Different cellular distribution of thioredoxin and subunit M1 of ribonucleotide reductase in rat tissues

Experimental

Different

Cell Research I63 (1986) 363-369

Cellular Distribution of Thioredoxin and Subunit Ribonucleotide Reductase in Rat Tissues

Ml of

HANS-ARNE HANSSON,’ BJGRN ROZELL,’ STEN STEMME,’ YLVA ENGSTRGM,’ LARS THELANDER,* and ARNE HOLMGREN3. * ‘Institute of Neurobiology, University of Giiteborg, S-40033 Giiteborg, ‘the Medical Nobel Institute, Department of Biochemistry, and ‘Department of Physiological Chemistry, Karolinska Institutet, S-10401 Stockholm, Sweden

The cellular distribution of thioredoxin and protein Ml of ribonucleotide reductase in adult rat tissues was investigated with immunohistochemical techniques using specific antisera. Tissues with high or low frequency of either mitotic or meiotic cell divisions were compared. Thioredoxin was demonstrated in many cells types that showed no detectable protein Ml of ribonucleotide reductase. A few cell types with protein Ml immunoreactivity also contained immunoreactive thioredoxin. However, in most cells no such co-localization could be demonstrated. This lack of correlation between cells containing subunit Ml of ribonucleotide reductase and the thioredoxin indicates that thioredoxin is not the physiological hydrogen donor for ribonucleotide reductase in rat tissues and that the expression of two enzymes is differently regulated. 0 1986 Academic press, IN.

The enzyme ribonucleotide reductase is required in all growing cells with DNA synthesis by supplying deoxyribonucleotides for DNA replication [l, 21. In Escherichia coli and mammalian cells the reduction of the four ribonucleotides requires electrons from NADPH which are transferred to ribonucleotide reductase through a series of sulfhydryl groups [l, 21. Originally [3], the reduced form of thioredoxin, a ubiquitous small (M, 12000) protein with a redox-active cysteine dithiolkystine disulfide was isolated as the immediate hydrogen donor for ribonucleotide reductase. Oxidized thioredoxin is regenerated by NADPH in a reaction catalysed by the FAD-containing enzyme thioredoxin reductase [4]. It was thought for some time that thioredoxin is the only carrier of reducing power to ribonucleotide reductase. However, a mutant of E. coli totally devoid of thioredoxin was found to form deoxyribonucleotides at normal rates [5]. This led to the isolation of a second hydrogen donor system, where reduced glutathione (GSH) is the reductant for ribonucleotide reductase in the presence of a small protein called glutaredoxin [5-71 also containing a redox-active dithiol. GSSG formed in the reaction is reduced by NADPH and glutathione reductase [6, 71. Glutaredoxin was subsequently demonstrated in calf thymus [8], using ribonucleotide reductase from this organ [9] and has been purified to homogeneity [ 101. * To whom offprint requests should be sent. Address: Department of Physiological Chemistry, Karolinska Institutet, Box 60400, 10401 Stockholm, Sweden. Copyright @ 1986 by Academic Press. Inc. All rights of reproduction in any form reserved 0014427/86 $03.00

364 Hansson et al.

Thioredoxin and glutaredoxin from calf thymus show no immunological crossreactivity [8] or cross-reactivity with their reduction systems (NADPH-thioredoxin reductase and GSH and NAGPH-glutathione reductase, respectively) [lo]. Ribonucleotide reductase has recently been isolated from calf thymus in highly purified form [9, 111. It consists of two non-identical subunits, protein MI and M2. Protein Ml is a dimer of il4, 170000 [9]. The enzyme is tightly coupled to DNA synthesis and normally shows very low or undetectable activity in adult liver [12-141. Thioredoxin, on the other hand, is present in tissues like liver irrespective of DNA synthesis [14-161and has in recent years been implicated in several reactions as a protein disultide reductase [17]. The relative contribution of the thioredoxin and glutaredoxin systems in ribonucleotide reduction by mammalian cells is not yet known. The aim of the present study was to localize thioredoxin and ribonucleotide reductase by immunohistochemical techniques in several tissues of adult rats to analyse possible co-localization, required for functional interactions. MATERIALS AND METHODS Animals Albino Sprague-Dawley rats (18&200 g) of both sexes were used. They were fed pellets and water ad libitum.

Antisera Thioredoxin was purified to homogeneity from rat liver [16] and used for immunization of rabbits [22]. The antiserum contained an antibody titre per ml, that completely inhibited the activity of 10 nmole of thioredoxin measured with an insulin disulfide reduction assay [16]. Mouse monoclonal antibodies against protein Ml of calf thymus tibonucleotide reductase were prepared as previously described 118, 191. Secondary antibodies labelled with either fluorescein or Texas Red were of commercial origin (Amersham International, Amersham, Bucks, UK, and Dakopatt A/S, Copenhagen, Denmark).

Preparation of Tissues for Immunojluorescence Tissues of anaesthetized rats were fixed by transcardial perfusion with freshly prepared buffered 4% formaldehyde, atIer an initial rinse with 0.15 M sodium chloride-O.01 M potassium phosphate, pH 7.2 (PBS). The tissues to be examined, i.e. testis, ovary, spleen, and thymus were identified, dissected and further futed by immersion in 4% formaldehyde for 30 min. After rinsing in 7.5% sucrose in PBS at 4°C overnight the specimens were frozen in isopentane chilled with liquid nitrogen. Sections, 7-10 urn thick, were prepared in a cryostat at about -20°C and allowed to attach to gelatine-coated glass slides.

Immunofluorescence Microscopy The slides with tissue sections were preincubated in normal horse serum (DAKO, Copenhagen) diluted 1: 50 in PBS for 30 min at +2O”C. Excess serum was carefully removed and the slides were then immediately incubated with rabbit antithioredoxin serum [22] diluted 1: 800 in PBS containing 2% normal horse serum, or mouse monoclonal antibodies against ribonucleotide reductase subunit Ml as described [19]. After 16 h at +4”C the slides were rinsed for 3x10 min in PBS and incubated with secondary antibodies [19]. After rinsing [19], the sections were mounted in buffered glycerol containingparaphenylendiamine [19] and examined in a Zeiss microscope equipped with ePiExp Cell Res 163 (1986)

Thioredoxin

and ribonucleotide

Table 1. Distribution

and staining intensity of thioredoxin reductase subunit Ml (RR) in rat organs

reductase

365

(T), and ribonucleotide

Testis Protein

Connective tissue

Leydig interstitial celI

-

+++ -

T RR-

Spermatogonia -0 +++

Primary spermatocytes -

Spermatids and sperms -

Sertoli cells + -

LI 11-14 % of seminiferous tubules contain positive spermatogonia OwrY Oocytes

Follicular cells

Cells in atretic follicles

Interstitial gland cells

Theta intema

Theta extema

+ -

+++ +++

-

++ -

++ +

-

Capsule and trabeculae

Medullary epithelial cells

Epithelial reticular cells

Lymphocytes

Plasma cells

-

+++ -

+++ -

++

+++ -

Mast cells -

Frotein

Capsule and trabeculae

Red pulp

T RR

-

++ +

Subcapsular lymphoid Central tissue artery +++ +

Protein T RR-

Connective tissue -

Germinal epithelium cells +++ -

Thymus Frotein T RRSpleen White Pulp +++

Marginal Reticular zone cells + +

++

-, Not detectable; +, slight; + +, moderate; + + +, strong immunohistochemical reaction.

illumination and filters for fluorescence microscopy. Both single- and double-labelled specimens were prepared and analysed. To check the specificity of the immunohistochemical reactions, the following controls were performed in each case. Incubation with primary antibodies absorbed with excess antigen, omission of either primary or secondary antibodies, and incubation with preimmune serum.

RESULTS By immunofluorescence both thioredoxin and protein Ml of ribonucleotide reductase could be localized in tissues from adult rats. In testis, ovary, thymus, spleen and the small intestine different cells positive for thioredoxin and ribonucleotide reductase were observed, which enabled the examination of the relative distribution of the two proteins (see table 1). Generally speaking, thioredoxin was much more widespread in cells, whereas ribonucleotide reductase was only found 24-868334

Exp Cell Res 163 (1986)

366 Hansson et al.

Fig. 1. Double immunofluorescence staining for (a) thioredoxin, and (b) protein Ml of ribonucleotide

reductase in a section of rat testis. The interstitial Leydig cells (15) are intensely stained for thioredoxin but lack ribonucleotide reductase, which is confined to spermatogonia. ST, Seminiferous tubule. (c-f) Double-labelled sections of rat ovary. The (c) follicular Q and (e) luteal cells show moderate to high reactivity for thioredoxin but (d-j) no co-variation in intensity with protein Ml of ribonucleotide reductase. Note the decrease in ribonucleotide reductase immunoreactivity with increased follicular size. The (c, 6) theta cells (arro&zead) show high reactivity for both antigens, while (e, j) the germinal epithelium (nrrow) only showed immunoreactive thioredoxin. 0, Ovum. (g) Control section of ovary with primary antibody omitted. (a, b)~ 125.

in cells with DNA synthesis. Both thioredoxin and ribonucleotide reductase were localized in the cytoplasm of cells. No reactivity could be demonstrated in the nucleus or extracellularly. All controls showed negative reaction (fig. 1g). Testis Sections from testis showed high reactivity for thioredoxin in all Leydig interstitial cells (fig. 1a). About 10% of the seminiferous tubules contained Exp Cell Res 163 (1986)

Thioredoxin and ribonucleotide reductase

367

Fig. 2. (a, b) Double-labelled section of rat thymus. The epithelial cells show (a) intense staining of thioredoxin, especially in the medulla (M), while (b) numerous mainly large lymphoid cells are highly positive for ribonucleotide reductase protein Ml. Capsule marked by arrow. (c-f) Double-labelled sections of rat spleen. Reticular cells contain (c, e) thioredoxin, while (d, j) numerous cells, mostly lymphocytes, show variable staining for protein Ml of ribonucleotide reductase, illustrating the partial lack of co-variation between the two examined proteins. cn, Central artery.

spermatogonia with low to moderate reactivity for thioredoxin beside slightly reactive Sertoli cells. However, ribonucleotide reductase reactivity could only be observed in spermatogonia, which were equally intensely stained within each individual tubule, but with variability between adjacent tubules (fig. 1 b). Ovary The ovary showed high reactivity of thioredoxin in the follicular, luteal and germinal epithelium cells, moderate in the interstitial gland cells and theta intema and faint in the oocytes (fig. 1 c, e). However, only the follicular cells and some cells in the theta intema contained ribonucleotide reductase (fig. 1d, fi. Thymus In thymus, high thioredoxin activity was demonstrated in reticular epithelial cells in the medulla and the cortex as well as in rare plasma cells (fig. 2a). Exp Cell Res 163 (1986)

368 Hansson et al. Ribonucleotide reductase could be demonstrated mainly in the numerous lymphocytes, distributed throughout the thymus, but most frequently in the cortex (fig. 2b). Spleen Spleen sections displayed similar patterns of cellular immunoreactivity as observed in thymus (fig. 2c-j). Thus, the reticular epithelial cells showed high thioredoxin immunoactivity (fig. 2c, e). Further, plasma cells could be demonstrated to be moderately stained. Ribonucleotide reductase, on the other hand, was mainly demonstrable in lymphocytes, especially in periarterial lymphatic sheaths and lymphatic nodules (fig. 2d, j). Small intestine In the small intestines, all epithelial lining cells both on villi and in the crypts showed moderate to high reactivity for thioredoxin, as did plasma cells and autonomic nerve cells. However, only epithelial cells in a narrow zone at the bottom of the Lieberkiihn intestinal crypts contained immunoreactive ribonucleotide reductase (not shown) DISCUSSION This immunohistochemical study demonstrates that thioredoxin is widespread in many cells of adult rats, irrespective of growth and DNA synthesis, in contrast to the ribonucleotide reductase subunit Ml that is confined to cells known to proliferate [19]. This is in accordance with previous results; i.e. demonstrating thioredoxin to be present in adult rat liver [13-161 known to almost lack detectable ribonucleotide reductase activity [12-141. Thioredoxin shows comparable levels in regenerating rat liver, whereas ribonucleotide reductase is induced at least 20-fold 48 h after partial hepatectomy [13]. Similar results were obtained in a series of rat hepatomas [14], where the fast-growing Novikoff hepatoma showed at least a 200-fold increase in ribonucleotide reductase activity compared with normal liver. However, thioredoxin and thioredoxin reductase were found to be present at comparable levels in normal liver and hepatomas [14]. This has lead to suggestions that thioredoxin is involved in several other metabolic functions; for a review see [17]. Although immunofluorescence is not a fully quantitative technique, it enables comparisons of relative levels in different cells of a tissue. The double staining of cells showed that high levels of thioredoxin and ribonucleotide reductase could be demonstrated in the same cells in the ovary, small intestine and lymphoid organs. However, in most tissues, no obvious correlation between thioredoxin and protein Ml of ribonucleotide reductase could be demonstrated as in the testis, thymus and spleen. The results from testis, i.e., showed high activity of ribonucleotide reductase in spermatogonia, where thioredoxin is very low and thioredoxin reductase hardly detectable [22]. Exp Cell Res 163 (1986)

Thioredoxin and ribonucleotide reductase

369

In vitro, purified ribonucleotide reductases from various systems [ 1,2] generally use thioredoxins as hydrogen donors. However, the results of Hopper & Iurlano [20] suggest that thioredoxin purified from rabbit bone marrow fails to serve as a hydrogen donor for its homologous ribonucleotide reductase. Additional evidence that thioredoxin is not essential for deoxyribonucleotide synthesis comes from the demonstration that the thioredoxin from Corynebacterium nephridii is not a hydrogen donor for its homologous ribonucleotide reductase [21]. Our results strongly suggest that thioredoxin is not the physiological hydrogen donor for ribonucleotide reductase in many cell types of the rat. So far, the tissue and cellular distribution of glutaredoxin in mammalian cells is unknown. Obviously, immunohistochemical studies of the cellular distribution of thioredoxin, glutaredoxin and ribonucleotide reductase will be of interest when antibodies against glutaredoxin become available. This investigation was supported by grants from the Swedish Medical Research Council, the Swedish Cancer Society, the Swedish Natural Research Council and the University of Gothenburg.

REFERENCES 1. Thelander, L & Reichard, P, Ann rev biochem 48 (1979) 113. 2. Holmgren, A, Curr top cell reg 19 (1981) 47. 3. Laurent, T C, Moore, E C & Reichard, P, J biol them 239 (1964) 3436. 4. Moore, E C, Reichard, P & Thelander, L, J biol them 239 (1964) 3445. 5. Holmgren, A, Proc natl acad sci US 73 (1976) 2275. 6. - J biol them 254 (1979) 3664. 7. - Ibid 254 (1979) 3672. 8. Luthman, M, Eriksson, S, Holmgren, A & Thelander, L, Proc natl acad sci US 76 (1979) 2158. 9. Engstriim, Y, Eriksson, S, Thelander, L & Akerman, M, Biochemistry 18 (1979) 2941. 10. Luthman, M & Holmgren, A, J biol them 257 (1982) 6686. 11. Thelander, L, Eriksson, S & Akerman, M, J biol them 255 (1980) 7426. 12. King, C D & Van Lanker, J L, Arch biochem biophys 129 (1%9) 603. 13. Larsson, A, Eur j biochem 11 (1%9) 113. 14. Elford, H L, Freese, M, Passamani, E & Morris, H P, J biol them 245 (1970) 5228. 15. Larson, G & Larsson, A, Eur j biochem 26 (1972) 119. 16. Luthman, M & Holmgren, A, Biochemistry 21 (1982) 6628. 17. Holmgren, A, Ann rev biochem 54 (1985) 237. 18. Engstriim, Y, Acta them Stand ser B (Org them them b&hem) 5 (1982) 343. 19. Engstriim, Y, Rozell, B, Hansson, H-A, Stemme, S & Thelander, L, EMBO j 3 (1984) 863. 20. Hopper, S & Iurlano, D, J biol them 258 (1983) 13453. 21. Meng, M & Hogenkamp, H P C, J biol them 256 (1981) 9174. 22. Rozell, B, Hansson, H A, Luthman, M & Holmgren, A, Eur j cell biol(1985). In press. Received June 28, 1985 Revised version received October 3, 1985

Printed

in Sweden

Exp Cell Res 163 (1986)