Rat GST 8-8 is expressed predominantly in myeloid origin cells infiltrating the gravid uterus

Inl. J. Biochem. Cell Bid. Vol. 29, No. 5, pp. 807-813, 1997

c m 1997 Elsevier ScienceLtd. All rights reserved Printed in Great Britain

Pergamon PII: S1357-2725(96)00173-2

1357-2725/97 $17.00 + 0.00

Rat GST 8-8 is Expressed Predom inantly in Myeloid Origin Cells Infiltrating the Gravid Uterus SANJAY AWASTHI,‘* SHARAD S. SINGHAL,‘,’ SANJAY K. SRIVASTAVA,3 MEENA CHAUBEY,’ JOHN T. PIPER,’ PIOTR ZIMNIAK, C. YALLAMPALLI,5 S. RAJARAMAN; YOGESH C. AWASTHI* ‘Departments of Internal Medicine, University of Texas Medical Branch, Galveston, TX 77555-1067, U.S.A., ‘Departments of Human Biological Chemistry and Genetics, University of Texas Medical Brunch, Galveston, TX 77555-1067, U.S.A., ‘Departments of Cancer Research Laboratory, Mercy Hosptial of Pittsburgh, Pittsburgh, PA 15219, U.S.A., 4Departments of Internal Medicine and Biochemistry & Molecular Biology, University of Arkansas for Medical Sciences, and McClellan VA Hospital, Little Rock, AR 72205, U.S.A., ‘Departments oj Obstetric & Gynecology, University of Texas Medical Branch, Galveston, TX 77555-1067, U.S.A. and 6Departments of Patholog.y, University of Texas Medical Branch, Galveston, TX 77555-1067, U.S.A. Previous studies from our laboratory have shown a relatively high expression of rGST8-8 in uterine tissues. This GST isozyme displays relatively high glutathione-peroxidase activity towards lipid-hydroperoxides and towards toxic 4-hydroxyalkenals generated from lipid peroxidation. Since the uterus is a unique organ, subject to oxidative stress due to infiltration by immune effector cells during gestation and because this infiltration is readily identifiable histologically, the studies reported herein were performed to localize the cell specific expression of rGSTS-8 to determine whether immune effector cells infiltrating the pregnant rat uterus specifically expressed rGST8-8. A 75 bp end-radiolabeled cRNA probe was prepared from the full length mGSTA4-4 cDNA from the region which is highly homologous with rGST8-8. This cRNA probe was used for in situ hybridization studies to localize rGST8-8 in specific cell types of gravid rat uterus. Results of these studies indicate that this GST isozyme is selectively expressed in myeloid origin cells such as monocytes/macrophages, and neutrophils infiltrating the uterine endometrium and in vascular walls. Selective expression of rGST8-8 in the myeloid origin cells, which are known to generate higher levels of reactive oxygen species, suggests that this GST isozyme plays an important role in the protection mechanisms against lipid peroxidation. 0 1997 Elsevier Science Ltd Keywords: Glutathione S-transferase peroxidation

Uterus

Oxidative stress

Macrophage

Lipid

Int. J. Biochem. Cell Biol. (1997) 29, 807-813

1NTRODUCTION

bio-active lipids by these cells has been postulated to mediate incompletely understood physiological and pathologic effects on uterine tissues, utero-placental interface or the fetus during normal and abnormal gestation, parturition and the post-partum period (Hunt, 1989, 1994; Lea and Clark, 1991: Uotila et al., 1991, 1993; Davidge et al., 1992; Jain et al., 1993; Moldovan et al., 1994). These immune effector cells are capable of generating large quantities of highly reactive and damaging reactive oxygen

During pregnancy, bone marrow derived immune effector cells are known to infiltrate heavily the endometrial stroma and myometrial connective tissues (Lawrence et al., 1980; Clark et al., 1987; Bulmer et al., 1988; Hunt, 1989, 1994; Lea and Clark, 1991). Production of growth factors, reactive oxygen metabolites and *To whom all correspondence should be addressed. Received 12 August 1996; accepted 9 December 1996. 807

species (Hoffbrand and Pettit, 1988), which in concert with the redox cycling of transitional metal ions can initiate or propagate the cascade of lipid peroxidation (Halliwell, 1991; Krinsky, 1992; Beuttner, 1993; Awasthi et (II., 1996). Lipid peroxidation gives rise to large quantities of lipid-hydroperoxides, lipid-hydroperoxy radicals and a,/?-unsaturated carbonyls (including 4-hydroxy-2-nonenal, 4HNE). The biologic half-lives of some electrophilic products of lipid peroxidation are sufficiently long to allow them to diffuse significant distances and cause deleterious effects such as acceleration of lipid peroxidation or DNA mutations in the cell of origin and surrounding cells (Baker and He, 199 1; Krinsky, 1992). The concerted action of enzymatic antioxidants such as superoxide dismutase and catalase as well as chemical antioxidants such as x-tocopherol and ascorbate significantly reduce the chain reaction of lipid peroxidation, but do not abrogate it (Niki, 1991; Meister, 1992; van Acker et al., 1993; Ames et al., 1993). GSTs are a large family of multi-functional enzymes which are a secondary defense against the toxic effects of exogenous and endogenously generated electrophilic toxins such as those produced during lipid peroxidation. The y-class of GSTs is a particularly important lipid peroxidation defense owing to their relatively high GSH-peroxidase activity towards lipid hydroperoxides (Awasthi et al., 1980; Mannervik and Danielson. 1988; Singhal et al., 1992; Hayes and Pulford, 1995). In the rat, a minor isozyme of the cc-class GST isozyme (rGST8-8) has been characterized that shows particularly high activity towards the electrophilic products of lipid peroxidation, 4-hydroxynonenal (Jensson of GST is et al., 1986). This isoform structurally, functionally and immunologically distinct from the known x, p and rr classes of GSTs and potentially represents a significant cellular defense mechanism against the toxic t-4-hydroxy-2-alkenals generated during peroxidation of polyunsaturated fatty acids (Jensson et al., 1986; Stenberg et ul., 1992). Because this GST isozyme represents only about 2% of total GST activity, its overall significance in protection of tissues from lipid peroxidation has not been understood. We have recently reported that rat uterine tissues contain a relatively high level of this GST isozyme (Singhal et al., 1996). Since the uterus is a unique organ subject to oxidative stress as a result of infiltration by immune effector cells and this infiltration is

readily identifiable histologically, the studies reported herein were performed to localize the cell specific expression of rGST8-8 to determine whether immune effector cells infiltrating the pregnant rat uterus specifically expressed rGSTS-8. MATERIALS

AND METHODS

We have previously cloned mGSTA4, the mouse homolog of rGST8-8, from a lung cDNA library and expressed it in prokaryotic and eukaryotic cells (Zimniak et al., 1992, 1994). A 75 bp sequence from mGSTA4 with minimal homology to other GSTs (Zimniak et al., 1994), but with 92% homology with the corresponding region of rGST8-8, was used to synthesize sense and anti-sense end-radiolabeled cRNA probes for in situ hybridization. The 75 bp probe was generated using the full length mGSTA4 cDNA, excised by EcoRl from pGEM3zf( + ) 5.7-12. PCR primers were designed to amplify a specific region of the mGSTA4 cDNA (nucleotide positions 387461). The PCR product was subcloned into Invitrogen TA cloning vector PCR”. The identity of the insert was verified by sequencing. The vector was linearized using Barn HI or Not I. Transcription was carried out using T7 or SP6 RNA polymerase (Promega). The transcribed RNA was end labeled with [x-~S] UTP using the Promega Riboprobe i/z vitro transcription system. After purification, the specific radioactivities of both the sense and were identical anti-sense probes (1.6 F 0.1 x 10’ cpm/iil). The labeled anti-sense or sense cRNA probes (I x 10h cpm) were diluted in 200 ~1 of hybridization buffer [50% formamide, 5 x SSC, 0.5% Genius (Boehringer Mannheim), 2 ~1 mercaptoethanol]. Briefly, timed pregnant Sprague-Dawley rats (Charles River, Wilmington, MA) were placed under anesthesia by intra-peritoneal injection of Ketamine (40 mg/kg body wt) and Xylazine (5 mg/kg body wt) and the abdomen and thorax were opened. The left ventricle was cannulated using a 25-gauge catheter, and a Harvard Pump was used to perfuse the animals with 20 ml PBS containing 1 mM EDTA and 1000 U/ml heparin, followed by 4% paraformaldehyde in PBS. Tissues were harvested and fixed in 4% paraformaldehyde for 4 hr. After washing with 20% sucrose in PBS for 12 hr with three to four changes and with DEPC-H?O, the sections were dehydrated using a graded series of increasing ethanol concentrations at room temperature

Cellular localization

of glutathione s-transferase

and embedded in paraffin. From the paraffin blocks, 6-,um-thick sections were cut using a Microtome Cryostat (International Equipment Co., Needham Heights, MA), placed on silanized slides and baked overnight at 40 C. Prior to in situ hybridization with radio-labeled sense and anti-sense probes, paraffin was dissolved from the sections by washing with xylenes and pre-treatments of the sections were performed according to a method described previously (Ausbel et al., 1994). After hybridization. slides were repeatedly washed with 2 x SSC at room temperature, 1 x SSC at 6O’C, 0.5 x SSC at 60’C and finally with 0.1 x SSC at 60°C. Sections were air dried and dipped in Ilford Nuclear Emulsion K 0.50 (Polysciences, Warrington, PA). The slides were allowed to incubate at - 7O’C for 14 days before developing in D 19 solution (Kodak). The slides were counter stained with hematoxylin and examined by light microscopy. Immunohistochemistry was performed on uterine sections using the method described by Kurita et ul. (1992) with some modifications. Paraffin was dissolved from sections by washing with ice-cold acetone for 5 min. The sections were allowed to air-dry, followed by washing thoroughly with TBS for 10 min. The slides were treated with normal goat serum (1: 100 dilution in TBS) for 30 min to block nonspecific binding. Slides were then incubated with specific polyclonal antibodies against recmGSTA4-4 at 4’C overnight in a humidified chamber followed by several washes with TBS. The slides were then incubated with FITC-conjugated goat anti-rabbit IgG (1:50 dilution in TBS) for 2 hr at room temperature in a humidified chamber, and washed thoroughly with TBS. The sections were counter stained with Evan’s blue dye (three drops in 10 ml TBS) for 5 min at room temperature, followed by several washes with TBS. The slides were mounted in a PBS/glycerol solution, and observed using an fluorescence microscope (Nikon Labophot, Tokyo, Japan). RESULTS

AND DISCUSSION

Exposure of the photographic emulsion to the radiolabeled anti-sense resulted in deposition of black granules of photographic emulsion. The corresponding sense riboprobe served as the control for non-specific binding. The background non-specific binding observed with the sense-ribonrobe used in the present studies was

X09

found to be minimal and without any specific pattern (Fig. 1A). With the anti-sense riboprobe, focal dark staining could be seen in the endometrium at low power (Fig. 1D). At higher power (200 x ), cells with darkly stained nuclei can be seen with both sense and anti-sense probes (Fig. 1B and E, sense and anti-sense riboprobes, respectively). At the highest magnification (630 x ) (Fig. 1C and F) granular deposits can be seen in a pattern consistent with a cytoplasmic distribution of signal only with the anti-sense riboprobe. The histologic appearance and pattern of distribution of these cells is consistent with their being perivascular monocyte/macrophages. Examination of uterine myometrium and endometrium at high power indicated that neither uterine myocytes nor endometrial cells were significantly stained by the anti-sense riboprobe. Cells resembling tissue macrophages found near myometrial blood vessels were stained. Although some cells histologically resembling or neutrophils eosinophils were stained, clearly unstained neutrophils were also demonstrable. Specific stain was also seen, associated with the walls of some small muscular arterioles, but the present results were insufficient to demonstrate clearly the presence of rGST8-8 in vascular endothelium. Localization of expression of rGST8-8 was also carried out by immunohistochemistry using polyclonal antibodies against recmGSTA4-4 (Fig. 2A and B). Specific (bright green-yellow) fluorescence was observed predominantly in the endometrial layer. Individual cells infiltrating the perivascular and interstial spaces between glandular endometrial cells as well as blood vessel walls were observed to be stained intensely with fluorescence. These results were consistent with those observed with in situ hybridization and indicate that rGST8-8 is expressed in certain granulocytic immune effector cells and probably also in blood vessel walls. The present observations indicate that the expression of rGST8 is at least an order of magnitude greater in myeloid-origin cells including monocyte/macrophages and neutrophils than surrounding parenchymal cells. These findings appear to be consistent with the belief that granulocytic cells are subject to a much greater level of oxidative stress than surrounding parenchymal tissues. Available evidence indicates that granulocytic immune effector cell infiltration of the uterine endometrium during pregnancy is a crucial part of the process of

810

Sanjay Awasthi (21trl.

gestation (Lawrence et ul., 1980; Clark et al., 1987, Bulmer rf ul., 1988; Hunt, 1989, 1994; Lea and Clark, 1991). Intimate association of genetically dissimilar organisms during ges-

tation appears to provoke an immunologic response in the uterus, which promotes gestation (Lawrence et nl., 1980). Infiltration of monocyte-macrophages closely associated with

Fig. I. Detection of rGST 8 expression in gravid rat uterine tissue by in situ hybridization. Sense and anti-sense direction end-radiolabeled riboprobes were prepared from a highly specific 75 bp sequence from mGST A4 cDNA. Histologic sections of appropriately prepared tissues were allowed to hybridize with the sense and anti-sense riboprobes. After stringent washes, they were developed with photographic emulsion for 2 weeks. Photomicrographs of sections incubated with the sense (A, B and C) or anti-sense (D, E and F) riboprobes are presented at 100,400 and 630 x magnification. Histologic landmarks (labeled using lowercase letters) include (a) longitudinal and (b) circumferential layers of uterine myometrium, (c) uterine endometrium, (d) endometrial stroma, (e) endometrial epithelium, (f) perivascular infiltration of monocyte/macrophages and (g) vascular lumen.

Cellular localization

of glutathione

s-transferase

811

Fig. 2. Immunohistochemical localization of rGST8-8 in rat uterus using (A) preimmune serum and (B) the antibodies against ret-mGSTA4-4. Figures are at 100 x magnification. Histologic landmarks indicated are (a) endometrial lining and (b) inner layer of myometrium.

extravillus trophoblasts has been demonstrated in the decidual gravid endometrium (Clark et ul., 1987; Bulmer et al., 1988); this infiltration is dynamic with respect to the stage of gestation (Lawrence et al., 1980; Clark et al., 1987; Bulmer et al., 1988; Uotila et ul., 1991; Jain 1993). Inter-cellular communication et d., through numerous macrophage origin cytokines has been shown to play a crucial role in fetal implantation, development and growth (Lea and Clark, 1991). In addition to cytokines, macrophages are known to produce metabolites of arachadonic acid through lipid peroxidation (Hunt, 1989, 1994). Temporal variations in blood levels of oxidative metabolites in gravid women (Uotila et al., 1991; Yallampalli et al., 1994a, 1994b) coincident with degree of macrophage infiltration in the uterus (Lawrence et al., 1980; Clark et al., 1987; Bulmer et al., 1988) suggest that the level of oxidative stress

caused by macrophages may be quite significant. Oxidative metabolites have been shown to affect the blood flow through their actions on vascular smooth muscle cells. Under certain conditions, abnormally high accumulation of monocyte-macrophages and the effects of the products of ensuing lipid peroxidation on uterine blood supply appear to be involved in the pathogenesis of spontaneous abortion, pre-eclampsia and pre-term labor (Clark et al., 1987; Bulmer et al., 1988; Hunt, 1989, 1994; Lea and Clark, 1991; Uotila et al., 1991, 1993; Moldovan et al., 1994). Association of conditions such as pre-eclampsia and pregnancy-induced hypertension with decreased antioxidants in the blood also tends to support the assertion that excessive oxidative stress at the utero-placental interface plays a significant role in gestational pathology (Davidge et al., 1992; Uotila et al., 1993).

812

Sanjay

Aw

Since the formation of lipid-hydroperoxides represents a rate controlling step in the free radical chain reaction of lipid peroxidation, any catalytic activity that limits the accumulation of lipid-hydroperoxides may attenuate the formation of toxic by-products. rGST8-8 can catalyze GSH-linked reduction of lipid hydroperoxides as well as GSH-conjugation of reactive x$-unsaturated carbonyls. Expression of this enzyme in granulocytic cells, which are a significant source of tissue oxidative stress, implies that it may function to reduce the formation of diffusible reactive lipid-peroxidation products from these cells, thus ameliorating toxic effects of inflammatory oxidative stress on surrounding tissues. In the context of the gestational process, which requires an environment relatively free of reactive mutagens, the function of this group of GSTs can be interpreted as providing protection to the fetus from the adverse effects of reactive toxins produced from lipid peroxidation. Acknoa,k~l~~ernrnrs-Supported in part by NIH grants CA27967 awarded by the National Cancer Institute to Y. C. A. and CA63660 awarded to S. A.

REFERENCES Ames B. N., Shigenaga M. K. and Hagen T. M. (1993) Oxidants, antioxidants, and the degenerative diseases of aging. Proc. Natl. Acad. Sri. U.S.A. 90, 7915.-7922. Ausbel F. M., Brent R., Kingston R. E.. Moore D. D., Smith J. A. and Struhl K. (1994) Curre/r/ prorocol.\ in Molecular Biology,, Vol. 2, Chapter 14. John Wiley, New York. Awasthi Y. C., Dao D. D. and Saneto R. P. (1980) Interrelationship between anionic and cationic forms of glutathione S-transferases of human liver. Bioclzenz. J. 191, l---10. Awasthi Y. C., Singhal S. S. and Awasthi S. (1996) Nutritior? and Crtnc,er (Edited by Watson R. R.). pp. 139-l 72. CRC Press, Boca Raton, FL. Baker M. A. and He S. Q. (1991) Elaboration of cellular DNA brakes by hydroperoxides. Frrc~ Rad. Biol. Med. 11, 563-572. Beuttner G. R. (1993) The pecking order of free radicals and antioxidants, lipid peroxidation, alpha tocopherol and ascorbate. Arch. Biochem. Biophys. 300, 535.-543. Bulmer J. N., Pace D. and Ritson A. (1988) Immunoregulatory cells in human decidua: morphology, immunohistochemistry and function. Reprod. Nutr. Der. 28, 159991613. Clark D. A., Croy B. A., Wegmann T. G. and Chaouat G. (I 987) Immunological and para-immunological mechansims in spontaneous abortion: recent insights and future directions. J. Reprod. Immunol. 12, l-12. Davidge S. T., Hubel C. A., Brayden R. D., Capeless E. C. and McLaughlin M. K. (1992) Sera antioxidant activity

in uncomplicated and preeclamptic pregnancies. Oh.st. Gynaecol. 79, 897-901. Halliwell B. (1991) Drug antioxidant effects. A basis for drug selection? Drugs 42, 5699605. Hayes J. D. and Pulford D. J. (1995) The glutathione S-transferase supergene family: regulation of GST and the contribution of the isoenzymes to cancer chemoprotection and drug resistance. Crit. Rev. Biochem. Mol. Biol. 30, 445600. Hoffbrand A. V. and Pettit J. E. (1988) Clirticcrl Hrrmrolo~~*. pp. 1-31. Gower Medical, New York. Hunt J. S. (1989) Macrophages in human uteroplacental tissues: a review. Am. J. Reprod. Immunol. 21, 119-122. Hunt J. S. (1994) Immunologically relevant cells in the uterus. Biol. Reprod. 50, 461466. Jain S., Thomas M., Kumar G. P. and Laloraya M. (1993) Programmed lipid peroxidation of biomembranes generating linked phospholipids permitting local molecular mobility: a peroxidative theory of fluidity management. Biochem. Biophys. Res. Commun. 195, 574-580. Jensson H., Guthenberg C., Alin P. and Mannervik B. (1986) Rat glutathione transferase 8-8. an enzyme efficiently detoxifying 4-hydroxy-alk-2-enals. FEBS Lett. 203, 207-209. Krinsky N. I. (1992) Mechanism of action of biological antioxidants. Proc. Sot. Exp. Biol. Med. 200, 248-254. Kurita Y., Tsuboi R., Ueki R.. Rifkin D. B. and Ogawa H. (1992) Immunohistochemical localization of basic tibroblast growth factor in wound healing sites of mouse skin. Arch. Dermatol. Res. 284, 1933197. Lawrence R., Church J. A., Richards W. and Borzy M. (1980) Immunological mechanisms in the maintenance of pregnancy. Ann. Allergy 44, 1666173. Lea R. G. and Clark D. A. (1991) Macrophages and migratory cells in endometrium relevant to implanation. Baill. Clin. Ohsr. Gynaccol. 5, 25-59. Mannervik B. and Danielson U. H. (1988) Glutathione transferases- -structure and catalytic activity. CRC Crir. Rev. Biochem. 23, 283-337. Meister A. (1992) On the antioxidant effects of ascorbate acid and glutathione. Biochent. Phrrrmacol. 44, 19051915. Moldovan N. I., Lupu F.. Moldovan L. and Simionescu N. (1994) 4-Hydroxynonenal induces membrane perturbations and inhibition of basal prostacyclin production in endothelial cells, and migration of monocytes. Cell Biol. ht. 18, 985-992. Niki E. (1991) Vitamin C as an antioxidant. World Rep. Nutr. Diet 64, l--30. Singhal S. S., Saxena M., Ahmad H., Awasthi S., Haque A. K. and Awdsthi Y. C. (1992) Glutathione S-transferases of human lung: characterization and evaluation of the protective role of the x-class isozymes against lipid peroxidation. Arch. Biochem. Biophys. 299, 2322241. Singhal S. S., Yallampalli C., Singhal J., Piper J. T. and Awasthi S. (1996) Purification and characterization of glutathione S-transferases of rat uterus. Znt. J. Blochem. Cell Biol. 28, 1271l1283. Stenberg G., Ridderstrom M., Engstrom A., Pemble S. E. and Mannervik B. (1992) CIoning and heterologous expression of cDNA encoding class alpha rat glutathione transferase X-8. an enzyme with high catalytic activity towards genotoxic a$-unsaturated cdrbonyl compounds. Biochem. J 284, 3 13-3 19. Uotila J., Tuimala R., Aarnio T., Pyykko K. and Ahotupa M. (1991) Lipid peroxidation products, selenium-depen-

Cellular localization

of glutathione

dent glutathione peroxidase and vitamin E in normal pregnancy. Eur. J. Ohst. Gynecol. Reprod. Biol. 42, 95.-100. Uotila J., Tuimala R., Pyykko K. and Ahotupa M. (1993) Pregnancy-induced hypertension is associated with changes in maternal and umbilical blood antioxidants. G~~~ecol. Ohsr. Incesr. 36, 153-l 51. van Acker S. A. B. E., Koymans L. M. and Bast A. (1993) Molecular pharmacology of vitamin E, structural aspects of antioxidant activity. Free Rud. Biol. Med. 15, 3 I l-328. Yallampalli C., Iaumi H., Byam-Smith M. and Garfield R. E. (1994a) An L-arginineenitric oxidecyclic guanosine monophosphate system exists in the uterus and inhibits contractility during pregnancy. Am. .1. Ohsr. Gr~~rcol. 170, 175-~185.

s-transferase

813

Yallampalli C., Byam-Smith M., Nelson S. 0. and Garfield R. E. (1994b) Steroid hormones modulate the production of nitric oxide and cGMP in the rdt uterus. Endocrinology 134, 1971-1974. Zimniak P., Eckles M. A., Saxena M. and Awasthi Y. C. (1992) A subgroup of class c( glutathione S-transferases: cIoning of cDNA for mouse lung glutathione S-transferase GST 5.1. FEBS Left. 313, 173-176. Zimniak P., Singhal S. S., Srivastava S. K., Awasthi S., Sharma R., Hayden J. B. and Awasthi Y. C. (1994) Estimation of genomic complexity heterologous expression and enzymatic characterization of mouse glutathione S-transferase mGSTA4-4 (GST 5.7). J. Rio/. (‘hem. 269, 992-1000.