Bioscience Reports, Vol. 6, No. 6, 1986

Superoxide Dismutase in Human Testis Preparations

Jan Johansson, Ella Cederlund, Sudhir B. Moodbidri, Anil Sheth and Hans Jiirnvall 1

Received May 23, 1986

KEY WORDS: Superoxide dismutase; testis

A protein fraction from human testis was structurally investigated. The main component of the fraction reported to contain inhibin-like activity was purified and analyzed by tryptic digestion. The peptides obtained identified the protein as an enzyme, superoxide dismutase, previously known to be present in seminal plasma. The results show that superoxide dismutase is a major enzyme, also of testicular material. They further demonstrate the importance of using pure fractions, and controls such as checks with structural analysis or synthetic peptides, in the work of elucidating the nature of inhibin and other hormonal peptides.

INTRODUCTION

Inhibin is the name given to the postulated hormone of gonadal origin that is supposed to suppress hypophyseal release of FSH (follicle-stimulating hormone) without affecting LH (luteinizing hormone) release (1). Initially, two types of polypeptide, purified because of supposed inhibin activity, were characterized from preparations of seminal plasma (~8). The two types of peptide have unrelated structures, origins unexpected for true inhibin, structures related to those of known non-inhibin polypeptides, and no inhibin bioactivities in assays using pituitary cell cultures (9). Consequently, these peptides have been considered not to be inhibin (9), in agreement with conclusions from other studies (10). Furthermore, preparations likely to contain the true hormone have been described (11-14), and the complete structure has been identified as a transforming growth factor homologue (15).

Department of Chemistry 1, Karolinska Institutet, S-104 01 Stockholm, Sweden, and Institute for Research in Reproduction, Parel, Bombay, 400 012 India. 1 To whom correspondence should be addressed.

535 0144-8463/86/0600-0535505.00/0 �9 Plenum Publishing Corporation

536 Johansson et al.

However, extracts of testicular and ovarian tissues have also been reported to contain other peptide components with possible inhibin-like activity and apparently different structural properties (16, 17). Consequently, it is important to characterize the proteins in gonadal extracts. In the present work, the major component of a preparation from human testicular material with supposed inhibin-like activity is characterized. The protein is shown to be human superoxide dismutase, establishing that the major polypeptide in this preparation is not the hormone inhibin. At the same time, the results show that superoxide dismutase is a main enzyme in preparations of testicular tissue. It has previously been found in seminal plasma, and has been interpreted to have protective functions (cf. 18-20).

MATERIALS AND METHODS

Protein preparation

The inhibin-like activity was purified from human testes essentially as described for preparation of seminal inhibin-like material (21). Briefly, the tissue was homogenized in 0.01 M phosphate buffer, pH 7.4, and centrifuged (20,000 x g; 60 min). The supernatant was mixed with ethanol (4~ 3 vol.). After centrifugation, washing with ethanol and ether, drying, solubilization on 0.05 M sodium acetate buffer, pH 4.0, and chromatography on Sephadex G-100 in that buffer, the inhibin-active fraction was concentrated by ultrafiltration and equilibrated with 0.05 M Tris HC1, pH 8.0. It was then subjected to DEAE-chromatography in the same buffer with a NaC1 gradient (0-0.5 M) to yield a major protein fraction with apparent inhibin activity.

This fraction was further purified by reverse phase high performance liquid chromatography on a C18 column eluted with acetonitrile as described (4). The elution pattern is shown in Fig. 1, and the major protein fraction was the starting material for structural characterization.

Structural analysis

The protein obtained was 14C-carboxymethylated, and digested with trypsin and CN Br. Peptides were purified by high performance liquid chromatography in the same way as for the intact protein above, and analyzed for total composition and amino acid sequence. Samples for sequence analyses were degraded with the manual dimethylaminoazobenzene isothiocyanate method, and, for longer peptides, were analyzed with liquid phase sequencer degradations and subsequent phenyl- thiohydantoin identification (6). Total compositions were determined after hydrolysis with 6 M HC1/0.5 % phenol for 24 h at 110~ in evacuated tubes.

Superoxide Dismutase in Testis 537

RESULTS

The main protein from a fraction with apparent inhibin-like activity was purified from human testis by homogenization, treated with organic solvents, exclusion chromatography, and ion-exchange chromatography, as described (16, 17, 21), followed by final reverse phase high performance liquid chromatography as shown in Fig. 1. The major component obtained corresponds to a single band on SDS/polyacrylamide gel electrophoresis. The protein was carboxymethylated with ~*C-Iabelled iodoacetate after reduction, and submitted to direct liquid phase sequencer degradation. No amino acid derivatives were detected, suggesting that the protein has a blocked c~-amino group.

I.C C~4

/ /

/

/

/ 1

I J

J / /

/

/

/

/

/

/

20 3'0 4; s'0 V0tume(mt)

Fig. 1. Final purification by reverse phase high performance liquid chromatography of the material analyzed. Chromatography on a column of Ultropac in 0.I G trifluoroacetic acid with a gradient (~B) of acetonitrile. The fraction pooled and further studied is indicated by the horizontal bar at the top_

50

25

I I

&

The total composition indicated that the protein did not contain methionine. This was verified by resistance to protein cleavage with CNBr. However, trypsin treatment of the carboxymethylated protein produced several peptides, and the subsequent purification by high performance liquid chromatography produced a pattern shown in Fig. 2. Ten peptides were obtained pure and their amino acid sequences were determined. Results are also summarized in Fig. 2. It is evident that the peptides match the structure of human superoxide dismutase (22), establishing that the testicular protein studied is this enzyme. In total, 88 positions were recovered, covering 58 ~o of the entire structure, including the entire C-terminal third of the human enzyme.

538 Johansson et al.

o

o

~%{- - -}

2 1

...... } g

o

wu~l~ ;e a,ueqJ0sqv(--}

2 >

. ~ Z = ' ~ ~=~

- { ..= ",,m

-~.-~ .> ~ z

,--~ ~ ~.~

r Z e o e o .

�9 . ~ ~ ~ ~:~u-~

= . ' ~ { ~ Z

~ . - ~ ~ , . ~

Superoxide Dismutase in Testis 539

DISCUSSION

Bioactivity

The present preparation was initially monitored for inhibin-like activity (16, 17, 2]). The multiplicity of reported peptides with inhibin-like properties, the elution pattern upon high performance liquid chromatography (Fig. i), the non- reproducibility of some earlier supposed inhibin activities (9) and the recent demonstration of apparently true inhibin (15), motivated a structural characterization of the present preparation. The structure obtained clearly identified the major component as the enzyme superoxide dismutase, establishing that the main component in this type of preparation is not related to inhibin. Thus, the situation with the testicular preparation is the same as with the preparation from seminal plasma, where the initially characterized peptides have been shown not to be true inhibins (9). Furthermore, scans for similarities between the superoxide dismutase structure now determined, and all reported inhibin structures do not reveal any significant similarities, suggesting that the present testicular protein is unrelated to inhibin.

Instead, the major component of the present preparation is the human testicular counterpart of the well-known enzyme superoxide dismutase. Since this is an active enzyme affecting several systems, and recently also discovered in a completely different context in another system (radical formation in an enzyme (23)), there is no reason to ascribe superoxide dismutase any structural or functional link with true inhibins. Thus, the present characterization further illustrates the widespread occurrence of superoxide dismutase and establishes that this enzyme is common also in testicular extracts. Superoxide dismutase is further known to be a common component in seminal plasma (19).

The combined results on this and previous preparations of inhibin-like activities (2 17), demonstrate difficulties in ascribing inhibin activities to components of biological preparations. None of the small proteins initially suspected and characterized represent possible inhibin products and are now considered not to be true inhibins. Instead, one peptide (2, 3) has been found to be related to a fragment of high molecular weight seminal vesicular kininogen (24), another peptide has been found to be related to sperm-coating antigen (6, 25), and a third preparation is now shown to have a major component of superoxide dismutase at the same time as the apparently true inhibin activity has been found to be homologous with the transforming growth factor (15).

The human superoxide dismutase structure

The peptides now recovered fit the amino acid sequence of human superoxide dismutase deduced from analysis ofa fibroblast-derived cDNA (22). At a few positions where earlier peptide reports (on the erythrocyte enzyme) have suggested minor discrepancies regarding the amidation status of residues at positions 11 (26), 52 and 53 (27), or the type of residue at position 17 (27), the present peptide data support the cDNA derived structure (22) rather than the earlier peptide analyses (26, 27). The nature of the previous differences suggests that they do not reflect a true

540 Johansson et al.

po lymorph i sm, a l though the analyses were per formed on mater ia l of different tissue origins. In any event, the present results prove tha t tes t icular superox ide d i smutase has a s t ructure in agreement wi th tha t previously indirect ly deduced for the enzyme from

f ibroblasts . This also means tha t the h u m a n enzyme, in re la t ion to the bovine form (28), has an extra cysteine residue, since Ser-109 of the bovine pro te in is replaced by a cysteine residue. Therefore, since superoxide d ismutase , like p l a sma a lbumin , has the

u n c o m m o n proper ty of exhibit ing bo th a disulfide br idge and a thiol group (28), the add i t i on of one further cysteine residue gives a var ia t ion in the ra t io of disulfides versus

thiols for this pro te in type.

A C K N O W L E D G E M E N T

This work was suppor t ed by grants from the Leo Research F o u n d a t i o n , and the

Swedish Cancer Society (project 1806).

REFERENCES

1. McCullagh, D. R. (1932). Science 76:19-20. 2. Ramasharma, K., Sairam, M. R., Seidah0 N. G., Chretien, M., Manjunath, P., Schiller, P. W.,

Yamashiro, D. and Li, C. H. (1984). Science 223:l 199-1202. 3. Seidah, N. G., Ramasharma, K., Sairam, M. R., and Chretien, M. (1984). FEBS Lett. 167:89-102. 4. Sheth, A. R., Arbatti, N., Caflquist, M, and JiSrnvall, H. (1984). FEBS Lett. 165:11-15. 5. Seidah, N. G., Arbatti, N. J., Rochemont, J., Sheth, A. R., and Chretien, M. (1984). FEBS Lett.

175:349-355. 6. Johansson, J., Sheth, A., Cederlund, E, and J~Srnvall, H. (1984). FEBS Lett. 176:21-26. 7. Arbatti, N. J., Seidah, N. G., Rochemont, J., Escher, E., Sheth, A. R., and Chretien, M. (1985). FEBS

Lett. 181, 57-63. 8. Li, C. H., Hammonds, R. G. Jr., Ramasharma, K., and Chung, D. (1985). Proe. Natl. Acad. Sci. USA

82:4041~4044. 9. Kohan, S., FriSysa, B., Cederlund, E., Fairwell, T., Lerner, R., Johansson, J., Khan, S., Ritzen, M.,

JiSrnvall, H., Cekan, S., and Diczfalusy, E. (1986). FEBS Lett. in press. 10. de Jong, F. H., and Robertson, D. M. (1985). Mol. Cell. Emlocrinol. 42, 95 103. 11. Robertson, D. M., Foulds, L. M., Leversha, L., Morgan, F. J., Hearn, M. T. W., Burger, H. G.,

Wettenhall, R. E. H., and de Kretser, D. M. (1985). Biochem. Biophys. Res. Commun. 126:220-226. 12. Miyamoto, K., Hasegawa, Y., Fukuda, M., Namura, M., Igarashi, M., Kangawa, K., and Matsuo, H.

(1985). Biochem. Biophys. Res. Commun. 129:396-403. 13. Rivier, J., Spiess, J., McClintock, R., Vaughan, J., and Vale, W. (1985). Biochem. Biophys. Res. Commun.

133 : 120-127. 14. Ling, N., Ying, S.-Y., Ueno, N., Esch, F., Denoroy, L., Guillemin, R. (1985). Proc. Natl. Acad. Sci. USA

82:7217-7221. 15. Mason, A. J., Hayflick, J. S., Ling, N., Esch, F., Ueno, N., Ying, S.-Y., Guillemin, R., Niall, H., and

Seeburg, P. H. (1985). Nature 318:659-663. 16. Sheth, A. R., Joshi, L. R., Moodbidri, S. B., and Rao, S. S. (1979). J. Repro& Fertil., Suppl. 26:71 85. 17. Vijayalakshmi, S., Bandivdekar, A. H., Joshi, L. R., Moodbidri, S. B., and Sheth, A. R. (1980). Arch.

Androl. 5:179-188. i8. Menella, M. R. F., and Jones, R. (1980). Biochem. J. 191:289-297. 19. Marklund, S. L. (1984). In: Oxygen Radicals in Chemistry and Biology (W. Bors, M. Saran, and D. Tait,

Eds.), de Gruyter, Berlin, New York, pp. 765 777. 20. Petkan, A., Chelack, W. S., and Pleskach, S. D. (1976). Int. J. Radiat. Biol. 29:297-299. 21. Thakur, A. N., Vaze, A. Y., Dattatreyamurty, B., Arbatti, N. J., and Sheth, A. R. (1978). Indian J. Expt.

Biol. 16, 854-856. 22. Sherman, L., Dafni, N., Lieman-Hurwitz, J., and Groner, Y. (1983). Proc. Natl. Acad. Sci. USA

80:5465 5469.

Superoxide Dismutase in Testis 541

23. Eliasson, R., JiSrnvall, H., and Reichard, P. (1986). Proc. Natl. Acad. Sci. USA, in press. 24. Lilja, H. and Jeppsson, O. (1985). FEBS Lett. 182:181-184. 25. Akiyama, K., Yoshioka, Y., Schmid, K., Offner, G. D., Troxler, R. F., Tsuda, R., and Hara, M. (t985).

Biochem. Biophys. Acta 829:288 294. 26. Barra, D., Martinin, F., Bannister, J, V., Schinina, M. E., Rotilio, G., Bannister, W. H., and Bossa, F.

(1980). FEBS Lett. 120:53-56. 27. Jabusch, J. R., Farb, D. L., Kerschensteiner, D. A., and Deutsch, H. F. (t980). Biochemistry 19:2310-

2316. 28. Abernethy, J. L., Steinman, H. M., and Hill, R. L. (1974). J. Biol. Chem. 249:7339-7347.