ELSEVIER PI1 SOO24-3205(97)00437-2

Life Sciences, Vol. 61, NO. 6, pp. 623429, 1997 cbpyright 0 1997 Ekvier scienec Inc.

Printed in the USA. All rights reserved omAnasp $17.00 t .al

PRESENCE OF A HUMAN-LIKE THYROID STIMULATING

HORMONE (TSH) IN CIONA INTESTINALIS

M. Maddalena Di Fiore o *, Lorena Perrone O and Antimo D’Aniello O

o Department of Biochemistry and Molecular Biology, Stazione Zoologica “A. Dohrn” Vii Comunale, 80121 Napoli, Italy

* Institute of Biology, Second University of Naples, Via Arena, 81100 Caserta, Italy

(Received in fmd form May 12,1997)

Summary

Using a monoclonal antiiody against human Thyroid Stimulating Hormone (TSI-I), we have found that the invertebrate Ciona intestinalis (phyhun Chordata, subphyhun Tunicata) contains a previously unreported protein which is imnnmoreactive for anti-human TSH. The amount of this hormone in the blood, endostyle and ovary of C. intestinalis was found to be 0.01*0.003, 1.05kO.2 and 3.6W.25 uIU/g of tissue, respectively, using the RIA method, and a value of 0.013~0.0043, l.leO.30 and 3.85kl.32 $IJ/g using an immuno-chemihumnescent method. In addition to possessing Wological properties, this protein is able to induce the synthesis of CAMP in rat thyroid cell culture (FRTLS) and in Chinese Hamster Ovary cells (CHO) transfected with the cDNA for hmnan TSH receptor. This indicates that the putative ciona TSH has the capability to react specifically with receptors for m TSH. Maximum concentrations of ciona TSH-like factor occur during periods of sexual maturity (between May to July), whereas very low concentrations were assayed during the rest of the year suggesting that this hormone may be involved in hormonal fttnction related to sexual maturity. From a phylogenetic point of view, the above data supports the hypothesis for a common origin of a thyroid hormonal system between Tunicata and vertebrates.

Key Words: thyroid stimulating hormone, CAMP, Ciona intestinalis, Tunicata

Ciona intestinalis is an ascidian species belonging to the subphylum of Tunicata, phyhun Chordata. On the basis of the morphology at larval stage, the Tunicata represent the most primitive vertebrate (1). A region of the pharynx, the endostyle, has been postulated to be the homologue of the vertebrate thyroid gland (2). In fact, a sob population of cells in the endostyle of Ciona intestinalis, accumulate iodine and also contain thyroid hormones T3 and T4 (3-11) as well as the thyroid hormone precursor, thyroglobubn (12). In addition, evidence for the existence for the triiodothyroniue (T3) receptor in the pharynx and alimentary canal has been reported (13).

Corresponding author: Dr. Antimo D’Aniello, Department of Biochemistry and Molecular Biology, St&one Zoologica “A. Dohrn”. Villa Comunale, 80121 Naples, Italy. Tel. xx 39 81 5833-249 Fax: xx 39 81 7641-355; E-mail: daniello@alpha.sm.it

624 TSH in Tunicata Ciona intestinalis Vol. 61, No. 6, 1997

ln the present work, using a mouse monoclonal antibody against human Thyroid Stimulating Hormone (TSH; Thyrotropin), we have found that C. intestinalis possesses a protein which is immunologically similar to human TSH. ln addition, like mammalian TSH, this protein is capable of stimulating tbe synthesis cylic AMP (CAMP) in rat thyroid cells of the cellular line FRTL-5 (14) and in Chinese Hamster Ovary cells (CHO) transfected with the cDNA for human TSH receptor (15, 16).

Materials and Methods

Preparation of crude extracts. Aduh C’io71u irztestinalis were collected from the Bay of Naples between October and June. The blood was collected from the heart and centrifuged at 30,000 g for 60 min. The supematant was concentrated 20-fold on an Amicon membrane concentrator with a cut-off of 10,000 Dalton (Amicon PM IO membrane). The solid tissues: endostyle, ovary, branchial sac and tunic, were homogenized in a ratio of 1:5 with phosphate buffer saline (PBS) (NaCl, 8 g/l; KCI, 0.2 g/l; Na2HP04, I .44 g/l and KH2PO4, 0.24 g/l; pH 7.4) containing a cocktail of protease inhibitors consisting of IO mM EDTA and Ieupeptin, pepstatin and aprotinin each at the concentration of 10 &ml. The homogenates were then centrifuged as above and the supematant was concentrated 5 fold using the same membrane concentrator as above.

Determination of the human-like TSH. The determination of the TSH in the crude extract or in the purified &actions of C’. intestinalis was performed by an immunological method using a monoclonal antibody against human TSH. Two reagent kits were used for the determination of the TSH in human blood: a radio immunoassay (RL4) by ICN Pharmaceuticals, Inc., Diagnostic Division, Costa Mesa. CA 92626, USA, and a chemiltiescent immunometric assay (lmmulite third generation TSH, solid phase, two-site) by Diagnostic Products Corporation, Los Angeles, CA 90045-5597, USA.

Induction of the synthesis of cyclic AMP (CAMP) in rat thyroid cells. This experiment was carried out using cultured rat thyroid cells, FRm5 cell line, grown routinely at 37” C in a chemically defined Coon’s modified Ham’s F- 12 medium (Sigma, Chemical Co.) supplemented with 5% newborn calf serum (heat inactivated) including the following six hormones: insulin, 10 l&ml; hydrocortisone, 1 umol/ml; transferrln, 5 ug/ml; glycil-L-hystidinsl-lysine acetate, 10 @ml; somatostatin, 10 &ml and thyrotropin (TSH), 100 uU/ml according the method of Ambesi- lmpiombato et al. (14). The cells were seeded for 4 days in a 12-well plate (Costar, Cambridge, MA) using 500,000 FRTL-5 cells/well and the above medium. After 3 days, the cells were switched to a similar medium, but deprived of the TSH hormone and maintained in this medium for 3 days. The medium was than aspirated and 2 ml of a new medium was added to each well. Then, 50 ul (2 mg/ml) of 3-isobutyl-l-methylxanthile (IBMX) (an inhibitor of phosphodiesterases) and 100 ul of the sample were added to each well, mixed and incubated at 37” C for 2 hours in a 5% CO2 I 95% air atmosphere mixing every 20-30 min. After incubation, the cells were washed two times with 2 ml of PBS (previously warmed to 37” C) and mixed. Finally, 1 ml of 0.25 M perchlotic acid (PCA) was added to each well, mixed and left overnight at 2-6” C. The cells were sonicated (20 cycles for 30 set) and centrifuged in Eppendorfmicrotubes for 5 min at 15,000 x g.

The supematant was neutralized (pH 6-8) with 2.5 M KOH (using 180 pl before and than 5 ul each time). The samples were cooled at 0” C for 30 min to obtain the maximum insolubilization of potassium perchlorate and then centrifuged for 5 min at 20,000 x g. The supematants were speed vacuum dried, and the residues were dissolved in 100 u1 of 0.05 M Tris-HCl buffer containing 4 mM EDTA, pH 7.5. These samples were again cooled at 0’ C and re centrifuged as above to precipitate the residue of potassium perchlorate if still present. Finally, the later supematants were used for the radio immunoassay determination of the CAMP.

Vol. 61, No. 6, 1997 TSH in Tunicata Ciona intestinalis 625

Znduction of CAMP in ovary cells transfected with the gene of the human TSH receptors. This

experiment was carried to know if ciona TSH in addition to the above activity it also possesses the capability to react with receptor for mammalian TSR Chinese Hamster Ovary cells (CHO) transfected with the recombinant human TSH receptor (CHG-R cells), clone p-09 according to the method of Chiovato et al. (16), were used in this study. CHO-R cells were cultured in Petri dishes in RPMI 1640 medium supplemented with 1 mmol/L glutamine, 10% fetal calf serum and 0.4 g/l gentamicin. CHO-R cells were harvested using trypsin-EGTA mixture and seeded (300,000 cells/well) in a 12 well plates (Costar, Cambridge, MA.). For the determination of CAMP the procedure described above was followed.

Gelfiltration on Sephadex G-ZOO of ovary homogenate Two ml of concentrated supematant of the ovary homogenate were applied to a Sephadex G-100 superfine column (2.4 x 40 cm) previously equilibrated in isotonic PBS. The column was ehrted with the same buffer at a flow rate of 1 mllmin collected in 5 ml fractions. Each fraction was concentrated to 0.5 ml using Amicon membrane concentrator and analyzed for protein concentration (O.D. 280 nm), TSH-like immunoreactivity and for the biological activity. The molecular weight was determined by gel 8hration according to the method of Am-hews (17). The column was calibrated using a mixture of the following purified proteins of di&rent molecular weights: phosphorylase b (rabbit muscle) 97 kDa; bovine serum albumin, 66.2 kDa; trypsin inhibitor (soybean), 20.1 kDa and cytochrome c (horse heart), 12.5 kDa. The molecular weights were plotted against the elution vohme, and determined by interpolation on this plot.

Results and Discussion

Immunological activity. This was determined using two immunological methods based on a monoclonal antibody anti-human TSH: a radio immunoassay method @IA) and a chemiltiescent method. Both methods gave positive results for the presence of the human-like TSH in blood, endostyle and ovary of C. intestinalis. The RL4 method gave a value of 0.0 1 @U/ml of blood and 1.05 and 3.6 1 uIU/g of endostyle and ovary respectively. The chemihtminescent method gave more or less the same results: 0.013 @J/ml of blood and 1.16 and 3.85 uIU/g of endostyle and ovary respectively (Table 1). We did not find any appreciable amounts in the branchial sac, pharynx, intestine and tunic. The ovary contained a higher concentration of putative TSH than the endostyle, but the endostyle possesses a higher specific activity expressed as the value of pBJ of TSWmg of protein extracted. To determine if the immunoreactivity for human-like TSH found in C. intestinalis tissues was due to interference from other compounds contained in the crude extract, a portion of the ovary extract was purified on a Sephadex G- 100 column (Fig. 1) and each fraction was analyzed for both immunological and biological activities. The resuhs demonstrated that both activities were found only in one protein peak corresponding to a molecular weight of about 30 kDa, similar to mammalian TSH (28 kDa). The fractions collected before and after this peak did not show any significant immunological and biological activity demonstrating that the immnnopositivity occurring in the ovary was due to a specific molecule that reacted with the antibody against the human TSH (Fig. 1).

Biological activity. In vitro experiments were also conducted on cultured thyroid cells to test for an increase of the intracellular synthesis of CAMP induced by ciona putative TSH. The results indicated that both endostyle and ovary extracts had this activity. In fact, when a known amount of a crude extract of endostyle or ovary were incubated for 2 hours with 500,000 FRTL-5 cells incubated in 2 ml medium, the intracellular concentration of CAMP increased from a basal v&e of 2.5 pmol to 5.5 pmol (increased 2.2, ~~0.05) in the presence ofthe endostyle extract and to a value of 11.4 pmol (increased 4.6 fold; pcO.01) for the ovary extract (Table 2). This experiment was also

626 TSH in Tunicata Ciona intestinalis Vol. 61, No. 6, 1997

TABLE 1 Human-like Thyroid Stimulating Hormone (TSH) from Cionu intestinalis

Radio immunoassay Chemiluminescent Protein Specific RIA method immunoassay (mg/W activity

uKJ/g wet tissue _ pIU/mg protein

0-v 3.61+1.25 3.85kl.32 3 1.255.3 0.11 Endostyle 1.05*0.20 1.1&0.30 5.lk1.05 0.20 Blood 0.01*0.003 0.013+0.004 0.2kO.05 0.05 Branchial sac co.01 <O.Ol Pharynx co.01 co.01 Intestine co.0 1 co.0 1 Tunic <O.OI co.01

The results represent the mean f S.D. from 5 individual experiments. Each experiment was performed on a pool of tissue from 5 animals.

0.5

0 1

o---o O.D. - TSH A-A CAMP

I 1 40 60 120 160

0 .30 I I 12

0 .25 10

i 0 .20

1 0.05

_I 0

2 1 0

Elution volume (ml) FIG. 1

Chromatographic gel filtration on Sephadex G-100 superfine of a crude extract of ovary of Cionn intestinalis. Two ml of the supematant from a homogenate concentrated 5-fold (proteins from 2 g of tissue) were applied on a 2.4 x 40 cm column equilibrated in PBS, pH 7.4. The column was eluted with the same buffer at a flow rate of 1 ml/min. Fractions of 5 ml were collected and on each fraction were determined the optical absorbance at 280 nm, TSH activity, and CAMP induction. O-O, O.D. at 280 nm; O----O, TSH activity; A -A, CAMP induction in thyroid cells. Insert: Plot of molecular weight of marker proteins against e&ion volume. Molecular wtights were determined by interpolation on this plot: 1. Phosphorylase b (rabbit muscle) 97 kDa; 2. Bovine serum albumin, 66.2 kDa; 3. Trypsin inhibitor (soybean), 20.1 kDa and 4. Cytochrome c (horse heart), 12.5 kDa.

Vol. 61, No. 6. 1997 TSH in Tnnicata Ciona intestinalis 627

TABLE 2 Content of CAMP in FRTL-5 cells, in Chinese Hamster Ovary cells (CHG-cells) and in CHG-cells transfected with the cDNA for human TSH receptors (CHO-R cells) following treatment with the

putative TSH extracted Born Ciona intestinalis tissues

FRTLS cells CHO cells CAMP Specitic cA.MP Specific pm01 a&vi+ pm01 activityd

CHO-R cells CAMP Specific pm01 activityd

Basal levelsa 2.5*0.3* - 2.9*0.5* - 2.8*0.4* -

Endostyleb (Crude extract)

5.5*0.9* 1’1.0 3.6*0.8* 1.2 6.3*0.9* 6.6

Gvaryb 11.4*1.8* 3.7 3.*0.6* 1.3 10.5*1.4* 4.4 (crude extract)

fiaryc 8.3&1.4* 27.6 3.5*0.6* 1.7 7.5*1.3* 25.0 (purified fraction)

The values represent the mean f SD obtained from 5 individual experiments. a Basal levels of the CAMP umtained in 500,000 cells. b Levels of CAMP contained in 500,000 cells induced by the action of the putative TSH extracted from 100 mg of each tissue of Ciona intestidis. c Values obtained using the purified fraction of the &ma putative TSH at the MW of 30 kDa obtained from the G-100 Sephadex column (see Fig. 1) of a supernatant of an homogenate of 100 mg of ovary. d The specific activity represents the ratio between the amollnt of total CAMP contained in 500,000 cells and the amount of the proteins (mg) extracted from 100 mg of tissue. Statist.ical analyses were performed using unpaired Student’s test. *p<O.Ol

G z p. ;;: 5 5- 5

0, I I

0 0.25 0.50

putative TSH (ulU) . . ~ FIG. 2

Relationship between the amount of C. intestinalis putative TSH and the CAMP synthesis in rat thyroid FRTL-5 cells. The values are mean f S.D. (~5 ) of the increase in the synthesis of CAMP from 500,000 cells induced by various amount of putative TSH obtained from the purified ovary &actions on Sephadex G- 100.

628 TSH in Tunicata Ciona inleslinalis Vol. 61, No. 6, 1997

carried out using the purified putative TSH fkaction obtained from the Sephadex G-100 column. Interestingly, we observed that the fractions purified from the G-100 Sephadex contained the highest TSH immunoactivity (Fig. l), and also the highest specik activity in the induction of the synthesis of CAMP (27.6 pmol of CAMP for 500.000 cells/total amount of protein contained in the purified fkaction).

Surprisingly, this also occurred in the induction of the synthesis of CAMP in the CHO-R cells tranfected with the cDNA for human receptor. In fact. as shown in table 2, crude extracts of endostyle and ovary increased the level of intracellular CAMP in these cells f?om the basal value of 2.8 pmol to 6.3 pmol (an increase of 2.25 fold over the basal value, p<O.Ol) and from 2.8 to 10.5 pmol (an increase of 3.75 fold over the basal value, pcO.0 1) respectively for the endostyle and ovary extracts (Table 2). This result indicates that the ciona putative TSH speciftcally is recognized by the receptors for the mammalian TSH. Finally, also in this last experiment it was observed that the purified putative TSH kaction from the ovary, obtained on G-100 Sephadex column, had a significantly higher value of specific activity in inducing the synthesis of CAMP compared to the crude extract. A value of 25 was obtained for the purified e-action. against a value of 3.4 obtained for the crude extract.

Finally, a dose response between putative TSH and the increase in the levels of the synthesis of CAM? exists in rat thyroid cells. In fact, as is shown in Fig. 2, when the thyroid cells FRTL-5 were incubated with various amounts of ciona putative TSH (0 to 0.5 pIU), a linear increase in the level of CAMP occurred.

Relationship between the occurrence of ciona putative TSH and sexual maturity. We found that the occurrence of the mammalian-like TSH in C. intestinalis was higher during the months of May to July, corresponding to the period in which ciona reaches maximum sexual maturity (in this period the average concentration of putative TSH in the blood is 3.7 f 1.3 pIU/ml). On the contrary in the months between October to March, the amount of putative TSH is very low (0.5 + 0.06 pIU/ml). This data indicates that ciona putative TSH could be implicated in some hormonal functions, related to sexual maturity.

Conclusion

In previous studies (3- 12), the endostyle of C. rr~testinalis has been related to the mammalian thyroid due to the discovery of the hormones T3, T4 and thyroglobulin in C. intestinalis (13). The present tiding of a pituitary-like hormone in ciona endostyle and ovary provides, along with earlier studies, further evidence for a similar hormonal activity between Tunicata and Vertebrates.

Acknowledgments

We are gratemy to Prof. Roberto Di Lauro f?om the Zoological Station of Naples (Italy) for his suggestions and for available FRTL-5 thyroid cell line. We also are gratefklly to Prof. Luca Chiovato of the Institute of Endocrinology, University of Pisa (Italy) and his collaborators for the suggestions of the experiments in the use of the Chinese Hamster Ovary cells transfected with the recombinant human TSH receptor CHO-R cells, clone JP-09.

References

I. E. J. W. BARRINGTON, The Biology of the Emichordata and Protocordata, Oliver and Boyd, Edinburgh (1965).

2. R. GUILLEMIN, Science 202 390-402 ( 1978). 3. G. SALVATORE, G. VECCHIO and V. MACCHIA, Comp. Rend. Sot. Biol. Paris 154,

Vol. 61, No. 6, 1997 TSH in Tunicata Ciona intestinalis 629

1380 (1960). 4. J. ROCHE, G. SALVATORE and G. RAMETTA, Biochem. Biophys. Acta 63 154-165

(1962). 5. E.J.W. BARRINGTON and A. THORPE, Gen. Camp. Endocrinol 3 373-385 (1965). 6. G. SALVATORE, Gen. Comp. Endocrinol 2 535-552 (1969). 7. S. SUZUKI and Y. KONDO, Gen. Comp. EndocrinoL u 402-406 (1971). 8. H. FUJITA and H. NANBA, Z. Zellforsch m,455-469 (1971). 9. A. DUNN, J. Exp. Zool. 188 103-124 (1974). 10. A. DUNN, Gen. Comp. Endocrinol 40 473-483 (1980). 11. A. DUNN, Gen. Comp. Endocrinol 40 484-493 (1980). 12. M.C. THORNDYKE, Nature 27161-62 (1978). 13. J.M.G. FREDRIKSSON, LEBEL and J.D LELOUP, Gen. Comp. Endocrinol 92 379-387

(1993). 14. F.S. AMBESI IMPIOMBATO, L.A. PARKS and H.G. COON, Proc. Natl. Acad. Sci. USA

zz 3455-3459 (1980). 15. J. PERRET, M. LUDGATE, T. LIBERT, C. GERARD, J.E. DUMONT and M.

PARMENTIER, Biochem. Biophys. Res. Common. 1711044-1055 (1990). 16. L. CHIOVATO, P. VITTI, G. BENDINELLI, F. SANTINI, E. FIORE, A. CAPACCIOLI, M.

TONACCHERA, C. MAMMOLI, M. LUDGATE and A. PINCHERA, J. Endocrinol. Invest. 11809-816 (1994).

17. P. ANDREW& Biochem. J. 96 595-606 (1965).