer

• Thermometer • Microscope

• Magnetic stirrer • Pipets and tips

• Timer

• Pipets and tips

Related protocols:

• H001: Slicing paraffin sections

• H002: Mounting sections on a microscope slide • H007: Heat induced antigen retrieval

• H003: Histological staining: Haematoxylin/Eosin

Procedure: Day 1

1. Check the provided datasheet of the antibody for: * the host in which the antibody is raised (e.g. rabbit) * target organism of the antibody (e.g. human) * Isotype of the antibody (e.g. IgG)

* Recommended working dilution of the antibody (e.g. 1:50) * predicted cross hybridization due to homology (e.g. mouse)

2. Make a plan how to optimize the new antibody. Include all the procedure variables that you need to test and think of positive and negative controls. The procedure variables are:

* Different tissue fixatives (4% PFA, Mod. Methacarn and dBouin); start to test all fixatives

by mounting

them all on one slide.

* Antigen retrieval (none, 1x, 2x or 3x); start to test “no antigen retrieval” in comparison to “1x

antigen retrieval”.

* Cell permeabilization with Triton-X (yes or no); Paraffin sections don’t need any

permeabilization (cells are

already cut open). But cells do need permeabilization if you want to detect an intracellular or intranuclear

marker. Permeabilization with Triton X-100 is irreversible. Use Triton X-100 in a range of 0.2-0.5%.

* Different antibody working dilutions

3. Prepare a sufficient amount of slides based on your plan; see protocols H001 (Slicing paraffin sections) and H002 (Mounting sections on a microscope slide). Mount on one slide next to each other PFA, Mod. Methacarn and dBouin fixated tissue. Let the slides dry overnight in a 37ºC incubator.

4. Prepare a sufficient amount of all the needed solutions for the Immuno.

Day 2

16

6. Hydrate the slides in 100% EtOH-absolute (2x), 96% EtOH (1x) and 70% EtOH (1x), each for 2 minutes. Slides can be stored in 70% EtOH in de cold room if necessary.

7. Rinse the slides shortly in demiwater. 8. Wash 3x 5 min. with PBS, slowly shaking. 9. No Antigen retrieval: Continue with step 11.

Antigen retieval with microwave: see protocol H007 (Heat induced antigen retrieval)

The process of antigen retrieval results in the exposure of the antigen to the antibody. Antigen retrieval is only necessary for tissue that has been fixated with formalin! There are also other (more severe) ways to retrieve the antigen, like DTT treatment. 10. Place the slide holder directly in PBS. Wash subsequently 2x 5 min. in PBS, slowly shaking.

H2O2 inactivates endogenous peroxidase and like this prevents false positive signals.

Spermatogonia contain endogenous peroxidase. 11. Wash 3x 5 min. with PBS, slowly shaking.

12. Blocking a-specific binding of the primary and secondary antibody:

Possible blocking buffers:

* Superblock (Commercial. Ingredients unknown; probably based on BSA!)

Remark: Anti goat or sheep antibodies have affinity with IgG’s in Bovine Serum Albumin (BSA).

This means that if you block with BSA (if not IgG-free), your secondary goat/sheep antibodies will non-specific bind to BSA and you’ll get a lot of BG. This may happen as well with Superblock (we don’t know if they used IgG-free BSA). It’s good to use always IgG-free BSA for immunoresearch. Pipet 100 µl on a slide and spread out with a pipet tip to cover all the tissue. Incubate for 1 hour in the humid slide box at RT.

13. Rinse the slides shortly and gently in PBS.

14. Incubation with first antibody and negative control

15. Pipet 100 µl on a slide and spread out with a pipet tip to cover all the tissue. Put the slides in a slide box with wet tissues and incubate overnight at 4ºC.

Day 3

1. Bring the slide box to RT before opening it (takes about half an hour).

2. Wash 3x 5 min. with PBS without shaking. Keep positive slides (with antibody) and negative slides (without antibody) in separate washing containers!!

3. Pipet 100 µl Powervision Poly-Hrp-anti Ms/Rb, anti-RT secondary antibody on a slide and spread out with a pipet tip to cover all the tissue. Incubate for 1 hour in a humid slide box at RT. Hrp = Horseradish Peroxidase. Do not use slides that were stored in PBS with Sodiumazide. Sodiumazide will react with Hrp and will cause BG.

17

Use the “DAB bright substrate kit”. This kit stains quicker and brighter and less DAB is needed. Take 1 ml buffer, add 1 drop Bright DAB, mix well. Cover the tissue with 100 µl solution. Keep track of the staining, sometimes less than 1 minute is sufficient!.

Check a positive slide with the “DAB microscope” (keep the slide wet!!). Place the slides back in the DAB solution whenever necessary. Place the slides in demiwater when the DAB staining is sufficient.

4. Stain the slides with Heamatoxylin (staining of the nuclei). Possible DAB staining in the nuclei should be visible. See protocol H003.

Immuno slides are normally only stained with Heamatoxylin and not with Eosin.

5. Dehydrate the slides in 70% EtOH (1x), 96% EtOH (1x), 100% EtOH (2x), Xylene I (1x) and Xylene II (1x), each for 2 minutes.

6. Embed the slides in Entellan.

18 Appendix 3: The effect of antigen retrieval on background staining in immunohistochemistry using modified Methacarn as a fixative and primary antibodies against zona occludens-1. A. Positive control with antigen retrieval. B. Negative control with antigen retrieval. C. Positive control without antigen retrieval. C. Negative control without antigen retrieval. Note that at the sections without antigen retrieval, the staining of the peritubular cells (black arrows) disturbs the contrast of the staining of the blood-testis barrier (white arrows). Also, the sections without antigen retrieval show more background staining.

19 Appendix 4: Serum negative controls of testicular organ cultures derived from B6D2F2 and C57B6/J mice yielded before culturing (zero days) and after 1 week, 2 weeks, 3 weeks, 4 weeks and 5 weeks of culturing. The C57B6/J negative controls after 1 and 2 weeks are missing because there was no tissue left to use as negative control.