CARBONIC ANHYDRASE ACTIVITY IN THE VAS DEFERENS OF ...

JOURNAL OF PHYSIOLOGY AND PHARMACOLOGY 2006, 57, Supp 8, 107–123 www.jpp.krakow.pl

J. KOTWICA, M.A. CIUK, E. JOACHIMIAK, S. ROWINSKI, B. CYMBOROWSKI, P. BEBAS*

CARBONIC ANHYDRASE ACTIVITY IN THE VAS DEFERENS OF THE COTTON LEAFWORM – SPODOPTERA LITTORALIS (LEPIDOPTERA: NOCTUIDAE) CONTROLLED BY CIRCADIAN CLOCK Department of Animal Physiology, Zoological Institute, Warsaw University, Warsaw, Poland The male reproductive tract of Lepidoptera is an ideal model for the study of the physiological role of peripheral clocks in insects. The latter are significant in the generation and coordination of rhythmic phenomena which facilitate the initial stages of sperm capacitation. This process requires the maintenance of pH in the upper vas deferens (UVD) aided by, among others, H+-ATPase. Our aim was to determine the potential involvement of carbonic anhydrase (CA) in this process, an enzyme tasked with generating protons subsequently utilized by H+-ATPase to acidify the UVD milieu in S. littoralis, during the time when the lumen of this organ is filled with sperm. We attempted to answer the question whether CA activity can be controlled by the biological oscillator present in the male reproductive tract of the cotton leafworm. Using PAGE zymography, the presence of CA was demonstrated in the UVD wall, but not in the luminal fluid nor in the sperm. Using histochemistry, it was shown that CA is active in the UVD epithelium, and that this activity varies throughout the day and is most likely controlled by an endogenous biological clock. Conversely, the application of CA inhibitors, acetazolamide and sodium thiocyanate, in conjunction with an analysis of H+-ATPase activity in the acidification the UVD environment shows that CA most likely does not play a direct role in the regulation of the pH in this organ. K e y w o r d s : carbonic anhydrase, circadian rhythm, biological clock, sperm maturation, insect reproduction

INTRODUCTION

Insects are a model organism used to study both the molecular underpinnings of circadian clock function, as well as physiological processes under clock

108 control (1 - 7). The unprecedented progress made in this area was made primarily using Drosophila melanogaster (8 - 11). The research performed on this species was the basis for the formulation of the molecular model of the cellular oscillator (8). The latter is composed of transcription-regulating proteins synthesized as a result of clock gene activity. Their rhythmic expression has been observed in cells of the CNS (comprising centers responsible for the coordination of behavioral rhythms) as well as in various other cells in the digestive, excretory and reproductive tract tissues (12 - 15). The expression of clock genes in tissues not directly connected with the nervous system is the basis for the multi-oscillator model (7, 16, 17). The existence of non-neural oscillators, the so-called peripheral oscillators, has also been documented in vertebrate model species such as the zebrafish, African clawed frog, mouse and rat (18 - 21). It was also determined that they are characterized by a large degree of autonomy, as evidenced by their ability to maintain rhythmic molecular oscillator gene expression in vitro, and thereby independently of the CNS oscillator. It was finally determined that cultured murine fibroblasts also maintain the rhythmic expression of all known clock genes and its profile is analogous to that observed in the suprachiasmatic nucleus, the location of the mammalian master clock (22). Despite the great interest in the roles of the peripheral clocks, there still is a relatively small amount of information on the characteristics of their rhythmic outputs. One of the more intensively studied clocks is the oscillator located in the tissues of the male reproductive tract of Lepidoptera. It generates the rhythm of the release of differentiated sperm cell clones (forming bundles) from the testes to the upper vas deferens (UVD) as well as their transport along the seminal tract. This phenomenon was first observed in Ephestia kuehniella and subsequently studied further in the pioneering research of J. M. Giebultowicz on Lymantria dispar, which proved the full autonomy of the peripheral oscillator generating the circadian rhythm of sperm bundle release from the testes of male reproductive systems cultured in vitro (23, 24). The ubiquity of this phenomenon was confirmed by her team and collaborating laboratories in other species: Homeosoma electellum, Manduca sexta, Lymantria monacha, Cydia pomonella and most recently Spodoptera littoralis (16, 24 - 29). The parameters of the rhythm of sperm bundle release are very similar in all these species. The process begins in the evening and lasts through the first half of the night. During this time the spermatozoa accumulate in the UVD, where they remain until the end of the night, at which point they are transferred to the seminal vesicles. The oscillator coordinating this process has been localized in the cells which form the base of the testis and those lining the UVD. This was achieved via the expression of one of the core clock genes period (per) as well as the distribution of its product, the PER protein (15). Disruption of this oscillator activity through culturing these insects in constant light disrupts the rhythm of sperm release even to the point of inhibition (27). This, in turn, results in decreased male fertility or sterility, which is plausibly explained by the lack of coordination between the presence of sperm

109 in the UVD and the secretory activity of the epithelial cells lining this organ (30 - 34). These cells synthesize and secrete nutrients, enzymes and a series of proteins essential to the early stages of posttesticular sperm maturation at specific time points throughout the day (30, 32, 35, 36). Our research on S. littoralis shows that UVD epithelial cells also influence the maintenance of the acid/base equilibrium of the environment in which the early stages of sperm maturation take place (37). The protein identified as responsible for the regulation of seminal fluid pH in the UVD is vacuolar H+-ATPase (VATPase). The circadian clock located in the male reproductive tract generates a rhythm of luminal pH through the regulation of the concentration and subcellular distribution of this proton pump. The fact that V-ATPase participates in the acidification of seminal fluid in S. littoralis allowed us to formulate a hypothesis regarding the analogy of this process to early sperm maturation between insects and mammals. In the latter group, it was shown that acidification of the luminal fluid in the epididymis is significant for sperm maturation as well as for keeping sperm immotile throughout their movement along the epididymis and the vas deferens (38, 39). V-ATPase is directly involved in this process along with carbonic anhydrase (CA) (40, 41). It still remains to be seen whether either one of these enzymes is regulated by a peripheral circadian clock in mammals. The existence of the latter is postulated on the basis of the expression of clock genes in the cells of some reproductive tract tissues. Due to the large number of unknowns pertaining to the peripheral clock output pathways regulating reproductive processes in insects, we undertook further analyses of the circadian regulation of protein activities in the UVD epithelium. Assuming that further analogies exist between insects and mammals in the maintenance of pH levels in the male reproductive tract, we concentrated on CA. Like in mammals, in the insect vas deferens it could well be an enzyme operating in conjunction with V-ATPase. Using a histochemical method specifically identifying CA activity, we analyzed the UVD of S. littoralis. We determined that this enzyme is active in its epithelial cells, meaning the same tissue in which we earlier identified V-ATPase. A detailed analysis of the CA activity profile showed that it is under circadian control. We also showed that the circadian activity of CA has no direct relationship with the acidification of the extracellular environment. MATERIALS AND METHODS

Histochemistry The research was performed on Spodoptera littoralis males cultured in LD 16:8, constant darkness (DD) and constant light (LL) as described previously (27, 31). For the latter two, the males were transferred into the new conditions a day prior to the imaginal molt. About 25 individuals were collected every 4 hrs. Following CO2 narcosis, testes and sperm ducts were

110 dissected out in 111.2mM NaCl, 33.5mM KCl, 2.7mM CaCl2, 2,9mM NaHCO3. Tissues were fixed in 4% glutaraldehyde in 0,1M PBS for 1h, and 3x0.5h in 0,1M PBS. Incubation was performed twice, each time for 12h, in two changes of 20% saccharose in PBS. All the above were performed at 4°C. The prepared tissues were placed on glass slides, embedded in cryo-mounting medium and frozen at -80°C. Histochemical sections were prepared using a Microm HM 505 N cryostat and microtome. Tissues were mounted onto the microtome at -24°C, cut into 10µm sections and affixed onto poly-L-lysine coated slides. CA was localized in tissues using the cobalt/phosphate method previously described (42). Tissues were incubated in Hansson solution, HS, (1.7mM CuSO4, 157.1mM NaHCO3, 11.6mM KH2PO4, 39mM H2SO4). Control groups additionally contained a CA inhibitor, 10-4M acetazolamide. HS was pipetted onto the slides, incubated for 30s and dried for 30s, repeated for 20 min. Next, the chromogenic reaction was performed by immersing the sections for 20 min. in 0.5% (NH4)2SO4 and rinsing in running water for 20 min. Rinsed tissues were dehydrated and imbedded in DePeX. Photomicrography was performed on a Nikon TE 200 Eclipse microscope with a Nikon DXM 1200 digital camera with Act–1 software. Densitometric analysis was performed using the OD measuring option of Adobe Photoshop 6.0CE which facilitates the quantification of the average density of a desired area. The measurements were performed on photomicrographs of the UVD epithelium in a square area of 4 cm2. In the relative scale used, maximum opacity was given 250 units and 0 units for white.

Zymography Experimental material was collected every 4h over one day. The animals were anaesthetized and testes with the UVD were dissected out. Only UVD walls were used in this procedure. These were kept frozen in protein extraction buffer (5mM Tris-HCl pH 6.8, 1% TritonX-100, 1% NP40) until needed. The material was sonificated and centrifuged for 15 min. at 18000 g. The amount of protein in the supernatant was assayed using the Bradford assay. 12%, PAGE gels without SDS (pH8.8) were used, with a 4%, pH 6.8 stacking gel. Wells were loaded 2 OD of total protein mixed with loading buffer (Laemmli LB without 2-mercaptoethanol). Following electrophoresis, the gel was rinsed for 20 min. in 10% isopropanol in 100mM Tris pH 8.2 and in 100mM Tris pH 8.2 (2 x 10min.) The gel was then incubated in 0,1% bromothymol blue in 100mM Tris pH 8.2 (1 x 30 min. at 4°C). The CA reaction was elicited by immersing the gel in CO2-saturated ddH2O. Local decreases in pH were evidenced by the formation of yellow bands. Bands were scanned in 8-bit monochrome on a scanner with transmitted light and saved in 300dpi TIFF format. To obtain mean band optical intensity densitometry was performed using the GelEval package from Frogdance Software (UK). The electrophoretic bands were selected with rectangular boxes 100x30 pixels, which encompassed whole bands. The densitometric units scored 1 for black (100% opacity) and 0 for white (0% opacity). Negative images were scanned, thus OD varied directly with enzyme activity.

pH variation in the vas deferens The procedures were performed on insects reared under LD 16:8 conditions. 30 males were collected every 4h and anaesthetized with CO2. Testes and sperm ducts were dissected out in clear Grace’s medium pH 7.0 and incubated for 0.5h so that the tissues could adapt to the new environment (the experiment made use of the UVD only). Next, the tissues were placed in culture plates with 200µl Grace’s medium. In order to verify that CA participates in the acidification of the extracellular environment, specific inhibitors of this enzyme were used: 10-4 M acetazolamide and 100mM sodium rhodate (NaSCN). The tissues were incubated in a Binder WTC chamber at LD

111 16:8, 26°C, and an atmosphere of 0.5% CO2, 30% O2, 69.5% N2. After 4h of incubation, the medium was sampled and its pH was measured using 0.250g/ml bromothymol blue (BTB) indicator. pH changes were measured colorimetrically on a Perkin Elmer Victor3 plate reader at 450nm. Medium incubated without tissues was used as a control.

Statistical analysis All data were analyzed for statistical significance by using ANOVA and Fisher’s least significant procedure (LSD intervals) at 95% confidence intervals. RESULTS

CA activity rhythm in the UVD epithelium – histochemical analysis Hansson staining of UVD tissues showed that this enzyme is active in the columnar epithelium lining this structure. The research was performed on tissues collected at various time points throughout the day and showed differences in the staining intensity of the tissues, indicating periodic variation of CA activity (Fig. 1a). Differences were also observed in the localization of staining within the cells in this tissue, which are evidence of periodic changes in the distribution of the enzyme. The CA activity peak was observed during midday at Zeitgeber time 4 (Zt4). At this time point, we also observed the accumulation of cobalt salt precipitates in the apical portion of epithelial cells (oriented towards the UVD lumen). Barely 4 hours later (Zt8), the tissues were characterized by the lowest reactivity. The staining level of the UVD cells was comparable to that of control cells treated with acetazolamide (Fig. 1d). Throughout the night, the staining intensity of the cells was moderate (Zt12 to Zt20) and stayed at the same level for half of the day (until Zt0). During this period, CA activity was observed throughout the cells without any discernible polarization which would be indicative of local accumulations of CA. These results demonstrate a clear rhythm of CA activity in this tissue. In order to ascertain whether this rhythm is under endogenous regulation, analogous procedures were performed in constant darkness. UVD tissue analysis showed that an almost identical rhythm occurs when the males were cultured for 2 days in DD (Fig. 1b). Here too, we observed a peak in enzyme activity with its characteristic accumulation in the apical portion of the cell during subjective midday, with a minimum at near-control levels just 4 hours later. After showing that the rhythm was endogenous and not just a response to altering day/night conditions, we showed that it is regulated by an endogenous oscillator. This was confirmed by exposing the insects for two days to constant light. At all time points through the day we observed comparable CA activity in the UVD (Fig. 1c). Neither did we observe the several hour long accumulation of the enzyme in the apical portion of the cells observed in LD and DD. In order to obtain maximally significant, comparable results, demonstrating actual differences in CA activities in the examined tissues, it is necessary to

112 a

b

c

d

0

4

8

12 time (hours)

16

20

Fig. 1. CA activity in the UVD epithelium identified using Hansson staining. (a) UVD epithelia of males reared under LD 16:8 conditions. A clear activity rhythm with a peak at Zt4 and an accumulation of cobalt salt precipitate indicating high enzyme activity in the apical portion of the cell, with a minimum or no activity at Zt8. Analogous images of UVD epithelia from males reared for two days in constant darkness (b) and constant light (c), which indicates the endogenous nature of the CA activity rhythm and the role of a biological oscillator in its generation; (d) the result of control reactions following treatment with acetazolamide, confirming the role of CA in the staining of the epithelium in the experimental variants. The images presented are representative of a minimum of 25 samples analyzed at each time point for each experimental variant. Scale bar, 20µm. Horizontal bars represent day (open), night (black filled) and former day (hatched) portions of the photo-regimes.

perform the Hansson staining each time under identical experimental conditions. The samples must always be processed simultaneously, over an identical period measured in seconds and the result may be influenced by even minimal differences in tissue size. Therefore, to obtain representative results, we performed the procedure on a large number of insects, no less than 25 per time point per variant. Subsequently, densitometric analyses were performed on the micrographs of the stained tissues. Finally, statistical analyses were performed of the quantified densities. The results obtained unequivocally demonstrate circadian variation of CA activity in the UVD epithelium which peak at Zt4 (Fig.

113 2a). At this time, the enzyme activity was two and a half times higher than at its minimum (Zt8) and was almost a third higher than at the remaining time points. A very close result was obtained analyzing the profile of circadian variation of CA activity in males kept for two days in constant darkness (Fig. 2b). In males kept for a period of two days in constant light on the other hand, densitometric analysis of the samples showed that CA activity levels remained at identical levels throughout the day (Fig. 2c). The specificity of this reaction was ascertained through the use of control material, tissues incubated in the presence of acetazolamide (Fig. 2 – white bars in each graph).

240

a

200 160 staining intensity of the UVD epithelium (relative units)

120 80 40

*

*

*

*

0 240

b

200 160 120 80 40

*

*

*

*

0 240 200 160 120 80 40 0

c

0

4

8 12 time (hours)

16

20

Fig. 2. CA activity rhythm in the UVD epithelium, results of the densitometric analysis of tissues stained using the cobalt method of Hansson for a minimum of 25 individuals per time point per experimental variant. The procedure was performed on males reared under LD 16:8 conditions (a), constant darkness (b) and constant light (c). The vertical axis represents relative staining intensity units from 0 to 250, where 0 represents no staining (white field), and 250 represents total staining (black field). Black bars represent the averaged results (+ SEM) for densitometric measurements of micrographs obtained of stained tissues from experimental animals, and white bars represent respective control groups treated with acetazolamide. * denote bars representing statistically significantly differing values obtained, both at maximal and minimal enzyme activity values (calculated by ANOVA and Fisher’s LSD intervals P