Formin 1 Regulates Microtubule and F-Actin Organization to Support Spermatid Transport During Spermatogenesis in the Rat Testis

Abstract

Formin 1 confers actin nucleation by generating long stretches of actin microfilaments to support cell movement, cell shape, and intracellular protein trafficking. Formin 1 is likely involved in microtubule (MT) dynamics due to the presence of a MT binding domain near its N terminus. Here, formin 1 was shown to structurally interact with α-tubulin, the building block of MT, and also end-binding protein 1 (a MT plus [+]-end-binding protein that stabilizes MT) in the testis. Knockdown of formin 1 in Sertoli cells with an established tight junction barrier was found to induce down-regulation of detyrosinated MT (a stabilized form of MT), and disorganization of MTs, in which MTs were retracted from the cell cortical zone, mediated through a loss of MT polymerization and down-regulation of Akt1/2 signaling kinase. An efficient knockdown of formin 1 in the testis reduced the number of track-like structures conferred by MTs and F-actin considerably, causing defects in spermatid and phagosome transport across the seminiferous epithelium. In summary, formin1 maintains MT and F-actin track-like structures to support spermatid and phagosome transport across the seminiferous epithelium during spermatogenesis.

In mammals, spermatogenesis takes place in the seminiferous tubules in the testis that produces approximately 200–300 million vs approximately 30–50 million sperm daily from each male since approximately 12 years of age vs approximately 45 days of age in humans and rats, respectively, well into adulthood (1, 2). In order to sustain such enormous cellular output, preleptotene spermatocytes differentiated from type B spermatogonia must be rapidly transported across the blood-testis barrier (BTB) between adjacent Sertoli cells near the basement membrane of the tunica propria, so that primary spermatocytes can prepare for meiosis I/II at the adluminal (apical) compartment behind the BTB. Furthermore, postmeiotic spermatids derive from meiosis must also be transported across the entire adluminal compartment so that fully developed spermatids (ie, spermatozoa) can line-up near the luminal edge to prepare for their release at spermiation (for reviews, see Refs. 3, 4). It is envisioned that extensive remodeling takes place at the Sertoli cell-cell and Sertoli-spermatid interface to accommodate these events of germ cell transport. Without the timely transport of preleptotene spermatocytes across the BTB and of spermatids across the adluminal compartment, spermatogenesis will be arrested, leading to infertility. Studies have shown that in order to support preleptotene spermatocyte and spermatid transport, cytoskeletons at the Sertoli cell-cell and Sertoli-spermatid interface conferred by basal and apical ectoplasmic specialization (ES) are rapidly reorganized. It is known that the ES is a testis-specific actin-rich ultrastructure typified by the presence of actin microfilaments that are bundled and sandwiched in between cisternae of endoplasmic reticulum and the Sertoli cell plasma membrane (for reviews, see Refs. 3, 5–8). Interestingly, studies in other epithelia have shown that besides the actin-based cytoskeleton, the microtubule (MT)-based cytoskeleton conferred by dimerized α- and β-tubulin is also critical in cellular and organelle transport by serving as the tracks, involving the MT-specific motor proteins (for reviews, see Refs. 9–12). Indeed, an extensive MT network is found near the actin microfilament bundles at the ES in the testis (for review, see Ref. 7). In short, germ cell transport is supported by both actin- and MT-based cytoskeletons. However, it remains virtually unknown the mechanism(s) nor the involving biomolecule(s) that coordinate the concerted efforts of actin- and MT-based cytoskeletons to support germ cell transport.

Formin 1 is a member of the formin family protein known to be involved in actin nucleation at the barbed end of an actin microfilament, rapidly creating long stretches of actin microfilament in mammalian cells to support cell motility and endocytic vesicle-mediated trafficking, and to confer scaffolding function (for reviews, see Refs. 13–16), including Sertoli cells in the testis as recently reported (17). It is of interest to note that along the polypeptide sequence of formins, such as formin 1 which contains a MT binding domain near its N terminus, formins contain conserved formin homolog domains near their C terminus to be used to nucleate actin microfilaments but also to mediate interactions with numerous proteins that regulate cytoskeletal function (for reviews, see Refs. 18–20). For instance, studies have shown that formins including formin 1 and inverted formin 1 (also known as formin homolog 2 domain containing 1, which is a MT-associated formin) are associated with MT-organizing center (21, 22), the ultrastructure in eukaryotic cells from which MTs emerge, illustrating formins are involved in MT function. Indeed, emerging evidence has shown that formins regulate MT dynamics in mammalian cells by stabilizing MT structures (for reviews, see Refs. 19, 20) via mechanism(s) remain to be fully understood. Here, we report the role of formin 1 in conferring the actin- and MT-based tracks to support spermatid and phagosome transport using a loss-of-function approach by RNAi, supporting the concept that formin 1 provides the necessary cross talk so that the 2 cytoskeletons are working in concert to confer spermatid and phagosome transport during the epithelial cycle of spermatogenesis.

Materials and Methods

Animals and antibodies

Sprague Dawley rats, both male pups at 18 days of age to be used on 20 days of age for the isolation of Sertoli cells for primary cultures and adult rats at 275- to 300-g body weight (∼60–70 d of age), were obtained from Charles River Laboratories. The use of animals for different experiments reported here was approved by The Rockefeller University Institutional Animal Care and Use Committee with protocol numbers 12-506 and 15-708H. Each 10-male pups were shipped with a foster mother, and rats were housed at The Rockefeller University Comparative Bioscience Center and with free access to standard rat chow and water and a 12-hour light, 12-hour dark cycle at 20 ± 1°C. Rats were euthanized by CO₂ asphyxiation using slow (20%–30%/min) displacement of chamber air with compressed CO₂ using a built-in regulator approved by The Rockefeller University Lab Safety and Comparative Bioscience Center. Antibodies were obtained commercially from different vendors and listed in Table 1.

Table 1.

Antibody Table

Antibodies	Host Species	Vendor	Catalog Number	Dilution
				IB	IF/IHC
Actin	Goat	Santa Cruz Biotechnology	sc-1616	1:200
Akt1	Rabbit	Cell Signaling Technology	2938S	1:500
Akt2	Rabbit	Cell Signaling Technology	2964S	1:500
α-Tubulin	Mouse	Abcam	ab7291	1:1000	1:500
β-Tubulin	Rabbit	Abcam	ab6046	1:1000	1:500
Detyrosinated-α-tubulin	Rabbit	Abcam	ab48389	1:500	1:100
EB1	Mouse	BD Biosciences	610534		1:100 (cells)
EB1	Rabbit	Santa Cruz Biotechnology	sc-15347	1:500	1:100 (testes)
Formin 1	Mouse	Abcam	ab68058	1:500	1:100
Katanin p-80	Rabbit	Santa Cruz Biotechnology	sc-292216	1:200
MARK4	Rabbit	Cell Signaling Technology	4834S	1:500
mDia1	Goat	Santa Cruz Biotechnology	sc-10885	1:500
p-Akt1-Thr308	Rabbit	Cell Signaling Technology	2965S	1:500
p-Akt1-Ser473	Rabbit	Cell Signaling Technology	4060S	1:500
p-Akt2-Ser474	Rabbit	Cell Signaling Technology	8599S	1:500
Vimentin	Mouse	Santa Cruz Biotechnology	sc-6260	1:1000
Mouse IgG-Alexa Fluor 488	Goat	Invitrogen	A11029		1:250
Rabbit IgG-Alexa Fluor 488	Goat	Invitrogen	A11034		1:250
Rabbit IgG-Alexa Fluor 555	Goat	Invitrogen	A21429		1:250
Mouse IgG-HRP	Bovine	Santa Cruz Biotechnology	sc-2371		1:250
Rabbit IgG-HRP	Bovine	Santa Cruz Biotechnology	sc-2370		1:250

Abbreviations: Akt, transforming oncogene of v-akt encoding Ser/Thr protein kinase Akt1 or Akt2, also known a protein kinase B (PKB); EB1, end-binding protein 1, a +TIP; HRP, horseradish peroxidase; MARK4, microtubule-associated protein (MAP)/microtubule affinity-regulating kinase 4; mDia1, mammalian Diaphanous-related formin.

Primary Sertoli cell cultures

Sertoli cells were isolated from testes of 20-day-old male pups and cultured in serum-free F12/DMEM supplemented with growth factors and gentamicin at 35°C with 95% air/5% CO₂ (vol/vol) in a humidified atmosphere as described (23). Sertoli cells were seeded at different cell density on Matrigel (1:7 diluted in F12/DMEM; BD Biosciences) coated: 1) 6-well dishes at 0.4 × 10⁶ cells/cm², with each well containing 5-mL F12/DMEM supplemented with growth factors (for immunoblotting [IB] and MT polymerization assay); and 2) round cover glasses (18-mm diameter) placed onto 12-well dishes at 0.05 × 10⁶ cells/cm² (for visualization of MT and MT binding/regulatory protein end-binding protein 1 [EB1], a plus [+]-end-binding protein, +TIP), and each well contained 2-mL F12/DMEM. These cell densities were selected based on pilot experiments in order to obtain enough proteins in cell lysates for immunoblotting or to better visualize MT, F-actin, and EB1 in Sertoli cells. Cover glasses were then gently removed from the 12-well dishes and processed for immunofluorescence (IF) microscopy at specified day. The day when Sertoli cells were seeded was designated as cultures on day 0. About 48 hours later (ie, d 2), Sertoli cells were subjected to hypotonic treatment as described (24) to lyse residual germ cells. The purity of these Sertoli cell cultures was greater than 98% with negligible contaminations of Leydig, peritubular myoid, and germ cells when specific markers of these cells were used for RT-PCR based on corresponding specific primer pairs or assessed by IB using corresponding specific antibodies as described (25). Thereafter, cells rinsed twice in F12/DMEM were cultured in fresh medium for an additional 24 hours before used (ie, on d 3) for different experiments.

Knockdown (KD) of formin 1 by RNAi in Sertoli cells cultured in vitro

Formin 1 KD in Sertoli cells cultured in vitro was performed as earlier described (17). In brief, Sertoli cells cultured alone on day 3 with an established functional tight junction (TJ)-permeability barrier were subjected to RNAi by transfecting cells with formin 1-specific siRNA duplexes (100nM) vs nontargeting negative control siRNA duplexes also at 100nM (Table 2) for 24 hours using Lipofectamine RNAi MAX transfection reagent (Thermo Fisher Scientific) (17). Thereafter, cells were washed twice with F12/DMEM and cultured for additional 24 hours, which were then terminated on day 5 to obtained cell lysate for IB or MT polymerization assay or fixed and processed for IF microscopy. For cell staining, cells were cotransfected with 1nM siGLO red transfection indicator (Dharmacon) to track successful transfection.

Table 2.

Primers Used for qPCR and RNAi Experiments

Gene	GenBank Accession Number	Nucleotide Position	Primer Sequence (5′-3′)	Expected Size (bp)
Formin 1	XM_006224644.2	Sense, 2828-2847	CCTCACATTATGGACACCAG	225
		Antisense, 3035-3052	CACAGAGTCATCCACGTT
GAPDH	NM_017008.4	Sense, 267-286	GCTGGTCATCAACGGGAAAC	112
		Antisense, 360-378	GGTGAAGACGCCAGTAGAC
siRNA Catalog Number	Formin 1 siRNA	Duplex Number	siRNA Sequence (5′-3′)
10620318	Fmn1RSS355680	262071C10	CAGGUCAAAUUUGAGGACCUCAUAA
10620319	Fmn1RSS355680	262071C11	UUAUGAGGUCCUCAAAUUUGACCUG

KD of formin 1 by RNAi in the testis in vivo

For in vivo KD of formin 1 by RNAi in the testis, male adult rats (275- to 300-g body weight) were used with n = 6 rats. Each testis (∼1.6 g in weight assuming a volume of ∼1.6 mL) of a rat was transfected with formin 1-specific siRNA duplexes (100nM) vs the other testis of the same rat with nontargeting negative control siRNA duplexes (also 100nM) served as a control. In short, each testis received a transfection solution of approximately 100 μL, which contained 0.16 nmol siRNA duplexes (suspended in sterile water, ∼0.32 μL), 0.32-μL PolyPlus in vivo-jetPEI (VWR International; using an N/P ratio of 6; note: N/P ratio is a measure of the ionic balance of the siRNA complexes, referring to the number of nitrogen residues (N) of jetPEI/oligonucleotide phosphate (P), in which jetPEI concentration is expressed in nitrogen residues molarity and 1 μg of oligonucleotide contains 3 nmol of anionic phosphate), and 100-μL 5% glucose solution (wt/vol) on day 0. This transfection medium was loaded onto each testis using a 28-gauage insulin syringe (1-mL capacity), in which the needle was gently administered from the apical end of the testis towards the distal end vertically, as the needle was withdrawn from the testis, transfection medium was gently released to avoid subtle changes in intratesticular pressure. Rats were terminated on day 4, which yielded reproducible phenotypes. This termination time was selected based on pilot experiments when testes were terminated on day 3, 4, and 8. The efficacy of transfection using PolyPlus in vivo-jetPEI as a transfection reagent was more than 80% based on scoring siGLO red transfection indicator with at least 5 separate red fluorescence aggregates randomly distributed inside the seminiferous epithelium of a tubule when approximately 150 tubules were randomly selected and scored from n = 3 rat testes.

IF or dual-labeled IF microscopy

IF or dual-labeled IF analysis was performed as earlier described (17, 26) using frozen cross-sections of testes (7 μm in thickness, obtained in a cryostat at −21°C) and corresponding specific primary antibodies and secondary antibodies at corresponding dilutions (Table 1). In selected experiments, such as IF staining for formin 1 in testes, cross-sections of paraformaldehyde (PFA)-fixed testes (4% PFA in PBS for 24 h) embedded in paraffin was used instead; and the steps of dewaxing and antigen retrieval were the same as the procedures used for immunohistochemistry (IHC) (see below). Primary antibody used for incubation with frozen sections or paraffin-embedded section were performed at 4°C for 16 hours using corresponding specific antibodies and appropriate working dilutions as listed in Table 1. Negative controls were performed using normal IgG of either mouse or rabbit (depending on the source of primary antibody for different experiments) and also the corresponding secondary antibody (Table 1), or the omission of secondary antibody (but with the primary antibody). F-actin was visualized by rhodamine-conjugated phalloidin as described (27). Cell nuclei were visualized by 4′, 6-diamidino-2-phenylindole (DAPI) (Invitrogen). Fluorescence images were obtained using an Olympus BX61 fluorescence microscope with images acquired using the built-in Olympus DP-71 digital camera and Olympus MicroSuite Five software package (Version 1224). Image files were analyzed using Photoshop in Adobe Creative Suite (version 3.0) such as for image overlay to assess protein colocalization.

Cell staining by IF microscopy

Sertoli cells were cultured on Matrigel-coated coverslips at 0.05 × 10⁶ cells/cm², and subjected to RNAi as described above. On day 5, cells were terminated for MT or EB1 staining in which Sertoli cells were fixed in ice-cold methanol for 5 minutes at −20°C. For visualization of F-actin, Sertoli cells were fixed in 4% (wt/vol) PFA in PBS (10mM sodium phosphate [pH 7.4 at 22°C], containing 0.15M NaCl) for 10 minutes at room temperature. After fixation, cells were washed in PBS 3 time, 5 minutes each, and permeabilized in Triton X-100 at 0.1% (vol/vol) in PBS for 10 minutes. Cells were then washed thrice in PBS, blocked in 4% BSA (wt/vol) in PBS for 30 minutes at room temperature, to be followed by an incubation with primary antibody (Table 1) for 16 hours at 4°C. Sertoli cells were then washed 3 times in PBS and incubated with fluorescence labeled secondary antibody for 30 minutes at room temperature. After rinse thrice in PBS, cells were mounted in Prolong Gold mounting solution with DAPI (Invitrogen) and examined using the Olympus BX61 fluorescence microscope.

Immunohistochemistry

For IHC, testis removed from rats were fixed immediately in Bouin's fixative (Polysciences), subjected to dehydration using ascending concentrations of ethanol before embedded in paraffin. Paraffin sections obtained as 5-μm thickness using a microtome were deparaffinized, rehydrated, and subjected to antigen retrieval in 10mM citrate buffer (pH 6.0 at 22°C) for 10 minutes in a microwave. Sections were then permeabilized by Triton X-100 at 0.1% (vol/vol) for 10 minutes and blocked in 4% BSA (wt/vol) in PBS for 30 minutes. Thereafter, sections were incubated with primary antibody for 16 hours at 4°C, to be followed by an incubation of biotinylated secondary antibody using IgG and then streptavidin-horseradish peroxidase (Invitrogen). Color visualization was performed using 3-amino-9-ethylcarbazole (AEC) substrate kit (Invitrogen), and cell nuclei were stained by hematoxylin. Sections were examined using the Olympus BX61 microscope and images were acquired as described above.

IB, coimmunoprecipitation (Co-IP), and RT-PCR

Lysates of Sertoli cells and testes were obtained by using IP lysis buffer (10mM Tris [pH 7.4 at 22°C], containing 0.15 M NaCl, 1% NP-40 (nonyl phenoxypolyethoxylethanol 40) [vol/vol], and 10% glycerol [vol/vol]) freshly supplemented with protease and phosphatase inhibitor cock-tails (Sigma-Aldrich) at a 1:100 dilution (vol/vol) (17, 26). Approximately 20 μg of protein of Sertoli cell lysate or approximately 60 μg of protein of testes lysate were subjected to SDS-PAGE. Immunoblot analysis was performed as earlier described (28) using corresponding specific antibodies (Table 1). Chemiluminescence was performed using a kit prepared in our laboratory as described (29). Co-IP was performed using lysates (∼600-μg protein) from testes to assess protein-protein interactions between formin 1 and α-tubulin vs formin 1 and EB1 as described (17, 26). RNA extraction for reverse transcription to obtain cDNAs, and PCR using specific primer pairs (Table 2) were performed as described (17). Quantitative PCR (qPCR) using SYBR Green Real-Time PCR Master Mixes (Thermo Fisher Scientific) was performed at The Rockefeller University Genomics Resource Center (New York, NY). The authenticity of PCR products were verified by direct DNA sequencing performed at Genewiz.

MT polymerization assay

A biochemical assay used to examine changes in MT polymerization after formin 1 KD in cultured Sertoli cells in vitro was performed on day 5 as detailed elsewhere earlier (30). In this assay, paclitaxel (PTX) (also known as Taxol, a MT stabilizing agent) (31) at 20μM vs CaCl₂ (known to induce MT depolymerization) (32) was included as the corresponding positive and negative controls. In brief, this assay quantified the ability of Sertoli cells with and without formin 1 KD to polymerize MTs in cell cytosols, in which the polymerized tubulins that formed MTs in the pellet vs the free tubulin monomers in the supernatant in both KD and control groups were visualized by IB using both an α- and β-tubulin specific antibody (Table 1).

Image analysis

To assess changes in protein localization in Sertoli cells after in vitro RNAi of formin 1, at least 200 cells were randomly selected and examined in control vs experimental group with n = 3 experiments. Fluorescence intensity of a target protein (eg, formin 1) in Sertoli cells, or in the seminiferous epithelium of testes after in vitro or in vivo silencing to assess efficacy of formin 1 KD, was quantified using ImageJ 1.45 (United States National Institutes of Health; http://rsbweb.nih.gov/ij) as described (33). At least 70 randomly selected staged tubules from cross-sections of a rat testis were examined with n = 3 rats and a total of more than 200 tubules. For Sertoli cells, at least 70 randomly selected cells were photographed and scored from each experiment, which was repeated 3 times using different batches of Sertoli cells and a total of more than 200 Sertoli cells were scored. Micrographs shown here are representative findings of an experiment, but each experiment was repeated at least 3 times using different batches of Sertoli cells or different rat testes and yielded similar results.

Statistical analysis

For studies using Sertoli cell cultures, triplicate coverslips (for fluorescence microscopy) or dishes (for IB, MT polymeriization assay) were used. For in vivo experiments, n = 6 rats were used with 6 control vs 6 treated testes for formin 1 KD with half testes used for frozen sections or IB and half testes used for paraffin-embedded sections for histological analysis or IHC excluding pilot experiments which were performed to identify the phenotypes and to optimize the experimental conditions/duration with n = 2–3 rats. Also, all samples within an experiment set were performed simultaneously to avoid interexperimental variations. Each data point (or bar graph) is a mean ± SD of n = 3–5 experiments (or n = 3–6 rats). For each experiment, data in treatment groups were normalized against the corresponding control, which was arbitrarily set at 1. Statistical analysis was performed using the GB-STAT software package (version 7.0; Dynamic Microsystems). Statistical analysis was performed by two-way ANOVA followed by Dunnett's test. In selected experiments, Student's t test was used for paired comparisons.

Results

Formin 1 structurally associates with actin microfilaments and MTs in the rat testis

To assess whether formin 1 is involved in MT dynamics in the testis, we first examined whether it colocalized with tubulins (note, α- and β-tubulin are building blocks of MTs) (for reviews, see Refs. 12, 34) in Sertoli cells by IF analysis using fluorescence microscopy with a specific antiformin 1 antibody earlier characterized (Table 1) (17), because ultrastructures of the ES and extensive network of MTs are found in Sertoli cells adjacent to each other in the seminiferous epithelium. As expected, formin 1 colocalized with β-tubulin and also F-actin in Sertoli cells (Figure 1A). We next used this antiformin 1 antibody for IF in cross-sections of adult rat testes (Figure 1, B–D). Formin 1 appeared as stalk-like structures of MTs in the adluminal compartment in cross-sections of testes, but most prominently in stages VI to early VII tubules, also apical ES, and basal ES (Figure 1, B and C), colocalized with β-tubulin and EB1 (also a plus-end tracking protein [+TIP] and a known MT stabilizing protein found in the rat testis) (30) that constituted the MT network in Sertoli cells (Figure 1C). The staining of formin 1 shown in Figure 1, B and C, was specific, because when the primary antibody was substituted by normal mouse IgG to be followed by an incubation with the corresponding secondary antibody, no green fluorescence was detected in cross-sections of testes (Figure 1B, upper panel). It is of interest to note that although formin 1 was prominently expressed at the apical ES and basal ES/BTB, and also the stalk-like structures in stage VI tubules and colocalized with β-tubulin and EB1, its expression at the basal ES/BTB and the stalk-like structure was considerably diminished but remained robustly expressed at the apical ES in stage VII tubules (Figure 1C), illustrating its stage-specific expression. Structural interaction of formin 1 with MT and also with EB1 was detected based on a study using Co-IP and corresponding specific antibodies (Table 1), where formin 1 association with actin, and EB1 interaction with α-tubulin served as positive controls vs normal mouse IgG in corresponding experiments served negative controls (Figure 1D). In short, these findings support the notion that there are structural association/interaction between formin 1 and tubulins (ie, MT network) in Sertoli cells in the testis.

Figure 1. A study to assess whether formin 1 structurally interacts with MTs in Sertoli cells in the seminiferous epithelium of rat testes.

A, Dual-labeled IF analysis was performed to assess whether formin 1 colocalized with β-tubulin, the building block of MTs, because MT is a polymer of α-/β-tubulin dimers, in Sertoli cells cultured in vitro with an established TJ-permeability barrier. Indeed, formin 1 (green fluorescence) colocalized with β-tubulin (red fluorescence, ie, MTs) and also F-actin (red fluorescence) in Sertoli cell cytosol. Cell nuclei were visualized by DAPI (blue). Scale bar, 20 μm, which applies to other micrographs in the same panel. B, Localization of formin 1 in the seminiferous epithelium of adult rat testes. PFA-fixed paraffin cross-sections of adult rat testes stained with formin 1 (green fluorescence), illustrating its localization at the apical ES and basal ES/BTB and also the stalk-like structures across the epithelium in tubules of stages VI–VII (lower panel). Formin 1 (green fluorescence) was not found in control sections (upper panel) incubated with normal mouse IgG, which substituted the primary antibody to be followed by an incubation with the corresponding goat antimouse IgG-Alexa Fluor 488 secondary antibody (Table 1), illustrating staining shown in the lower panel was specific for formin 1. Scale bar, 150 μm, which applies to the other micrograph. C, Stage-specific expression of formin 1 (green fluorescence) in the seminiferous epithelium of adult rat testes using PFA-fixed paraffin cross-sections was noted in which formin 1 staining was limited to stage VI-VII tubules, colocalized with β-tubulin (red fluorescence) and EB1 (red fluorescence) in particular the MT-based stalk-like structures in stage VI tubules but considerably diminished in stage VII tubules. At stage VI, formin 1 was robustly expressed at the apical ES and basal ES/BTB, colocalized with β-tubulin and EB1. At stage VII, although formin 1 remained prominently expressed at the apical ES and colocalized with β-tubulin and EB1, its expression considerably reduced at the basal ES/BTB. Cell nuclei were visualized by DAPI (blue). Scale bar, 40 μm, which applies to other micrographs of the same experiment. It is noted these findings are representative findings of an experiments which was repeated 3 times using different rat testes and yielded similar results. D, Lysates of rat testes were used for Co-IP, illustrating structural interaction between formin 1 and α-tubulin (left panel), formin 1 and EB1 (middle panel), as well as EB1 and α-tubulin (right panel) vs normal mouse IgG serving as negative controls in corresponding experiments. Interactions between formin 1 and β-actin vs EB1 and α-tubulin served as the corresponding positive controls. IgG, both the heavy and light chains, shown here served as the IgG input control. Data shown here are representative results of an experiment with n = 3 experiments which yielded similar results.

KD of formin 1 in Sertoli cells by RNAi perturbs the organization MTs

Formin 1 was silenced by RNAi using formin 1-specific siRNA duplexes vs nontargeting negative control siRNA duplexes (Table 2). KD of formin 1 by approximately 90% as confirmed by IB (Figure 2A, left vs right panels) and qPCR (Figure 2A, middle bar graph in right panel) was found to have no apparent effects on the expression of MT-binding and regulatory proteins including EB1 (a +TIP that stabilizes MT in the testis) (30), mammalian Diaphanous-related formin (mDia; a MT stabilizing and nucleation protein and a member of the formins family) (35, 36), katanin (a MT cleavage protein essential for spermatogenesis) (37, 38), because the steady-state level of these proteins remained relatively unaltered when assessed by IB using corresponding specific antibodies (Table 1). However, a consistent and considerably down-regulation of detyrosinated-α-tubulin (ie, removal of C-terminal Tyr by exposing Glu at the newly formed C terminus, which is known to cause MT stabilization by rendering MT less dynamic [39–41], thus, the MTs in these cells after formin 1 KD were less stable) was detected in Sertoli cell cytosol (Figure 2A), which was also supported by findings of IF microscopy (Figure 2B, left vs right panels). Also, MT no longer stretched across the Sertoli cell cytosol after formin 1 KD vs control cells (Figure 2B). Collectively, these findings suggest that formin 1 KD led to reorganization of MTs, so that they tended to fail to stretch across the Sertoli cell cytosol to have more stabilized MTs to confer cellular functions provided by the MT network.

Figure 2. Formin 1 KD by RNAi perturbs MT organization in Sertoli cells cultured in vitro.

A, Sertoli cells cultured in vitro for 3 days with an established TJ-permeability barrier was transfected with formin 1-specific siRNA duplexes (formin 1 siRNA) vs nontargeting negative control siRNA duplexes (Ctrl siRNA) for 24 hours to KD formin 1 by at least 80% when assessed by IB on day 5 (left and right top panel) and also by qPCR using specific formin 1 primer pair (Table 2) (right panel, second bar graph), without any apparent off-target effects, because none of the MT-associated and regulatory proteins vs cytoskeletal proteins were affected except detyrosinated α-tubulin, which was consistently and considerable down-regulated. GAPDH served as a protein loading control. Each bar in the histogram is a mean ± SD of n = 3–5 independent experiments. **, P < .01. B, IF staining confirmed that the expression of formin 1 (green fluorescence) in Sertoli cells was silenced by at least 80%, which also accompanied by a considerable down-regulation of the fluorescence signal of detyrosinated α-tubulin, the stabilized MTs. It was noted that formin 1 KD also perturbed MT organization in which MTs (visualized by α-tubulin staining, green fluorescence) no longer stretched well across the cell cytosol. Cell nuclei were visualized by DAPI (blue). Each bar in the histograms on the right panel is a mean ± SD of n = 4 independent experiments, such as those shown in the left panel by randomly scoring at least 70 Sertoli cells in each experiment. Scale bar, 20 μm, which applies to all other micrographs in B. **, P < .01.

KD of formin 1 in Sertoli cells cultured in vitro by RNAi induces retraction of MTs from cell cortical zone through considerable loss of MT polymerization activity and down-regulation of Akt1/2 signaling function

We next performed detailed analysis on changes in MT organization, probing the underlying molecular mechanism(s) that impeded MT organization in the Sertoli cell epithelium after formin 1 KD. After formin 1 KD, a persistent retraction of MTs from the Sertoli cell cortical zone was detected, because few MTs were detected at or near the cell-cell interface in the treatment vs control group (Figure 3A, left panel), which was also accompanied by the retraction of EB1, a MT plus [+]-end stabilizing protein (+TIP) (11, 42, 43), from the cell cortical zone at or near the Sertoli cell-cell interface (Figure 3A, right panel), consistent with findings shown in Figure 2A that there was a considerable reduction in detyrosinated α-tubulin expression, illustrating MTs in formin 1 KD cells are less stable where MTs failed to stretch across the cell due to a loss in stability. These changes were also likely to be the result of a considerable reduction in MT polymerization in Sertoli cells after formin 1 KD based on a biochemical polymerization as described (30) using α- or β-tubulin (Figure 3B). In this assay, PTX (also called Taxol, a known MT stabilizing agent that inhibits MT depolymerization) (31) and CaCl₂ (known to induce MT depolymerization by blocking MT assembly) (32) served as the positive and negative control to induce and to block MT polymerization, respectively, as noted in the pellet (ie, containing MTs) vs the supernatant (ie, containing free α- or β-tubulins). In short, formin 1 KD induced significant loss of MT polymerization by perturbing the assembly of α- and β-tubulin into MTs (Figure 3B, lower vs upper panel). Furthermore, formin 1 KD in Sertoli cells also led to a significant down-regulation on the expression of p-Akt1-Thr³⁰⁸ and p-Akt1-Ser⁴⁷³, as well as p-Akt2-Ser⁴⁷⁴ (Figure 3C), consistent with earlier reports that Akt1/2 are involved in MT dynamics (44, 45) in mammalian cells. These data further support the notion that formin 1 is intimately related to MT organization, its expression appears to be necessary to confer MT polymerization, possibly through Akt1/2 signaling function.

Figure 3. Formin 1 KD in Sertoli cell epithelium in vitro perturbs MT organization through mislocalization of EB1, and considerable reduction in MT polymerization and down-regulation of Akt1/2 signaling function.

A, After formin 1 KD in Sertoli cell cytosol by at least 80% (see Figure 2), defects in MT organization were noted, in which α-tubulin (green fluorescence) no longer stretched across the entire Sertoli cell cytosol but retracted from the cell cortical zone, such that α-tubulin signal was considerably diminished at the Sertoli cell-cell interface (see enlarged orange and red boxed areas in the left panel). These defects in MT organization possibly mediated by a mislocalization of EB1, a +TIP known to stabilize MT (42), along the MTs, because considerable reduced EB1 was also detected at the Sertoli cell-cell interface (see enlarged orange and red boxed areas in the right panel). Scale bar, 20 μm, and 8 μm in insets, which apply to corresponding micrographs and insets in A. B, MT polymerization assay was performed, in which considerable down-regulation on the ability of cell lysates to induce MT polymerization was noted in Sertoli cells after formin 1 KD when α- and β-tubulin found in the pellet (ie, polymerized MTs; note: because α- and β-tubulins are constituents of MTs via their polymerization, for reviews, see Refs. 7, 11) was significantly reduced (see also the bar graph in lower panel) but not in the supernatant, which quantified the free α- and β-tubulins as monomers. PTX and CaCl₂ served as the corresponding positive and negative controls, which promoted and inhibited MT polymerization, respectively, in the biochemical assay (see text for details). Each bar graph in the bottom panel is a mean ± SD of n = 3 independent MT polymerization experiments, such as those shown on the top panel. C, Formin 1 KD in Sertoli cells also induced considerable down-regulation of Akt1/2 signaling function, in which p-Akt1-Thr³⁰⁸ and p-Akt1-Ser⁴⁷³, as well as p-Akt2-Ser⁴⁷⁴, were considerably down-regulated (see also bar graph shown in the lower panel). Each bar graph is a mean ± SD of n = 3 independent experiments. *, P < .05; **, P < .01.

KD of formin 1 in the testis in vivo perturbs spermatid and phagosome transport

We next examined the effects of formin 1 KD on MT function in the testis in vivo using Polyplus in vivo-jetPEI as a transfection medium for efficient KD of formin 1. When formin 1 was silenced considerably in the testis in vivo without affecting the expression of EB1, mDia, and α-tubulin, except that detyrosinated α-tubulin, the stabilized MTs, was consistently down-regulated (Figure 4, A and B), consistent with in vitro findings shown in Figure 2A; two obvious phenotypes were immediately noted in the testis in 4 days after a single transfection of formin 1-specific siRNA duplexes vs testes transfected with nontargeting negative control siRNA duplexes. First, extensive spermatid loss was detected, in which elongated spermatids located near the tubule lumen in stage VI, VII, and early VIII tubules in formin 1 KD were emptied into the tubule lumen vs control testes as shown in this stage VI–VII tubule (Figure 4C, blue arrowheads). Interestingly, many spermatids remained embedded deep inside the epithelium when premature spermiation occurred in these tubules (see boxed insets in Figure 4C, right vs left panel). However, apical ES that anchored these elongated spermatids onto the Sertoli cell deep inside the epithelium was found to be disrupted (see Figure 5 that showed the F-actin network was equally disrupted for spermatids embedded deep inside the epithelium in the formin 1 KD tubules) as recently reported (17), yet these spermatids failed to be transported to the tubule lumen. In fact, elongated spermatids remained entrapped in the epithelium as noted in the red-, green-, and blue-boxed areas and persisted even in stage IX tubules when elongated spermatids should not be present (Figure 4C). Furthermore, phagosomes derived from the engulfed residual bodies that should have been transported to the base of the epithelium in stage IX tubules for their eventual degradation via the lysosomal pathway (Figure 4C, green arrowheads) (46) remained clearly visible in the apical compartment (Figure 4C, red arrowheads). These observations unequivocally confirm that there were defects in the transport of elongated spermatids and phagosomes after formin 1 KD.

Figure 4. Formin 1 KD in the testis in vivo perturbs the transport of elongated spermatids and residual bodies/phagosomes across the seminiferous epithelium.

A, Using PolyPlus in vivo-jetPEI as the transfection medium with a transfection efficiency up to approximately 80% because red fluorescence aggregates of siGLO red transfection indicator were found in 80% of the tubules randomly scored from transfected testes. A study by immunoblotting using testis lysates, formin 1 was found to KD by approximately 40% in the testis in vivo, without perturbing the expression of several MT regulatory proteins, such as EB1 and mDia, except that detyrosinated α-tubulin was down-regulated, consistent with the finding in vitro shown in Figure 2A. Each bar in the histogram on the right panel is a mean ± SD of n = 3 rats. **, P < .01. B, Fluorescence staining of formin 1 in cross-sections of tubules was considerably down-regulated after formin 1 KD by transfecting testes on day 0, and rats terminated on day 4 as noted here. Each bar in the histogram shown on the right panel is a mean ± SD of n = 3 rats, illustrating a considerable loss in formin 1 fluorescence signal, at least an approximately 60% reduction vs control, in the seminiferous epithelium after its KD. Cell nuclei were visualized by DAPI staining. **, P < .01. Scale bar, 40 μm; inset at 100 μm, which apply to the other micrograph and inset in this panel. C, Histological analysis of the testes after formin 1 KD by transfecting testes with formin 1-specific siRNA duplexes vs nontargeting negative control duplexes in control testes (Ctrl siRNA) on day 0 and rats terminated on day 4 (n = 3 rats). Micrographs on the left panel are cross-sections of control testes, illustrating stage IV–V, IX, and X tubules, in which elongating spermatids were being transported from the base of the epithelium to near the tubule lumen (see also the colored boxed areas magnified and shown in insets) in a stage IV–V tubule, whereas phagosomes (green arrowheads) were rapidly transported to the base of the epithelium in IX and X tubules for lysosomal degradation. Micrographs on the right panel are cross-sections of testes after formin 1 KD, illustrating in this VI–VII tubule that elongating/elongated spermatids near the tubule lumen were depleting from the seminiferous epithelium (blue arrowheads); however, numerous elongating/elongated spermatids were found to remain entrapped deep inside the seminiferous epithelium in this stage VI–VII tubule as shown in insets, which are the corresponding colored boxed areas of the lower magnified micrograph. More important, these elongating/elongated spermatids remained entrapped inside the epithelium even in late stage IX tubules when sperms had undergone their release into the tubule lumen at spermiation, supporting the notion that there was a lack of tracks to allow these spermatids to be transported to near the tubule lumen for their release. Also, phagosomes failed to be transported to the base of the epithelium for their eventual lysosomal degradation (see red arrowheads) Scale bar, 80 μm; inset at 30 μm, which apply to corresponding micrographs in top panel; 40 μm in lower panel which applies to all micrographs in this panel.

Figure 5. Formin 1 KD in the testis in vivo perturbs the organization of polarized actin microfilaments in the seminiferous epithelium.

Using frozen sections of testes, seminiferous tubules at stage II–III, IV–V, VI–VII, and VIII from control testes transfected with nontargeting negative siRNA duplexes (Ctrl siRNA) were shown here (2 left panels), in which F-actin (red fluorescence) that appeared as track-like structures was found to associate with elongating or elongated spermatids being transported across the epithelium, but these track-like structures not associated with elongated spermatids were also noted in stage VIII tubules (see yellow arrowheads). The basement membrane was annotated with a dashed white line, and F-actin at the BTB was clearly detected (see white arrowheads), which laid close to the basement membrane; however, F-actin intensity at the BTB was considerably diminished at stage VIII due to its remodeling to facilitate the transport of preleptotene spermatocytes across the BTB. However, after formin 1 KD, elongating/elongated spermatids no longer associated with F-actin prominently, both across the seminiferous epithelium and at the BTB, and the track-like structures were considerably diminished in the seminiferous epithelium. Scale bar, 30 μm, which applies to other micrographs in this figure.

KD of formin 1 impairs the track-like structures conferred by F-actin for spermatid transport

F-actin appears as “stalk-like” ultrastructures in the seminiferous epithelium is known to work in concert with MTs to facilitate the transport of cellular structures (eg, endocytic vesicles) (for a review, see Ref. 47), including developing spermatids during spermiogenesis (for a review, see Ref. 48). After formin 1 KD, these stalk-like ultrastructures serving as the tracks in different stages of the epithelial cycle of II–III, IV–V, VI–VII, and VIII found in control testes after transfection with nontargeting negative control siRNA duplexes were grossly collapsed, and barely visible in formin 1 KD testes (Figure 5). These findings suggest that the absence of the tracks mediated by actin microfilaments impeded the transport of elongated spermatids embedded deep inside the seminiferous epithelium even though the apical ES anchorage device had already been grossly affected, because F-actin no longer served as the anchoring site for these elongated spermatids (Figure 5). In short, these elongated spermatids with disrupted apical ES failed to be transported to the adluminal edge of the tubule lumen to undergo spermiation due to the absence of the F-actin conferred tracks.

KD of formin 1 impairs the tracks conferred by MTs for spermatid transport

MTs are known to serve as the tracks for the transport of cellular organelles (for reviews, see Refs. 10, 49, 50) such as spermatids during spermiogenesis as well as residual bodies and phagosomes (for reviews, see Refs. 3, 7, 11, 12). These track-like ultrastructures conferred by MTs were clearly visible across the seminiferous epithelium when visualized by IHC using an α-tubulin antibody (Table 1), some of which were tightly associated with elongating/elongated spermatids but some were found on their own without spermatids in control testes transfected with nontargeting negative control siRNA duplexes (Figure 6A). These MT-based track-like structures also stained by IHC using an EB1 antibody, appeared as track-like ultrastructures across the seminiferous epithelium in control testes (Figure 6B). It is noted that both antibodies have earlier been characterized in our laboratory (30). However, KD of formin 1 grossly impaired the α-tubulin conferred (Figure 6A) and also EB1 stabilized (Figure 6B) MTs across the epithelium, in which the track-like structures were considerably down-regulated and their obvious absence thus impeded spermatid and phagosome transport across the epithelium during the epithelial cycle. These findings thus support the notion that formin 1 is necessary to confer the track-like ultrastructures maintained by polarized MTs. These findings also offer an explanation for the presence of residual step 19 spermatids being trapped inside the seminiferous epithelium in stage X–XII tubules as noted in Figure 6, A and B.

Figure 6. Formin 1 KD in the testis in vivo perturbs the organization of polarized MTs and EB1 in the seminiferous epithelium.

A and B, IHC was used to examine changes in the localization and the track-like structures of MT (A) or EB1 (B) in the seminiferous epithelium of control (Ctrl siRNA) vs formin 1 KD (Formin 1 siRNA) testes. In control testis, MT-conferred tracks visualized by either α-tubulin (A) or EB1 (a +TIP known to stabilize MT) (B) staining was noted to be associated with elongating/elongated spermatids (blue arrowheads, stage VII tubule in A and II–III tubule in B) or in their absence (green arrowheads, stage VII tubule in A and VIII tubule in B), which were found in virtually all stages of the epithelium cycle, supporting the notion that MT served as tracks to support the transport of spermatids and phagosomes across the epithelium. However, after the KD of formin 1 by RNAi, these MT-conferred track-like structures were considerably diminished in or absent from the seminiferous epithelium, and defects in spermatid transport were noted after formin 1 KD. Scale bar, 40 μm, which applies to all other micrographs in A or in B.

Discussion

MTs conferred tracks in the seminiferous epithelium are necessary to support elongating/elongated spermatid transport

For the last few decades, it has been suggested that the transport of preleptotene spermatocytes across the BTB and of elongating/elongated spermatids across the adluminal compartment rely on polarized MTs, with assistance of polarized actin microfilaments in the Sertoli cells, involving a testis-specific anchoring device known as the ES, via a mechanism that remains poorly understood until to date (for reviews, see Refs. 4, 6, 11, 51). This is due to the lack of information and understanding on the role of actin- and MT-based binding and regulatory proteins that are working in concert to confer the plasticity to actin microfilaments and MTs. Using formin 1 KD in the rat testis as a model, it was first shown that this actin nucleation and bundling protein is involved in maintaining the homeostasis of F-actin network in the seminiferous epithelium (17). For instance, formin 1 KD by RNAi in the rat testis impedes spermatid adhesion by causing gross disruption of apical ES that induces premature release of spermatids from the epithelium in stage VI, VII, and early VIII tubules. However, it was noted that elongated spermatids remained embedded deep inside the seminiferous epithelium even though the anchoring device apical ES had been compromised due to the F-actin network disruption (17). Here, the earlier observation was further expanded. The failure of the elongated spermatids embedded deep inside the epithelium with compromised apical ES to be transported to the edge of the tubule lumen to undergo spermiation is likely due to the missing track-like structures conferred by MTs and also actin filaments that are necessary to support proper spermatid transport. As such, even though the apical ES function was compromised (17), these elongated spermatids failed to leave the epithelium but remain entrapped through stage IX and later staged tubules until they were eventually removed from the epithelium by Sertoli cells through phagocytic activities. As expected, the transport of phagosomes was also impeded. Although the precise molecular mechanism that affects spermatid transport remains to be elucidated, which likely involved motor proteins such as MT-specific motor protein kinesins which serve as walking machines to support cargo (ie, spermatids, phagosomes) transport along MTs, involving consumption of ATPs as recently reported in other mammalian cells (for a review, see Ref. 49). Nonetheless, our findings illustrate for the first time that formin 1, a known actin nucleator and an actin bundling protein (for reviews, see Refs. 13, 18), also modulates MT dynamics, possibly by providing the necessary cross talk to confer plasticity to the actin- and MT-based filaments.

Because formin 1 is not a transcription factor, it is not likely that it plays in role in regulating the expression of other genes pertinent to the BTB integrity. As such, even though formin 1 modulates the organization of MTs, its KD has no effects on expression of proteins known to affect MT organization, such as EB1 as noted here. However, it is of interest to note that its KD by RNAi modulates F-actin (17) and also MT organization, which in turn impedes F-actin- and MT-dependent endocytic vesicle-mediated protein trafficking (for reviews, see Refs. 34, 52, 53), thereby affecting the cellular distribution of EB1 and other actin- and/or MT-binding and regulatory proteins, because formin 1 dysfunction through its KD would perturb proper intracellular cargo trafficking. Indeed, EB1 distribution in Sertoli cells was grossly affected as noted here. Although the precise mechanism how formin 1 affects these cellular events will requires further investigation, the finding that formin 1 KD impedes the expression of Akt1/2 activated forms is of great interest. The oncogenic Akt protein kinase (also known as protein kinase B) is known to mediate a wide range of cellular functions in response to external cues such as growth and development, as well as changes in the stage of the epithelial cycle during spermatogenesis in the testis (for reviews, see Refs. 44, 54). In fact, Atk/protein kinase B was earlier shown to localize with elongating/elongated spermatids at the apical ES that are being transported across the seminiferous epithelium (55). Our finding that a disruption of MT organization after formin 1 KD that led to a down-regulation of phosphorylated (p)-Akt1-Thr³⁰⁸, p-Akt1-Ser⁴⁷³, and p-Akt2-Ser⁴⁷⁴ are also consistent with earlier findings that Akt1 activity is down-regulated after MT disruption in HEK293 and HeLa cells (for a review, see Ref. 44). Because studies have shown that Akt can be regulated by up-stream signaling kinases, such as mammalian target of rapamycin complex 2 (mTORC2) and phosphoinositide-dependent kinase 1 (for reviews, see Refs. 56, 57), it appears that the formin 1-mediated Akt activation can involve an array of signaling pathways including mTOR. This notion is supported by recent studies that mTORC1 and mTORC2 are displaying antagonistic effects on the BTB dynamics in which mTORC1 promotes Sertoli cell BTB remodeling (ie, making the Sertoli cell TJ barrier leaky) (58) through Akt signaling function through matrix metalloprotease 9 (59), which also perturbs the actin microfilament organization at the Sertoli cell BTB (60). On the other hand, mTORC2 promotes Sertoli cell BTB integrity by making the Sertoli cell TJ barrier tighter (61) through gap junction communication and reorganization of the F-actin at the cell cortical zone to strengthen the TJ barrier. However, it is not known in these earlier studies whether mTORC1 and mTORC2 also modulate MT dynamics, which should be carefully evaluated in future studies. Taking collectively, these findings have demonstrated that formin 1 is working in concert with Akt1/2, likely involving mTOR signaling complex to modulate F-actin but also MT organization. Work is also in progress to assess whether the use of constitutively active Akt1/2 mutants can modulate formin 1 expression and/or MT (or actin microfilament) organization in Sertoli cells, which will provide some of the needed information on the physiological relationship between formin 1 and Akt1/2 signaling function.