Chemicals

Rabbit polyclonal antibody anti-human cathepsin D (sc-10725), rabbit polyclonal antibody anti-human PTP1B (protein tyrosine phosphatase, sc-14021) and MCF7 whole cell lysate were from Santa Cruz Biotechnology (Palo Alto, CA, USA). All commercial antibodies were chosen on the base of their conserved sequences, using for the comparisons the database of insects (taxid:6960) of the National Center for Biotechnology Information (NCBI). Alexa Fluor 488- and 568-conjugated goat anti-rabbit IgG antibodies (Molecular Probes, Carlsbad, CA, USA); Tissue-Tek embedding medium (OCT) (Miles, Elkhart, IN, USA); Fluorsave (Calbiochem, Darmstadt, Germany); Western Lightning Plus-ECL, enhanced chemiluminescence substrate (PerkinElmer, Waltham, MA, USA) and electrophoresis protein standards (New England Biolabs, Ipswich, MA, USA) were from indicated commercial sources. Assay kits for serine/threonine phosphatase (V2460) and tyrosine phosphatase (V2471), as well as MMLV reverse transcriptase were obtained from Promega (Heidelberg, Germany). Primers were from Sigma Genosys (Houston, TX, USA); MasterPure RNA Purification Kit was obtained from Epicenter Biotechnologies (Madison, WI, USA) and Brilliant SYBR Green qPCR Master Mix was a product of Stratagene from Agilent Technologies, Inc. (Santa Clara, California, USA). All other chemicals were from Sigma-Aldrich (St. Louis, MO, USA). The fluorogenic peptide substrate Abz-AIAFFSRQ-EDDnp (Abz, orthoaminobenzoic acid, EDDnp, ethylenediamine-2,4-dinitrophenyl) was a kind gift from Dr. Maria Aparecida Juliano (Universidade Federal de São Paulo, Brazil).

Insects

Experiments were carried out using insects taken from a colony of anautogenous D. maxima, maintained at 28°C, 70% relative humidity and 8:16 h light:dark photoperiod. Insects free of T. cruzi or Blastochritidia triatomae were fed on hen blood [30] according to the recommendations of the National Institute of Parasitology (National Health Ministry, Argentina) [31].

Standardized conditions of insect rearing for the study were previously described [27, 28]. Briefly, fifth-instar females were separated from males before feeding. Upon molting, each female was placed in individual containers with two fed males during 48 h. The presence of the spermatophore in the container was considered as evidence of successful mating. Each mated female was maintained in an individual jar until it was able to ingest a blood meal (days 10–12 post-ecdysis), which resulted in a 3.0–5.5 fold increase in the body weight of the insect. Thereafter, egg laying was daily monitored and both, the beginning and the end of the oviposition period, which was confirmed if females did not lay eggs during five consecutive days, were recorded. Females were monitored at least for 32 days after the end of the oviposition (post-vitellogenesis). During post-vitellogenesis females did not receive any further blood meal [27].

Experimental approaches were performed by sampling fat bodies, ovaries and hemolymph at representative days of the reproductive cycle as previously described: (a) pre-vitellogenesis (day 2 post-ecdysis, unfed period); (b) vitellogenesis (days 4–6 post-blood feeding); (c) post-vitellogenesis (days 10–12 and days 30–32 after the end of oviposition for early and late atresia, respectively) [27, 28].

Ethics Statement

Housing conditions and manipulation of hens followed current protocols of the Centro de Investigaciones en Bioquímica Clínica e Inmunología (CIBICI-CONICET-Universidad Nacional de Córdoba) animal facility, in accordance with the guidelines published by the Canadian Council on Animal Care with the assurance number A5802-01 delivered by the Office of Laboratory Animal Welfare (National Institutes of Health). The protocol was authorized by the Animal Care Committee of CIBICI-CONICET-Universidad Nacional de Córdoba (Exp. 15-UNC-05-54976). According to institutional policy, no specific number of approval was assigned since no infective species, human blood or animal sacrifice were involved in the study. The animal facility at the Centro de Investigaciones en Bioquímica Clínica e Inmunología (CIBICI-CONICET-Universidad Nacional de Córdoba) is a dependency of the Argentine National Ministry of Science (Sistema Nacional de Bioterios, MINCyT, http://www.bioterios.mincyt.gob.ar). Briefly, hens were housed individually in a room at 25°C, with food and water available ad libitum, under a light-dark cycle of 12 h each. Cages were cleaned daily and hens’ health was regularly checked by a veterinary physician. Special care was taken in their handling in order to minimize stressful conditions. During the process of bug feeding (30 min, 4 insects/animal), hens were maintained in a special dark room and secured to the table by soft bands. Each of the hens participated in feeding purposes only once in a month.

Tissue and hemolymph sampling

Ovaries and fat bodies from females at different days of the reproductive cycle were carefully dissected out under cold phosphate buffered saline (PBS, 6.6 mM Na₂HPO₄/KH₂PO₄, 150 mM NaCl, pH 7.4), using a standard stereoscope with an optic fiber light source.

The hemolymph was individually collected from immobilized females with a Hamilton syringe and placed in ice-cold microtubes in the presence of 10 mM Na₂EDTA and 5 mM dithiotreitol (DTT) as previously reported [27]. For western blot assays, a mixture of protease inhibitors containing 1 μM aprotinin, 0.5 μM N-α-tosyl-L-lysine chloromethyl ketone (TLCK) and 1 mM benzamidine was added to each sample. All the hemolymph samples were centrifuged at 10,000 x g for 5 min at 4°C to remove hemocytes. Thereafter, samples were employed for the biochemical assays, previous protein determination [32].

RNA extraction and Reverse Transcription/quantitative PCR (RT-qPCR)

Taking into account the role of fat body and ovarian follicular cells in the synthesis of yolk protein precursors, transcriptional levels of DmCatD were determined in fat bodies and ovaries sampled at different stages of the reproductive cycle of D. maxima. For RNA extraction, dissected tissues of three females were pooled and the MasterPure RNA Purification Kit was used according to the manufacturer´s protocol. To eliminate genomic DNA, samples were treated with DNAse provided in the kit. RNA integrity was evaluated by electrophoresis in a 1% agarose gel. cDNA was synthesized from 2 μg of total RNA by reverse transcription reaction using the MMLV reverse transcriptase protocol. Real time PCR (qPCR) analysis was performed using an ABI Prism 7500 sequence detection system (Applied Biosystems, Foster City, CA, USA) and SYBR green chemistry as described [33]. The 2^−ΔΔCt method was used to quantify relative changes in gene expression using 18S ribosomal RNA (18S rRNA) as internal control. Reactions were carried out in triplicate using, for each 15 μl reaction, 5 μl of cDNA template (equivalent to 300 ng of total RNA). For each pair of primers, a dissociation plot resulted in a single peak, indicating that a single cDNA product was amplified (S1A and S1B Fig). Moreover, re\actions were subjected to agarose gel electrophoresis to check amplification specificity. All reactions showed an amplification efficiency higher than 98%. Specific target amplification was confirmed by automatic sequencing (Macrogen, Seoul, Korea). The primers sequences for DmCatD (96 pb) were: sense, 5`-GGTGAAGCACTGCTAGTGTCG-3´ and antisense, 5´-AGCTTGATCTTCGGCAATTG-3´. The primer sequences for the internal control 18S rRNA (104 bp) were: sense, 5´-TCGGCCAACAAAAGTACACA-3´ and antisense, 5´-TGTCGGTGTAACTGGCATGT-3´. The design of primers was based on the available mRNA sequences of related species.

DmCatD expression by western blot

Fat bodies and ovaries from females dissected at different stages of the reproductive cycle were individually homogenized in buffer Tris-HCl-Na₂EDTA (50 mM Tris, 1 mM EDTA, 0.1% Triton X-100, pH 7.5) containing a mixture of protease inhibitors (1 μM aprotinin, 0.5 μM TLCK and 1 mM benzamidine). The first centrifugation was performed at 2,500 x g for 10 min at 4°C and the floating fat cake and pellet were discarded. The resulting material was subjected to a second centrifugation at 20,000 x g for 20 min at 4°C and the supernatants collected and used for western blotting. Protein concentration in tissue homogenates was determined according to Bradford [32].

For western blots, protein extracts from fat body and ovary homogenates as well as hemolymph (40 μg each one) were subjected to 10% SDS-Tris-Tricine gel electrophoresis [34] and then transferred onto a nitrocellulose membrane for 1 h as described elsewhere [35]. The membrane was incubated with Ponceau S and the images were registered. After rinsing with 10 mM Tris-HCl buffer containing 150 mM NaCl (TBS, pH 7.5), the nitrocellulose was blocked with TBS-0.1% Tween 20 containing 5% non-fat milk at room temperature. The antibodies were diluted in TBS-0.1% Tween 20 containing 5% BSA and used in the incubation steps as follows: polyclonal anti-cathepsin D antibody (primary antibody, dilution 1:300, at 4°C, overnight) and HRP conjugated goat anti-rabbit IgG (secondary antibody, dilution 1:2000, for 1 h at room temperature). All incubations were carried out with gentle rocking. The immunoreactive bands were detected by ECL, according to the manufacturer’s instruction. Densitometric analyses of fat bodies and ovaries western blot profiles were performed using ImageJ software. For each lane, the intensity of the bands compatible with pro-DmCatD and DmCatD and the amount of total proteins loaded and stained with Ponceau S were considered (S2A and S2B Fig). MCF7 whole cell lysate derived from the MCF cell line was used as a positive control of western blot assays.

Measurement of DmCatD activity

The activity of DmCatD peptidase upon the specific synthetic fluorogenic substrate Abz-AIAFFSRQ-EDDnp was assayed either in aliquots of hemolymph or in tissue homogenates (fat bodies or ovaries) containing 30 μg of proteins according to the protocol previously reported [27]. Reactions were monitored every 11 s for 30 min in an F-Max fluorometer (Molecular Devices Inc., Silicon Valley, CA, USA) with 320/430 nm excitation/emission filters. Results were expressed as relative units of fluorescence (RUF)/μg protein/min.

Measurement of acid phosphatase activity

The activity of acid phosphatase in ovarian tissues was assayed measuring the hydrolysis of p-nitrophenyl phosphate (pNPP) to p-nitrophenol (pNP) as reported previously [28]. Briefly, the tissues were individually homogenized in 20 mM sodium acetate buffer pH 4.0 containing 10 mM DTT, 10 mM Na₂EDTA, 2 mM phenylmethylsulfonyl fluoride (PMSF), 0.01 mM pepstatin and 50 μM soybean trypsin inhibitor (SBTI). Homogenates were then centrifuged at 14,000 x g for 5 min at 4°C and the resulting supernatants were subjected to protein quantification [32]. Enzymatic assays were performed using 5 mM pNPP as substrate and 30 μg of protein homogenates as enzyme source. The amount of pNP released into the medium was registered at 405 nm. Specific activity was expressed as nmol pNP/mg protein/min. Acid phosphatase activity was also determined by pre-incubating the homogenates from females at early atresia with NaF (an inhibitor of serine/threonine phosphatases), Na₃VO₄ (an inhibitor of tyrosine phosphatases) or with Na⁺/K⁺ tartrate (a broad spectrum phosphatase inhibitor) at a final concentration of 1 mM each (30 min, at 37°C). Thereafter, assays were carried out as previously stated.

On the other hand, enzymatic assays using specific synthetic phosphopeptide substrates were performed to determine if the activity of acid phosphatase found in ovarian tissue during follicular atresia corresponded to either, a protein tyrosine phosphatase or a protein serine/threonine phosphatase. The assays were carried out on microplates provided in the commercial kits for serine/threonine phosphatase and tyrosine phosphatase following their manufacturer’s instructions. Three-five μg of protein extracts from ovaries at early and late atresia were added to the incubation media. The generation of free phosphate was determined by measuring the absorbance of a complex formed by phosphate, molybdate, and malachite green using the Multi-Mode Microplate Reader Sinergy HT (BioTek Instruments, Winooski, VT, USA) at 600 nm [36]. The specific activity was expressed as pmol phosphate/μg protein/min. To further characterize the class of acid phosphatase, aliquots of ovarian homogenates at both atretic stages were pre-incubated for 30 min with NaF, Na₃VO₄ or Na⁺/K⁺ (1 mM each, 30 min at 37°C). Then, the enzymatic assays for tyrosine or serine/threonine phosphatases were performed as described above.

Localization of DmCatD and acid phosphatase in ovarian tissue

Immunofluorescence assays for DmCatD and acid phosphatase were directed to assess their localization in ovarian tissue, particularly in oocytes, due to the proposed physiological role of acid hydrolases on vitellin degradation. For the assays, ovaries from females in vitellogenesis and early follicular atresia sampled as described previously were fixed in 4% paraformaldehyde in PBS, transferred to sucrose/PBS and then embedded in OCT and frozen in liquid nitrogen [28]. Thereafter, tissue cryosections (8 μm thick) were obtained with a Cryotome E Cryostat (Thermo Fisher Scientific Inc., Waltham, MA, USA) and placed onto poly-L-lysine-treated glass slides. For the assays, the slides were blocked with 1% BSA, 0.1% Triton X-100, 5% fetal bovine serum in PBS during 1 h, and then incubated overnight at 4°C with the anti-cathepsin D antibody (dilution 1:100) or the anti-tyrosine phosphatase antibody (dilution 1:50). Slides were washed three times with PBS and then incubated for 1 h at 37°C with either Alexa Fluor 568-conjugated goat anti-rabbit IgG antibody (for DmCatD, dilution 1:1000) or Alexa Fluor 488-conjugated goat anti-rabbit IgG antibody (for tyrosine phosphatase, dilution 1:1000). In all cases, the antibodies were diluted in PBS containing 1% BSA. Finally, the slides were rinsed with PBS, air-dried, mounted in Fluorsave and viewed with an Olympus Fluoview 300 laser scanning confocal microscope (Olympus, Tokyo, Japan) equipped with a 543 and 488 nm lasers.

Anti-vitellin antibody and labeling with FITC

The polyclonal anti-vitellin antibody was obtained by inoculating New Zealand rabbits with purified vitellin as described previously [37]. For co-localization assays, anti-vitellin antibody was labeled with fluorescein isothiocyanate (anti-Vt-FITC, 5 mg/ml) in dimethyl sulfoxide according to Hermanson [38].

Co-localization of DmCatD and acid phosphatase with vitellin in ovarian tissue

These set of studies were directed to obtain evidence about co-localization in yolk bodies of acid hydrolases and vitellin, which in turn would be of functional relevance in the process of yolk protein degradation during atresia. For co-localization assays, ovarian cryosections from vitellogenic females were blocked and then subjected to indirect immunofluorescence assays for DmCatD and tyrosine phosphatase as stated above. After rinsing with PBS, the tissue cryosections were incubated with anti-Vt-FITC in PBS/1% BSA (dilution 1:40) for 1 h at 37°C. Finally, the slides were rinsed with PBS and processed for laser confocal microscopy (Olympus Fluoview 300) as previously described.

In another set of experiments, after the immunofluorescence protocol was performed as described above, the potential interaction of both acid hydrolases with vitellin in the yolk bodies of vitellogenic oocytes was tested by the fluorescence resonance energy transfer (FRET) method [39]. In the fluorescent microscopy, FRET was detected by comparing the donor fluorescence intensity in the same sample before and after destroying the acceptor by photobleaching. If FRET was present, a resultant increase in donor fluorescence will occur on photobleaching of the acceptor. Experiments were performed in an Olympus Fluoview 300 laser scanning confocal microscope, applying a 15 s pulse of high-intensity laser at 543 nm to bleach the fluorescent signal of the enzymes (acceptors). If any interaction leading to energy transfer to a donor (vitellin) were present in the yolk bodies, photobleaching of the acceptor would lead to an increase in the fluorescent signal of the donor molecule because it would no longer be quenched by the acceptor [39]. Results were analyzed by processing a series of pre-bleaching and post-bleaching images, using the FluoView v5.0 software.

In vitro vitellin proteolysis assays

In order to determine the participation of DmCatD and acid phosphatase in the degradation of vitellin during follicular atresia, in vitro proteolysis assays were performed with purified vitellin as substrate [37] and, as enzyme source, protein homogenates from ovaries at early follicular atresia obtained as described for western blot assays.

The pH dependence of in vitro vitellin proteolysis was analyzed by incubating 30 μg of vitellin and 70 μg of protein homogenates at 37°C during 12 h using the buffer systems containing 200 mM NaCl and 5 mM Na₂EDTA as follows: 50 mM ammonium formate (for pH 3.0 and 3.5), 50 mM sodium acetate (for pH ranging from 4.0 to 5.0) and 50 mM sodium phosphate (for pH 6.0 and 7.0). Reactions were stopped by the addition of SDS-PAGE sample buffer and then boiled for 5 min [10, 40]. Aliquots containing 50 μg of the reaction mixture were fractionated by SDS-PAGE (7.5%) and then subjected to western blot. The main subunits of vitellin (Mr ~170 kDa and 174 kDa, detected as a single immunoreactive band) and its proteolytic fragments were probed with anti-vitellin antibody (dilution 1:5000) as previously described [27]. Bands were visualized by ECL.

Inhibition of vitellin proteolysis in vitro was analyzed by performing the assays according to conditions stated above, in a reaction medium containing 50 mM sodium acetate, 200 mM NaCl, 5 mM Na₂EDTA (pH 4.0), and in the presence or absence of acid phosphatase inhibitors (2 mM Na₃VO₄ or 10 mM Na⁺/K⁺ tartrate) or an specific inhibitor of aspartic peptidases (80 μM pepstatin A). Controls containing aliquots of ovarian homogenate and without addition of purified vitellin were performed to rule out a significant contribution of vitellin by atretic ovaries to the in vitro system.

Statistical analyses

The results are presented as the mean ± Standard Error of the Mean (SEM) from at least three independent experiments, including at each point the ovaries, fat body and hemolymph from females (individuals or pools) as stated in each section. Graphs and statistical tests were performed using GraphPad Prism 6.0 and GraphPad Instat 3.0. (GraphPad Software, San Diego, CA, USA). All datasets passed normality and homoscedasticity tests. Multiple group analysis was conducted by one-way ANOVA. The Student-Newman-Keuls multiple-comparisons test was used as a post-test. A P value < 0.05 was considered statistically significant.

DmCatD expression in fat body and ovarian tissue

In order to analyze the gene expression of DmCatD in fat bodies and ovarian tissues of D. maxima females at different stages of the reproductive cycle, its transcription profile by qPCR and protein expression by western blot were assessed.

A single PCR product of the expected size and sequence was amplified (S3A–S3C Fig). When a BLAST analysis was performed (Basic Local Alignment Search Tool) [41] using the default setting, the best match retrieved (89% identity) was the cathepsin D (mRNA) from Triatoma infestans (GenBank: GQ871498.1), a related triatomine species. As shown in Fig 1, DmCatD transcripts were detected in both tissues. However, in the fat body, the highest mRNA levels were found at pre-vitellogenesis and vitellogenesis. On the other hand, DmCatD mRNA was upregulated in vitellogenesis when compared to that from pre-vitellogenic and atretic ovaries (Fig 1A). At late atresia, when all ovarioles had terminal follicles in advanced degeneration and oocytes are resorbed, DmCatD mRNA levels were downregulated in ovarian tissue (Fig 1B). Similar amplification patterns were observed in fat bodies and ovarian tissues when conventional PCR assays were performed (S4A and S4B Fig).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Relative DmCatD mRNA quantification evaluated by RT-qPCR.

Total RNA was extracted from fat bodies (A) and ovaries (B) of females at different stages of the reproductive cycle. Quantification of 18S ribosomal RNA was used as internal control. Results expressed as mean ± SEM indicate the fold of change respect to the mRNA levels at pre-vitellogenesis (n = 3). (A), *P<0.01 vs. early and late atresia. (B), *P<0.05 vs. pre-vitellogenesis and early atresia; **P<0.001 vs. pre-vitellogenesis, early and late atresia.

https://doi.org/10.1371/journal.pone.0130144.g001

The changes in the expression of DmCatD protein in fat bodies and ovaries were evaluated by western blot, employing a commercial anti-cathepsin D antibody that recognizes both, the immature proenzyme (pro-DmCatD) and the active form of cathepsin D (DmCatD). Thus, in both tissues, two main immunoreactive bands of molecular weights compatible with pro-DmCatD (~43 kDa) and DmCatD (~25 kDa) were detected at all reproductive stages. Minor faint bands corresponding to either proteolytic fragments or nonspecific immunoreactive signals were also detected (Fig 2A and 2C). Densitometric analyses performed on western blot patterns showed that in fat bodies and ovaries, expression of pro-DmCatD was maximal at vitellogenesis. Pro-DmCatD protein expression then decreased in both tissues at early and late atresia, in coincidence with increased levels of DmCatD (Fig 2A and 2C). These results were in agreement with the increasing levels of DmCatD activity found in fat bodies and ovaries in early and late atresia (Fig 2B and 2D).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Protein expression and activity of DmCatD in fat bodies and ovaries of D. maxima.

The expression of pro-DmCatD and DmCatD was analyzed by western blot (A and C, top panels). Homogenates of fat body (A) and ovarian tissue (C) obtained at pre-vitellogenesis, vitellogenesis, early and late follicular atresia (lanes 2–5 respectively, 40 μg each) were probed with a polyclonal anti-cathepsin D antibody. As a positive control, 0.2 μg of whole cell lysate of MCF7 was loaded in the lane 1. The arrows indicate pro-DmCatD (~43 kDa) and DmCatD (~25 kDa). The western blots shown were a representative experiment of three independent assays. Densitometric analyses of three independent western blots (A and C, bottom panels) were performed taking into account the intensity of the bands compatible with pro-DmCatD or DmCatD detected in each lane and the amount of total proteins loaded and stained with Ponceau S. Results are expressed as mean ± SEM (n = 3). (A), *P<0.01 vs. pre-vitellogenesis, early and late atresia; (C), *P<0.001 vs. vitellogenesis, early and late atresia; **P<0.001 vs. vitellogenesis and early atresia. (B and D), activity of DmCatD in fat bodies (B) and ovarian homogenates (D) throughout the reproductive cycle. The reactions were performed using 10–30 μg of protein homogenates according to conditions stated in Material and Methods. Results are expressed as relative units of fluorescence (RUF)/μg protein/min and are the mean ± SEM (n = 3). (B), *P<0.05 vs. pre-vitellogenesis, early and late atresia. (D), *P<0.001 vs. pre-vitellogenesis and vitellogenesis.

https://doi.org/10.1371/journal.pone.0130144.g002

DmCatD in the hemolymph

When DmCatD was analyzed in the hemolymph of D. maxima females at different stages of the reproductive cycle by western blot, only the immunoreactive band compatible with pro-DmCatD was detected. Western blot showed low levels of pro-DmCatD at pre-vitellogenesis and increased levels of the protein at vitellogenesis and early atresia (Fig 3A, lanes 2–4). At late atresia, the amount of pro-DmCatD in the hemolymph noticeably decreased, being observed as a faint immunoreactive band (Fig 3A, lane 5).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 3. Protein expression and activity levels of DmCatD in the hemolymph of D. maxima.

(A), western blot was performed to evaluate the expression of pro-DmCatD and DmCatD. The hemolymph was obtained at pre-vitellogenesis, vitellogenesis, early and late follicular atresia (lanes 2–5 respectively, 40 μg each). In lane 1, 0.2 μg of whole cell lysate of MCF7 was used as a positive control. The arrow indicates pro-DmCatD (~43 kDa). The western blot shown was a representative experiment of three independent assays. (B), specific activity of DmCatD in the hemolymph. Assays were performed as stated in Materials and Methods. Results expressed as relative units of fluorescence (RUF)/μg protein/min are the mean ± SEM (n = 3). *P<0.05 vs. pre-vitellogenesis, early and late atresia.

https://doi.org/10.1371/journal.pone.0130144.g003

Although in the hemolymph no immunoreactive band compatible with DmCatD was detected in our experimental conditions, all the hemolymph samples obtained from females at different stages of the reproductive cycle displayed activity upon a specific fluorogenic substrate of DmCatD. The activity was significantly high at vitellogenesis (Fig 3B). However, it is noteworthy that the activity levels of DmCatD in the hemolymph were lower than those registered in fat bodies and ovarian tissues (Fig 2B and 2D).

Acid phosphatase activity in ovarian tissue

Acid phosphatase activity upon yolk proteins seems to increase their susceptibility to proteolysis during embryogenesis [42]. In addition, follicular atresia in D. maxima is characterized by partial vitellin proteolysis and significant activity levels of acid phosphatase in ovaries [27, 28]. In this context, we have investigated the class of acid phosphatase present in ovarian tissue homogenates, focusing in the atretic stages.

In the assays using p-NPP as substrate, significantly higher activities of acid phosphatase were detected at early and late follicular atresia compared to pre-vitellogenesis and vitellogenesis, although the highest activity was found at days 10–12 post-vitellogenesis (early atresia). Homogenates of ovaries sampled at pre-vitellogenic and vitellogenic stages displayed similar low levels of acid phosphatase activity (Fig 4A). Moreover, in homogenates of tissues at early atresia, the high activity upon p-NPP was strongly inhibited by Na₃VO₄ (an inhibitor of tyrosine phosphatases, ~77% inhibition) and Na⁺/K⁺ tartrate (a broad spectrum phosphatase inhibitor, ~87% inhibition). Inhibition by NaF, an inhibitor of serine/threonine phosphatases, was also observed although to lesser extent (~51%, Fig 4B). When the inhibitors were tested at higher concentrations, no significant changes on their inhibitory effect were observed. These results strongly suggested that the main responsible of the high acid phosphatase activity found in ovaries at atretic stage was a tyrosine phosphatase.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 4. Activity of acid phosphatase in the ovarian tissue of D. maxima.

(A), the enzymatic assays were performed using 30 μg of ovarian homogenates from females at representative days of the reproductive cycle and measuring the amount of p-nitrophenol (pNP). The results are expressed as specific activity (nmol of pNP/mg protein/min) and each point represents the mean ± SEM (n = 3). *P<0.05 vs. pre-vitellogenesis, vitellogenesis and early atresia; **P<0.001 vs. pre-vitellogenesis and vitellogenesis. (B), the effect on acid phosphatase activity of several inhibitors was analyzed by pre-incubating protein homogenates of ovaries at early atresia with Na₃VO₄ (inhibitor of tyrosine phosphatase), NaF (inhibitor of serine/threonine phosphatase) or Na⁺/K⁺ tartrate (a broad spectrum phosphatase inhibitor). *P<0.05 vs. Na₃VO₄ and Na⁺/K⁺ tartrate; **P<0.001 vs. Na₃VO₄, NaF and Na⁺/K⁺ tartrate. Specific activity of tyrosine phosphatase (C) and serine/threonine phosphatase (D) in ovarian homogenates from females at early and late atresia. The assays were performed using specific phosphopeptides as substrates either in the presence or in the absence of NaF, Na₃VO₄ or Na⁺/K⁺ tartrate. The specific activities are expressed as pmol phosphate/μg protein/min. Each point represents the mean ± SEM (n = 3). (C), activity in early atresia: **P<0.001 vs. Na₃VO₄ and Na⁺/K⁺ tartrate; activity in late atresia: *P<0.05 vs. Na₃VO₄ and Na⁺/K⁺ tartrate. (D), activity in early atresia: *P<0.01 vs. NaF and Na⁺/K⁺ tartrate; activity in late atresia: *P<0.01 vs. NaF and Na⁺/K⁺ tartrate; **P<0.001 vs. NaF and Na⁺/K⁺ tartrate.

https://doi.org/10.1371/journal.pone.0130144.g004

To further determine the class of acid phosphatase present in atretic ovarian tissues, enzymatic assays employing specific phosphopeptides as substrates of tyrosine phosphatase and serine/threonine phosphatase were performed (Fig 4C and 4D). Results showed that at early atresia, the tyrosine phosphatase activity levels were approximately twice higher than those registered at late atresia. The high activity levels of tyrosine phosphatase in atretic homogenates were significantly inhibited by Na₃VO₄ (~87% and 76% inhibition in early and late follicular atresia, respectively) whereas no significant inhibition by NaF (a serine/threonine phosphatase inhibitor) was observed (Fig 4C). However, treatment of homogenates with NaF inhibited the activity of serine/threonine phosphatase (~69% and 78% inhibition in early and late atresia, respectively). No effect of Na₃VO₄ on the activity of this class of acid phosphatase was observed (Fig 4D). As expected, tyrosine and serine/threonine phosphatase activities were significantly inhibited with Na⁺/K⁺ tartrate, a broad spectrum phosphatase inhibitor.

Localization of DmCatD and acid phosphatase in ovarian tissue

Localization of DmCatD and tyrosine phosphatase in ovarian tissues at different stages of the reproductive cycle of D. maxima was assessed by indirect immunofluorescence assays, using commercial antibodies. In vitellogenic follicles, the fluorescent signal for both acid hydrolases was clearly associated to yolk bodies (Fig 5A). By contrast, in early atresia, oocytes of follicles undergoing incipient degeneration displayed a homogeneous punctate fluorescent pattern for DmCatD and tyrosine phosphatase (Fig 5B). Moreover, in early atresia, DmCatD was also observed in the basal area of the follicular epithelium, whereas tyrosine phosphatase signal was localized in the perioocytic space of terminal follicles (Fig 5B).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 5. DmCatD and tyrosine phosphatase localization in ovarian follicles by indirect immunofluorescence.

Ovaries from females at vitellogenesis and early atresia were incubated with an anti-cathepsin D antibody (red) or with an anti-tyrosine phosphatase antibody (green) and processed as stated in Materials and Methods. Cryostat sections were analyzed by scanning laser confocal microscopy. (A), in vitellogenic follicles, the fluorescent signal for both acid hydrolases was associated to yolk bodies (arrowheads). (B), at early atresia, oocytes of follicles undergoing incipient degeneration displayed a homogeneous punctate fluorescent pattern for DmCatD and tyrosine phosphatase (arrowheads). At early atresia, DmCatD was also observed in the basal area of follicular epithelium (asterisk), whereas tyrosine phosphatase signal was localized in the perioocytic space of terminal follicles (arrow). Insets correspond to DIC images at lower magnifications. Oo, oocyte; FE, follicular epithelium; YB, yolk bodies. Bars: 10 μm. Similar results were observed after examination of 4–5 ovarioles per ovary in each separate experiment (n = 3).

https://doi.org/10.1371/journal.pone.0130144.g005

Additionally, DmCatD was evaluated in fat bodies sampled from females at vitellogenesis and early follicular atresia. At both stages DmCatD showed a fluorescent punctuated pattern homogeneously distributed thorough the cytoplasm of the trophocytes but the intensity of the signal was less evident at early atresia (S5 Fig).

Co-localization and interaction of DmCatD and acid phosphatase with vitellin

The co-localization of DmCatD and tyrosine phosphatase with vitellin was further investigated by immunofluorescence, analyzing yolk bodies of vitellogenic oocytes. Results showed a partial co-localization of both enzymes with vitellin (Fig 6, merge images). However, it seems that such co-localization of DmCatD and vitellin was restricted to some peripheral areas of yolk bodies (Fig 6A, merge) whereas a more heterogeneous pattern of co-localization was observed between tyrosine phosphatase and vitellin (Fig 6B, merge).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 6. Co-localization of DmCatD and tyrosine phosphatase with vitellin in yolk bodies of vitellogenic oocytes.

Vitellogenic ovaries were processed for immunofluorescence as stated in Materials and Methods. (A-B), the vitellin (Vt) signal is displayed in green and the enzyme signal (DmCatD or tyrosine phosphatase), in red. In merged images, the partial co-localization of Vt/DmCatD (A) and Vt/tyrosine phosphatase (B) is indicated with arrowheads. Similar results were obtained in three separate experiments. Bars: 2 μm.

https://doi.org/10.1371/journal.pone.0130144.g006

On the other hand, when FRET was used to test the interaction between both acid hydrolases and vitellin, a resultant increase in donor green fluorescence (vitellin) was observed after photobleaching of the red fluorescence of the acceptors (DmCatD or tyrosine phosphatase) (Fig 7A and 7B, post-bleach). These results indicated that both enzymes and vitellin were at a distance of 1–10 nm, supporting their potential interaction.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 7. Interaction of DmCatD and tyrosine phosphatase with vitellin in yolk bodies of vitellogenic oocytes.

Vitellogenic ovaries were processed for immunofluorescence as stated in Materials and Methods. Images at the left of the panels (A) and (B) show the Z-projections calculated from maximum intensity signals derived from all focal planes of the image stacks for vitellin (Vt, green) and DmCatD or tyrosine phosphatase (both red). Donor images (Vt signal) pre- and post-photobleaching of the corresponding acceptors, DmCatD and tyrosine phosphatase, show FRET. The white boxes indicate the pre- and post-bleaching areas. Each of the arrows point out the increase in the fluorescence intensity of Vt. A gray scale image pseudocolored with the color gradient is shown within each image to enhance the visualization of the increase in the fluorescence intenstity. Similar results were obtained in 3 separate experiments. Bars: 5 μm for image on the left panel (B) and 2 μm for the other pictures.

https://doi.org/10.1371/journal.pone.0130144.g007

Involvement of DmCatD and acid phosphatase in the proteolysis of vitellin

In D. maxima, high activity levels of acid hydrolases were detected in homogenates of atretic ovaries. In addition, the process of follicular atresia in this species was associated to vitellin degradation [27]. Therefore, in vitro proteolysis assays using purified vitellin and homogenates of ovaries at early atresia as the enzyme source were conducted to establish the participation of DmCatD and acid phosphatase in vitellin degradation.

The profile of proteolytic fragments of the main vitellin subunits, visualized by western blot, demonstrated that vitellin degradation was pH-dependent, and that extensive proteolysis occurred at pHs ranging from 3.0 to 4.0 (Fig 8A). It was also shown that negligible amounts of vitellin were present in ovarian homogenates (Fig 8A, lane 2).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 8. pH dependence and effect of DmCatD and acid phosphatase inhibitors on vitellin (Vt) proteolysis.

Homogenates from atretic ovaries were used as the enzyme source and purified Vt as substrate. (A), left panel, purified Vt and homogenates of ovarian tissues without incubation (lane 1 and 2, respectively). Right panel, in vitro pH dependence of Vt proteolysis. The reaction was carried out for 12 h at 37°C in different reaction buffers, according to the pH evaluated (3.0 to 7.0). In all cases, Vt degradation was evidenced by western blot using an anti-Vt antibody, after fractionation of proteins by 7.5% SDS-PAGE. (B), Vt and ovarian homogenates were incubated for 12 h at 37°C in a reaction medium at pH 4.0 either in the absence (lane 3) or in the presence of pesptatin A (inhibitor of aspartic peptidases, lane 4), Na₃VO₄ (inhibitor of tyrosine phosphatases, lane 5) or Na⁺/K⁺ tartrate (a broad spectrum phosphatase inhibitor, lane 6). Purified Vt and ovarian homogenate (lanes 1–2, respectively) were incubated in the same conditions and showed as controls. The arrows in the panels indicate the main subunits of purified Vt (Mr ~170 kDa and 174 kDa) visualized as a single immunoreactive band.

https://doi.org/10.1371/journal.pone.0130144.g008

When the in vitro assays were performed at pH 4.0, which is nearly the optimum pH for the activity of DmCatD and acid phosphatase, vitellin was widely degraded and few proteolytic fragments were observed as faint bands in the western blot (Fig 8B, lane 3). However, vitellin proteolysis was strongly inhibited in the presence of pepstatin A (an inhibitor of aspartic peptidases), Na₃VO₄ (an inhibitor of tyrosine phosphatase) and Na⁺/K⁺ tartrate (a broad spectrum phosphatase inhibitor) (Fig 8B, lanes 4–6, respectively), showing similar patterns to that of purified vitellin subjected to the conditions of incubation (12 h, 37°C) (Fig 8B, lane 1). NaF, an inhibitor of serine/threonine phosphatase had no effect in preventing vitellin proteolysis. Taken together, these results showed that, at least under our in vitro conditions, DmCatD and an acid phosphatase, most likely a tyrosine phosphatase, are required to promote vitellin proteolysis.