Tartrate resistant acid phosphatase as a marker for scale resorption in rainbow trout, Oncorhynchus mykiss: effects of estradiol-17p treatment and refeeding.
Fish Physiology and Biochemistry vol. 14 no. 4 pp 329-339 (1995) Kugler Publications, Amsterdam/New York
Tartrate resistant acid phosphatase as a marker for scale resorption in rainbow trout, Oncorhynchus mykiss: effects of estradiol-17p treatment and refeeding Petra Persson', Yasuaki Takagi, 2 and Bj6rn Thrandur Bj6rnssonl 'Fish Endocrinology Laboratory, Department of Zoophysiology, G6teborg University, Medicinaregatan18, S-413 90 G6teborg, Sweden; 2 0tsuchi Marine Research Center, Ocean Research Institute, University of Tokyo, Akahama, Otsuchi, Iwate 028-11, Japan Accepted: January 28, 1995 Keywords: calcium regulation, teleost fish, osteoclastic activity, resorption site, TRACP, ACP
Abstract In teleosts, a considerable part of the body calcium is found in the scales. Associated with the scales are osteoblasts and osteoclasts, and during periods of high calcium demand such as during sexual maturation or starvation, the scales can be resorbed and thereby act as an internal calcium reservoir. In mammalian bone tissue, the activity of an acid phosphatase (ACP) isoenzyme, tartrate resistant acid phosphatase (TRACP), can be used as a marker for osteoclastic activity. In the present study, an evaluation of TRACP as a marker for osteoclastic activity in teleost scales has been performed. ACP and TRACP was histologically localized at resorption sites around the edge of the scales as well as at resorption holes in the scales. The optimal conditions for biochemical measurements of ACP and TRACP activity were found to be pH 5.3, 10 mM paranitrophenylphosphate, incubated for 30 min at room temperature, and 10 mM tartrate added when required. Using TRACP as a marker, estradiol-170 (E 2) was found to increase the proportion of scales being resorbed, as well as the number and size of resorption sites per scale. Also, the scales of E 2-treated fish showed weaker staining for calcium. Together, the obtained data indicate that estradiol-17P induces osteoclastic activity in teleost scales, resulting in increased resorption of the scales. A period of refeeding following a period of starvation did not have detectable effects on the scale osteoclastic activity and scale resorption.
Introduction The teleost scale is a calcified tissue which contains forming cells (osteoblasts) (Yamada 1961, 1971; Ouchi et al. 1972; Bereiter-Hahn 1993) and resorptive cells (osteoclasts) (Ouchi et al. 1972; Bigler 1989; Sire et al. 1990) similar to those found in teleost, avian and mammalian bone. The scales of some teleosts can contain as much as 20% of the total body calcium (Berg 1968; Fleming 1974), and
are thus a potential internal calcium reservoir during periods of increased calcium demand, such as during sexual maturation and starvation (Berg 1968). Little is known about the regulation of osteoblast and osteoclast activity in teleost bone and scale tissue, and until recently, the involvement of multinucleated osteoclasts in teleosts bone resorption was debated. Characteristic of mammalian bone osteoclasts is the enrichment in acid phosphatases (ACP), partic-
Correspondence to: P. Persson, Department of Zoophysiology, Goteborg University, Medicinaregatan 18, S-413 90 G6teborg, Sweden
330 ularly that of tartrate resistant acid phosphatase (TRACP), an isoenzyme which is not blocked by tartrate (Vaes 1988). Therefore, TRACP activity is often used as a marker for osteoclastic activity in mammalian bone tissue (Vaes 1988). Rainbow trout (Oncorhynchus mykiss) bone osteoclasts have also been shown to contain TRACP, and the enzyme has been histologically identified at the surface of resorption lacunae in close vicinity to osteoclasts (Takagi and Kaneko 1995). During sexual maturation in female teleosts, the demand for calcium increases. This is due to an estradiol-170 (E 2)-induced increase in plasma vitellogenin (VTG) and VTG-bound calcium levels. The VTG is transported to the oocytes where it is taken up and stored as yolk. Vitellogenesis can also be induced by injecting juvenile fish with E2, and such studies on E 2-treated juvenile rainbow trout show an increase in the uptake of calcium from water (Persson et al. 1994) as well as an increase in the calcium mobilization from scales (Persson et al. 1994; Carragher and Sumpter 1991), whereas calcium excretion and mobilization from bone are not affected (Persson et al. 1994; Carragher and Sumpter 1991). Calcium mobilization from scales, but not from bones, were also reported in E2 -treated goldfish, Carassiusauratus, and killifish, Fundulus heteroclitus (Mugiya and Watabe 1977). Similarly, calcium was mobilized from scales when female tilapia (Tilapiaesculenta) reached sexual maturity (Garrod and Newell 1958) and scales have been seen to be resorbed when both female and male masu salmon reach sexual maturity (Ouchi et al. 1972; Takagi 1990). Another condition which can cause scale resorption in teleosts is starvation (Yamada 1956; Ikeda et al. 1974), but no studies concerning the mechanisms and endocrine regulation behind this have been carried out. The present study tests the hypothesis that estradiol-17[ induces scale resorption by activating osteoclasts directly or indirectly. In order to do this, an evaluation of TRACP as a marker for osteoclasts in teleost scale tissue was performed. Also, the effect of refeeding on scale osteoclastic activity was investigated.
Material and methods Fishand generalproceduresforA CP-measurement Juvenile rainbow trout (38 + 2 g) were obtained from a local hatchery, Antens laxodling AB. The fish were acclimated in aerated, recirculating fresh water ([Ca] 0.5 mM) at 10-12°C for at least 5 days prior to experiments. Prior to the experiments, the fish were fed 1.5% body weight day-' (EWOS, ET90 Vextra 2.0P). In the course of the experiments, the fish were killed either by a blow to the head, or by anaesthezia (tricaine methanesulphonate, 100 mg 1-1 water; buffered to pH 7.0 with NaHCO 3). The fish were blotted dry and the mucus scraped off with a scalpel. The scales were then removed using a scalpel, homogenized in a 0.1 M sodium acetate (NaAc) buffer (pH 5.0; 50 mg scales ml-1 buffer) using a glass-glass motorized homogenizer. The scales were kept on ice throughout, except during the enzyme assays which were conducted at 22°C. ACP and TRACP activity measurements were conducted by adding 50 ll of scale homogenate to 350 1lof 10 mM para-nitrophenylphosphate (pNPP) in a NaAc buffer. When measuring the TRACP activity, 20 mM tartrate was included in the pNPP solution. After incubation, the reaction was stopped by adding 200 gl of 2 M NaOH, and the solution was then centrifuged for 5 min at 3600 x g. The absorbance of the supernatant (250 1l) was measured in duplicate at 405 nm on a plate reader (EL 340, Bio Kinetics Reader, BIO-TEK Instruments) using 96-well microplates. The absorbance was converted into the amount of produced para-nitrophenol (pNP), using a standard curve for pNP.
Effects of pH The pH optima of ACP and TRACP were determined using five fish. The scales were homogenized in 0.1 M NaAc buffer (pH 5.0, 50 mg scales ml- ' buffer). The enzyme activities were measured using 10 mM pNPP in NaAc buffer with pH adjusted to
331 5.1, 5.2, 5.3, 5.4 and 5.5, respectively. The incubation time was 30 min at 220C. When TRACP activity was measured the pNPP solution contained 20 mM tartrate.
pNPP saturation and tartrateinhibition curves Scales from three fish were homogenized as above (50 mg scales ml- l buffer) and the pH of the pNPP solution was adjusted to 5.3. The pNPP concentrations tested were 0, 1.25, 2.5, 5, 10 and 20 mM. These substrate concentrations were tested in the presence of 0, 10, 20 and 40 mM tartrate. The incubation time was 30 min at 220C. Based on the results of this initial tartrate inhibition study, a second tartrate inhibition experiment was performed using 1.25, 2.5, 5, 10 and 20 mM tartrate. Scales from three fish were homogenized as above (40 mg scales ml-1 buffer), and the assay conditions were 10 mM pNPP, pH 5.3 and 30 min incubation time at 22°C.
Incubation time In order to evaluate the incubation time, scales from three fish were sampled and homogenized as above. The pH of the 10 mM pNPP solution was adjusted to 5.3. When TRACP activity was measured, the pNPP solution contained 20 mM tartrate. The incubation times used were 10, 20, 30, 60, 120, 180 and 240 min at 22 0C.
Effects of temperature The temperature dependency of ACP and TRACP activities was determined using three fish. The scales were homogenized as above (40 mg scales ml- buffer). The assay conditions were 10 mM pNPP solution (pH 5.3), 30 min incubation time, and when required, 20 mM tartrate was included.
Intra- and interassay variation Scales from 3 fish were collected, pooled, divided into five batches, and treated as described above. Five assays were performed, and in each, the ACP and TRACP activity of the five batches were determined. The assay conditions were 10 mM pNPP solution (pH 5.3), 30 min incubation time at 22°C and when required, 20 mM tartrate was added.
Scale resorption In order to check whether TRACP was found only at resorption sites, scales were stained for ACP and TRACP using the Sigma acid phosphatase kit (Procedure No. 387). The scales were sampled as described above and fixed in 10% formalin in 0.05 M cacodylate buffer or acetone. Prior to staining, the scales were rinsed in deionized water. After staining for lh, the scales were rinsed in tap water and treated with glycerol for 5 min before being mounted on micro slides. When staining for TRACP, 20 mM tartrate was added to the staining solution.
Linearity of the assay The linearity of the assay was determined using scales from three fish. The scales were sampled and treated as above. The scale homogenates (50 mg scales ml-') were diluted to 40, 25, 12.5 and 6.25 mg scales ml-1 with NaAc buffer (pH 5.0) and ACP and TRACP activities were measured. The assay conditions were 10 mM pNPP solution (pH 5.3), 30 min incubation time at 220 C, and when required, 20 mM tartrate was included.
Calcium mobilization In order to check whether calcium had been mobilized at the resorption sites, scales were stained for calcium using the alizarin red S method (Dahl 1952). The scales were sampled as mentioned above and fixed in 10% formalin in 0.05 M cacodylate buffer, pH 7.4. After rinsing the scales in deionized water, they were stained in alizarin red S for 2-3
332
100
0
ACP
0
1
98
.5
96
4-
co le
94
92
90
5.0
·
·
·
5.1
5.2
5.3
5.4
I
I
5.5
5.6
pH Fig. 1. Effects of pH on the activity of acid phosphatase (ACP) and tartrate resistant acid phosphatase (TRACP). Data are presented as means SEM (n = 5).
min. The scales were then treated with acetone and glycerol before being mounted.
Scaleformation Scales were stained for alkaline phosphatase (ALP) in an attempt to identify osteoblastic activity. Scales were sampled as described earlier, and fixed in 10% formalin in 0.05 M cacodylate buffer at pH 7.4, and stained using a 0.05 M 2-amino-2methyl-1,3-propanediol solution (Sigma) containing 1 M naphtol AS BI phosphate sodium salt (Sigma) and 1.3 M fast blue RR (Sigma). Rainbow trout intestine was used as a positive control for this method (Johnston et al. 1994).
ed controls (n=6). The E2 -treated fish received two additional E2 -injections (5 mg kg-', 2 ml kg-1), 5 and 10 days after the initial injection. The starved fish were killed 15 days after the first E 2-injection, whereas the fed fish were killed 23 days after onset of refeeding. Blood was collected from the caudal vessels into heparinized syringes. It was immediately centrifuged to obtain plasma, and total plasma calcium was analyzed spectrophotometrically by calcium arsenazo III (Sigma; procedure no. 588) using a plate reader. All scales from the left side of the body were sampled, thoroughly mixed, and ACP and TRACP activities were measured. The tartrate sensitive acid phosphatase (TSACP) was assessed by subtracting TRACP from ACP. Also, all scales from the right side of the body were sampled, thoroughly mixed, and fixed in 10% formalin in 0.05 M cacodylate buffer at pH 7.4. After fixation, the scales were stored in 0.05 M cacodylate buffer at pH 7.4. Scales from all fish were stained for ACP, TRACP and calcium as described above. The livers were removed and weighed to calculate the liver somatic index (liver weight x bwt-l)100. Statistical analysis First order regression was performed when assessing the incubation time, temperature dependency and linearity of the ACP and TRACP assay. Differences between untreated starved fish, E2 -treated starved fish and fed fish were assessed using a oneway ANOVA followed by Dunnett's test. Statistical significance was set at p
