- Email: [email protected]
Ribonuclease from cytosolic fraction of human erythrocytes
Clinica &mica Elsevier
19
Acta, 154 (1986) 19-28
CCA 03362
Ribonuclease from cytosolic fraction of human erythrocytes Barbara
Czajkowska,
Jerzy W. Naskalski
* and Jan Sznajd
Department of Biochemical Diagnostics Medical Academy, Krakbw (Poland) (Received
March 15th, 1985; revision July 3rd. 1985)
Key words: Acid RNase; Elythrocyte
cytosol fraction; Poly -C avid RNase; Ribonuclease inactive protein complex
Ribonuclease (RNase) activity is detectable in only one third of specimens of human erythrocyte haemolysates. On the other hand, treatment of erythrocytic cytosoles with sulphosalicylic acid reveals an inhibitor-bound RNase activity which is present in all erythrocyte specimens studied. The level of the erythrocyte inhibitor-bound RNase activity is comparable to that in human lymphocytes. Isolated RNase from the cytosolic fraction of human erythrocytes is poly-C avid RNase with maximum activity at pH 6.5. The enzyme is resistant to treatment with strong acids and heating up to 95°C. Molecular filtration of the erythrocyte RNase shows that it is composed of two fractions differing in molecular mass, 19000 and 15 000. No difference in enzymic properties between these fractions was found. The general properties of erythrocyte cytosolic RNase are much like those of acid RNases of human granulocytes and lymphocytes. As the erythrocytes do not metabolize RNA no function for the inhibitor-bound RNase can be suggested. Assuming that the observed erythrocyte RNase is the residual enzyme, persisting in the cell since it was functioning in the nucleated erythrocyte precursors, one may surmise that levels of free and inhibitor-bound erythrocyte RNase activity may be related to the normality or abnormality of erythrocyte maturation.
Human mature erythrocytes are deprived of cytoplasmic structures usually containing the ribonuclease (RNase) and it is a commonly held opinion that they do not reveal any measurable RNase activity. Moreover, erythrocytes do not synthesize l
To whom Kopernika
correspondence should be addressed 15, 31-501 Krakbw, Poland.
0009-8981/86/$03.50
at:
0 1986 Elsevier Science Publishers
Katedra
Diagnostyki
B.V. (Biomedical
Biochemicznej
Division)
A.M.
Ul.
20
proteins nor metabolize nucleic acids, and have no functional need to possess any nucleases [l]. Several authors have observed some RNase in rat and rabbit mature erythrocytes [2-41. This activity was due to the presence of an RNase bound to an unidentified inhibitory protein [3,4]. Since immature erythroid cells exhibit an easily detectable RNase activity, which then disappears when the red cell matures 13-61, the question arises whether the disappearance of erythrocyte RNase may provide some information on normality or abnormality of the erythrocyte maturation process. In this report free and inhibitorbound RNase activity in normal human blood erythrocytes is described. Materials and methods
Material for the study consisted of erythrocytes isolated from fresh human blood, obtained from male and female blood donors aged 18-50 yr, and using ACD (trisodium citrate 44 mmol/l, citric acid 62 mmol/l in water glucose solution 70 mmol/l) added to blood in a 1:5 (v/v) proportion as anticoagulant. Isolation of the erythrocytes was performed within 3 h of drawing blood. The erythrocyte isolation procedure and further handling of the test materials was performed in a temperature range 0-4°C unless otherwise described. Blood specimens were centrifuged for 10 min at 900 X g. The supernatant fraction containing plasma and leukocytic buffy coat was discarded. The sediment consisting of the erythrocytes was suspended in 7 vol of NaCl(154 mmol/l) solution and again centrifuged. The NaCl wash procedure was repeated three times. The erythrocytes thus obtained were resuspended in Na-phosphate buffer solution (310 mOsm) and centrifuged at 3000 x g for 15 min. The supematant buffer solution was discarded and 7 vol of 20 mOsm Na-phosphate buffer, pH 7.4, were added. After 30 s of gentle stirring the haemolysate was centrifuged for 2 min at 9000 x g. The supernatant, which consisted of the haemolysate and the erythrocyte ghosts, was collected for further processing. The sediment was discarded. The erythrocyte supernatant was centrifuged for 30 min at 30000 x g to remove debris and the erythrocyte membrane fraction. Thus the final erythrocyte stuff was free from detectable elements of erythrocyte structure [7]. Assay of RNase activity was performed at pH 6.5 using yeast RNA as substrate. The procedure of assaying RNase activity was based on that of Anfinsen et al [8] modified for assaying low RNase activities as follows: 50 ~1 of the erythrocyte supematant and 50 ~1 of the 0.1 mol/l Na-phosphate buffer, pH 6.5, were placed on the bottom of the microcentrifuge tube and 100 ~1 of 0.6% RNA solution in the same buffer were added. The tubes were placed in a water bath at 37°C for 60 min. The tubes were transferred to an ice bath and 300 ~1 of uranyl acetate 6 mmol/l in perchloric acid 185 mmol/l was added. The precipitate, still at 0°C was centrifuged 1200 X g for 15 min and 100 ~1 aliquots of supernatant were diluted with 1 ml of distilled water. The concentration of RNA digestion products was estimated by measuring the optical density at 260 nm. Each assay was performed twice. Final values were obtained by subtracting the ‘blank’ which was obtained by omitting the 37°C water bath step, from the ‘test’.
21 The results expressed as an increase in optical density at 260 nm when plotted against the concentration of bovine pancreatic RNase exhibited a straight proportionality versus the square root of RNase concentrations [9]. RNase activity was expressed in units calculated as concentration of the standard bovine pancreatic RNase (Boehringer Mannheim) displaying the same activity as the specimen studied. One unit was defined as the activity of 1 mg/l of bovine pancreatic RNase. Phosphodiesterase activity was measured by the procedure of Sischeimer and Koerner [lo] using Ca-bis-p-nitrophenylphosphate (Sigma, London, UK). Protein concentration was determined by the Lowry et al [ll] procedure. Ion exchange chromatography of erythrocyte RNase was performed at 4°C using CM cellulose (microgranular, preswollen CM-cellulose C 52, Whatman). Prior to use, the ion exchanger was converted into the Na-salt by batch treatment with 0.5 mol/l NaOH and then equilibrated with 0.01 mol/l Na-phosphate buffer, pH 6.0. The cellulose was packed in a glass column (32 x 1.5 cm) and washed with 200 ml of the starting buffer, pH 6.0. The column flow rate was 6-7 ml/h. Gel exclusion chromatography was performed using Sephadex G-75 suspended in Na-phosphate buffer 10 mmol/l, pH 7.8 with added NaCl 600 mmol/l packed into a glass column (3.3 x 70 cm) and eluted continuously with the same buffer at 15 cm hydrostatic pressure. The column elution rate was 12 ml/h. Three to four millilitres of the concentrated RNase samples were applied to the column and 2.5 ml fracuons were collected. Isolation of the erythrocyte RNase The final erythrocyte supematant was treated with water saturated with ammonia sulphate at 4°C. To one volume of erythrocyte supematant, 1.22 volumes of saturated ammonia sulphate was added upon stirring to obtain approximately 2.1 mol/l solution, and the mixture was left to stand for 12 h, and the sediment formed containing the was centrifuged at O”, 8 000 X g for 20 min. The supematant haemoglobin was rejected. The sediment, rich in bound-inactive RNase was dissolved in 20 mOsm Na-phosphate buffer, pH 7.4 (1:3 v/v) and used for further study. To liberate the bound RNase the dissolved sediment was treated with sulphosalicylic acid (SSA) 1 mol/l to obtain pH 2.0 (approximately 0.1 vol of 1 mol/l SSA to 1 vol of the solution). The mixture was neutralised by microtitration with NaOH 1 mol/l, and centrifuged for 3 min at 15 000 x g at room temperature. The supernatant exhibiting easily detectable RNase activity was used for further study. The solution of released RNase was dialysed against Na-phosphate buffer 10 mmol/l, pH 6.0, and applied to the CM cellulose C 52 column as described above. The RNase was retained in the column. The retained RNase was eluted with Na-phosphate buffer 10 mmol/l, pH 6.0, with linearily increasing NaCl concentrations (O-600 mmol/l) [12]. RNase was eluted from the column as a single peak of activity at NaCl 370 mmol/l (Fig. 1). The fractions displaying RNase activity were collected and concentrated to one tenth of the initial volume with an Aquacide II dehydrator (Calbiochem).
Effluent
volume
ml
Fig. 1. Chromatography of human erythrocyte RNase on CM cellulose C-52 at pH 6.0. Heavy solid line and circles represent RNase activity measured at pH 6.5. Thin solid line represents protein concentration measured as light absorption at 280 nm. Slanted broken line represents NaCl concentration gradient.
Concentrated RNase solution was divided into 3- to 4-ml portions which were applied to the Sephadex G-75 column. RNase eluted from the column as one peak of activity with a small shoulder closely following the main peak (Fig. 2). The fractions from the main peak area and the declining slope of the shoulder were collected separately and designated as RNase E, and E,, respectively. These preparations of erythrocyte RNase were subjected to further study (Table I). Properties of isolated erythrocyte RNase Assuming no interactions between the Sephadex gel matrix and the RNase molecules the calculated molecular masses of RNase E, and E, and 19ooO and 15 000, respectively. 1.5 i
250
300
350 Effluent
LCQ volume
ml
Fig. 2. Chromatography of the erythrocyte RNase on Sephadex G-75 column. Heavy solid line and circles represent RNase activity. The arrows show elution volumes of human albumin and hog heart cytochrome c used as molecular mass standards.
23 TABLE Summary
I of ribonuclease
Procedure
Erythrocyte haemolysate (NH,),SO, precipitation and SSA treatment Concentration with Aquacide II CM cellulose column chromatography Sephadex G-75 column chromatography RNase E, h RNase E, ’
isolation
procedure
from human
erythrocytes
Total RNase activity (mu)
Purification ratio
Recovery (W
1
27500 a
loo =
24500
160
89
21ooo
_
76
16500
2400
52
9600 3300
2ooo 3ooo
20 7
a Calculated from RNase activity obtained h Mean of the four runs of the procedure.
upon haemolysate
treatment
with SSA.
Both erythrocyte RNases showed the same maximum of enzyme activity at pH 6.5 with an ability to degrade yeast RNA in a pH range 4.1-8.0. Heating of the erythrocyte RNases to 95°C at pH 6.5 caused 50% decrease in enzyme activity in approximately 10 min. Treatment of the isolated erythrocyte RNases with HCl, H,SO, and HClO, (100 mmol/l) at room temperature for 10 min did not cause any significant inactivation of the enzyme, however, combined heating 95°C and acid treatment produced approximately 80% inactivation in l-2 min for both RNase fractions. Therefore, performed study of the properties of RNases E, and E, showed no difference between these two fractions. It is conceivable that they are two molecular forms of the same enzyme differing in the molecule fragments which do not contribute to the enzymic properties of the protein [13]. The presence of EDTA 1 and 3 mmol/l in the RNase reaction media and pCMB 1 mmol/l had no effect on enzyme activity. The Cu2+ and Mg*+ ions at concentration of 300 mmol/l decreased the erythrocyte RNase activity of 78 and 65%, respectively. The inhibitory action of Cu*+ and Mg*+ ions was abolished upon EDTA 3 mmol/l treatment. Zn*’ and Mn*’ at concentrations of 300 mmol/l had no effect on enzyme activity. The most readily degraded homopolynucleotide was the poly-C (Table II). Poly-U was also degraded, however, at only two-thirds of the poly-C rate. No degradation of poly-A, poly-G, poly-AG nor Ca-bis-p-nitrophenylphosphate was observed. RNase activity in human etythrocyte
cytosol fraction
RNase activity was detected only in seven out of twenty haemolysates. The observed free RNase activity was in the range of 0.6-2.0 mU (Table III). The same
24
TABLE
II
Erythrocyte
cytosolic
Polynucleotide Relative reaction rate a
RNase
ability
to degrade
of synthetic
homopolynucleotides
Poly-c
Poly-u
Poly-A
Poly-G
Poly-AG
1
0.73
0
0
0
a Relative reaction rate was calculated as a ratio of the amount of nucleotide liberated respective polynucleotide to the amount of cytydylic acid liberated from poly-C.
in 1 min from the
haemolysates treated with SSA all showed some RNase activity. These activities are presented in Table III. The level of bound erythrocyte RNase activity in human erythrocyte was close to that in healthy and leukemic human lymphocytes [9]. No correlation between the sex and age of blood donors and erythrocyte RNase activity was observed.
TABLE
III
Free and bound Subject
ribonuclease Sex
in cytosol Age
fraction
Free RNase (mU/mg protein)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 a Calculated b Calculated
m m f f m m f m m m m m E” f f f m f m per mg protein per mg protein
26 18 42 36 31 29 23 37 31 40 23 22 39 50 35 29 32 23 27 34
of healthy
1.5 0 0
0 1.4 2.0 0 0 0.5 1.1 0 0.6 0 0.7 0 0 0 0 0 0
human
erythrocytes
Total RNase after SSA treatment (mU/mg protein) a
600 720 330 610 690 860 1420 250 1190 920 270 780 1710 1620 350 620 250 410 320 220
remaining after SSA treatment. in haemolysate before SSA treatment.
Bound RNase (mU/mg protein) 2.4 2.9 1.9 2.4 1.7 1.5 2.4 5.8 5.4 1.8 1.2 9.2 2.6 7.3 1.8 1.7 5.8 5.2 1.8 7.3
b
25
The erythrocyte protein fraction obtained by precipitation with ammonia sulphate 2.1 mol/l, rich in bound-inactive RNase, displayed a concentration of bound RNase approximately ten times higher than the whole haemolysate. Discussion The results obtained in this study show that erythrocytes contain RNase. The properties of the erythrocyte RNase show that this is the poly-C avid, thermostable RNase similar to those isolated from human granulocytes [14,15] and lymphocytes [9]. The general properties of the human erythrocytes RNase are similar to those of spleen acid RNase [16]. The erythroid precursor cells, by comparison with the mature erythrocytes, displayed easily detectable RNase activity [2-6, 18-211, which decreases and finally disappears while the reticulocyte transforms into a mature erythrocyte [3-6,211. RNase activity in matured erythrocytes has been noticed by Rost [21]. Later Kraft and Shortman [17] reported the presence of RNase and RNase inhibitor in rat erythrocytes. The inhibitor bound RNase was also found in rat reticulocytes [4], heme being found to be a weak RNase inhibitor [2]. In a more extensive study Burka [3,4] described rabbit erythrocyte RNases which existed both in free cytosolic and inhibitor bound forms. The RNase inhibitor complex was dissociated upon treatment with 4 mol/l urea [3,4]. The membrane-bound RNase in rabbit erythroid cells was also found [3]. The results of our study of human erythrocyte RNase are in accord with the quoted data on the appearance of RNase in rat and rabbit erythrocytes. Approximately, two-thirds of the human erythrocyte specimens studied did not display any measurable RNase activity (Table I). All erythrocyte specimens possessed inactive protein bound RNase, liberation of this enzyme requiring the use of procedures producing denaturation of the inhibitory protein which is less stable than the RNase during treatment with acids and high temperature. Neuwelt et al [16], studying physicochemical and immunological properties of the RNase from human cells and body fluids, suggested the existence of two general forms of this enzyme: alkaline RNase, similar to the pancreatic RNase with predominant activity against poly-C and maximum activity at pH 7.8; and acid RNase of the spleen type with predominant activity against poly-C and maximum activity at pH 6.6. Because the erythrocytes consist of more than 60% of the cellular mass of the unperfused spleen, and the procedure of the acid extraction of ribonucleases from spleen homogenates most probably liberates the erythrocyte RNase along with RNases from the spleen specific constituents, the ‘spleen RNase’ should consist of a mixture of the erythrocyte RNase with very similar RNases originating from spleen lymphoid cells and phagocytes. Therefore, to describe the properties of human RNases, well characterized neutrophil RNase [14,15] or lymphocyte RNase [9] should be used as a primer, rather than spleen RNase, which probably originates from several cellular species. The function of intracellular RNase consists of degradation of some fragments of newly synthesized ribonucleic acid molecules and degradation of RNA no longer
26
necessary for the cell [l]. RNases of lysosomal type also participate in digestion of RNA engulfed into the phagosome. Since mature erythrocytes do not metabolize RNA nor synthesize protein, no specific function of the erythrocyte free and bound RNase could be suggested. One may surmise that binding of RNase to an inhibitor is a step in the RNase removal process from the cell. A similar process was suggested for plasma alkaline RNase which is taken up by liver cells, irreversibly bound to the RNase inhibitory protein and digested by proteolytic enzymes [22]. Therefore, RNase described in this paper may be the residual enzyme deprived of any specific function in the cell. On the other hand, the activity of the free and inhibitor-bound erythrocyte RNase may reflect some phenomena still not studied and related to maturation and duration time of the erythrocyte in the blood. Acknowledgement
The authors are greatly indebted to Miss Anna Okrajni for her excellent technical assistance. References 1 Levy CC. Roles of RNases in cellular regulatory mechanisms. Life Sci 1985; 17: 311-316. 2 Farkas W, Marks FA. Partial purification and properties of a ribonuclease from rabbit reticulocytes. J Biol Chem 1968; 243: 6464-6473. 3 Burka ER. Erythroid cell RNase: activation by urea and localization to the cell membrane. J Clin Invest 1971; 50: 60-68. 4 Burka ER. RNase activity in erythroid cell lysates. J Clin Invest 1969; 48: 1724-1730. 5 Hulea SA, Denton MJ, Amstein HVR. Ribonuclease activity during erythroid cell maturation. FEBS Lett 1975; 51: 346-350. 6 Hulea SA, Arstein HRV. Intracellular distribution of RNase during erythroid cell maturation. Biochem Sot Tram 1975; 3: 911-916. 7 Dodge JT, Mitchel C, Hanhan DJ. The preparation and chemical characteristics of hemoglobin-free ghosts of human erythrocytes. Arch Biochem Biophys 1963; 100: 119-130. 8 Anfinsen CB, Redfields RE, Choate WL. Page J, Carrol WR. Studies on the gross structure, cross linkage and terminal sequences of ribonuclease. J Biol Chem 1954; 207: 201-210. 9 Celihski A, Naskalski J, Sznajd J. Ribonuclease deficiency in lymphocytes of patients with chronic lymphocytic leukemia (CLL). Folia Haemat (Leipzig) 1982; 109: 201-212. 10 Sischeimer RL, Koemer JF. Purification of venom phosphodiesterase. J Biol Chem 1952; 198: 293-296. 11 Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurement with the Folin phenol reagent. J Biol Chem 1951; 193: 265-275. 12 Naskalski J. A comparison of the isoenzymic patterns of urinary ribonucleases in healthy subjects and patients with chronic granulocytic leukemia. Polish Med J 1972; 11: 827-834. 13 Beitema JJ, Gaestra W, Scheffer AJ, Weeling GW. Carbohydrate in pancreatic ribonucleases. Eur J Biochem 1976; 63: 441-448. 14 Akagi K, Yamanaka M, Murai K. Omahe T. Purification and properties of acid ribonucleases in human serum and leukocytes. Cancer Res 1978; 38: 2163-2167. 15 Sznajd J, Naskalski JW. Ribonuclease from human granulocytes. Biochim Biophys Acta 1973; 302: 282-292. 16 Neuwelt EA, Boguski MJ, Frank JJ, Procter-Appich K, Levy CC. Possible sites of origin of human plasma ribonucleases as evidenced by isolation and partial characterization of ribonucleases from several human tissues. Cancer Res 1978; 38: 88-93.
21 17 Kraft N, Shortman K. A suggested control function for the animal tissue ribonuclease-ribonuclease inhibitor system, based on studies of isolated cells and phytohaemagglutinin-transformed lymphocytes. Biochim Biophys Acta 1970; 217: 164-175. 18 Goto S, Mizuno D. Relationship between RNase and its corresponding inhibitor in rat reticulocytes. Arch Bicchem Biophys 1971; 145: 71-77. 19 Izaak G, Karsai A, Eylon E, Hershla CH. Ribonucleic acid production and breakdown in synchronised erythroid cohortes from actinomycin treated animals. J Lab Clin Med 1971; 77: 923-930. 20 Rowley PT, Barnes F. A reticulocyte RNase activity declining with cell maturation. Fed Proc 1966; 25: 645-656. 21 Rost G, Hahn Ch. Eigenschaften und Verkommen eines Ribonuklease Hemmestoffes in stromafrien Hemolysat rater Blutkorperchen. Acta Biol Med (FRG) 1959; 3: 276-283. 22 Bartholenys J, Peter-Joris C, Baudhin P. Hepatic nucleases. Extrahepatic origin and association of neutral liver ribonuclease with lysosomes. Eur J Biochem 1975; 60: 385-393.
19
Acta, 154 (1986) 19-28
CCA 03362
Ribonuclease from cytosolic fraction of human erythrocytes Barbara
Czajkowska,
Jerzy W. Naskalski
* and Jan Sznajd
Department of Biochemical Diagnostics Medical Academy, Krakbw (Poland) (Received
March 15th, 1985; revision July 3rd. 1985)
Key words: Acid RNase; Elythrocyte
cytosol fraction; Poly -C avid RNase; Ribonuclease inactive protein complex
Ribonuclease (RNase) activity is detectable in only one third of specimens of human erythrocyte haemolysates. On the other hand, treatment of erythrocytic cytosoles with sulphosalicylic acid reveals an inhibitor-bound RNase activity which is present in all erythrocyte specimens studied. The level of the erythrocyte inhibitor-bound RNase activity is comparable to that in human lymphocytes. Isolated RNase from the cytosolic fraction of human erythrocytes is poly-C avid RNase with maximum activity at pH 6.5. The enzyme is resistant to treatment with strong acids and heating up to 95°C. Molecular filtration of the erythrocyte RNase shows that it is composed of two fractions differing in molecular mass, 19000 and 15 000. No difference in enzymic properties between these fractions was found. The general properties of erythrocyte cytosolic RNase are much like those of acid RNases of human granulocytes and lymphocytes. As the erythrocytes do not metabolize RNA no function for the inhibitor-bound RNase can be suggested. Assuming that the observed erythrocyte RNase is the residual enzyme, persisting in the cell since it was functioning in the nucleated erythrocyte precursors, one may surmise that levels of free and inhibitor-bound erythrocyte RNase activity may be related to the normality or abnormality of erythrocyte maturation.
Human mature erythrocytes are deprived of cytoplasmic structures usually containing the ribonuclease (RNase) and it is a commonly held opinion that they do not reveal any measurable RNase activity. Moreover, erythrocytes do not synthesize l
To whom Kopernika
correspondence should be addressed 15, 31-501 Krakbw, Poland.
0009-8981/86/$03.50
at:
0 1986 Elsevier Science Publishers
Katedra
Diagnostyki
B.V. (Biomedical
Biochemicznej
Division)
A.M.
Ul.
20
proteins nor metabolize nucleic acids, and have no functional need to possess any nucleases [l]. Several authors have observed some RNase in rat and rabbit mature erythrocytes [2-41. This activity was due to the presence of an RNase bound to an unidentified inhibitory protein [3,4]. Since immature erythroid cells exhibit an easily detectable RNase activity, which then disappears when the red cell matures 13-61, the question arises whether the disappearance of erythrocyte RNase may provide some information on normality or abnormality of the erythrocyte maturation process. In this report free and inhibitorbound RNase activity in normal human blood erythrocytes is described. Materials and methods
Material for the study consisted of erythrocytes isolated from fresh human blood, obtained from male and female blood donors aged 18-50 yr, and using ACD (trisodium citrate 44 mmol/l, citric acid 62 mmol/l in water glucose solution 70 mmol/l) added to blood in a 1:5 (v/v) proportion as anticoagulant. Isolation of the erythrocytes was performed within 3 h of drawing blood. The erythrocyte isolation procedure and further handling of the test materials was performed in a temperature range 0-4°C unless otherwise described. Blood specimens were centrifuged for 10 min at 900 X g. The supernatant fraction containing plasma and leukocytic buffy coat was discarded. The sediment consisting of the erythrocytes was suspended in 7 vol of NaCl(154 mmol/l) solution and again centrifuged. The NaCl wash procedure was repeated three times. The erythrocytes thus obtained were resuspended in Na-phosphate buffer solution (310 mOsm) and centrifuged at 3000 x g for 15 min. The supematant buffer solution was discarded and 7 vol of 20 mOsm Na-phosphate buffer, pH 7.4, were added. After 30 s of gentle stirring the haemolysate was centrifuged for 2 min at 9000 x g. The supernatant, which consisted of the haemolysate and the erythrocyte ghosts, was collected for further processing. The sediment was discarded. The erythrocyte supernatant was centrifuged for 30 min at 30000 x g to remove debris and the erythrocyte membrane fraction. Thus the final erythrocyte stuff was free from detectable elements of erythrocyte structure [7]. Assay of RNase activity was performed at pH 6.5 using yeast RNA as substrate. The procedure of assaying RNase activity was based on that of Anfinsen et al [8] modified for assaying low RNase activities as follows: 50 ~1 of the erythrocyte supematant and 50 ~1 of the 0.1 mol/l Na-phosphate buffer, pH 6.5, were placed on the bottom of the microcentrifuge tube and 100 ~1 of 0.6% RNA solution in the same buffer were added. The tubes were placed in a water bath at 37°C for 60 min. The tubes were transferred to an ice bath and 300 ~1 of uranyl acetate 6 mmol/l in perchloric acid 185 mmol/l was added. The precipitate, still at 0°C was centrifuged 1200 X g for 15 min and 100 ~1 aliquots of supernatant were diluted with 1 ml of distilled water. The concentration of RNA digestion products was estimated by measuring the optical density at 260 nm. Each assay was performed twice. Final values were obtained by subtracting the ‘blank’ which was obtained by omitting the 37°C water bath step, from the ‘test’.
21 The results expressed as an increase in optical density at 260 nm when plotted against the concentration of bovine pancreatic RNase exhibited a straight proportionality versus the square root of RNase concentrations [9]. RNase activity was expressed in units calculated as concentration of the standard bovine pancreatic RNase (Boehringer Mannheim) displaying the same activity as the specimen studied. One unit was defined as the activity of 1 mg/l of bovine pancreatic RNase. Phosphodiesterase activity was measured by the procedure of Sischeimer and Koerner [lo] using Ca-bis-p-nitrophenylphosphate (Sigma, London, UK). Protein concentration was determined by the Lowry et al [ll] procedure. Ion exchange chromatography of erythrocyte RNase was performed at 4°C using CM cellulose (microgranular, preswollen CM-cellulose C 52, Whatman). Prior to use, the ion exchanger was converted into the Na-salt by batch treatment with 0.5 mol/l NaOH and then equilibrated with 0.01 mol/l Na-phosphate buffer, pH 6.0. The cellulose was packed in a glass column (32 x 1.5 cm) and washed with 200 ml of the starting buffer, pH 6.0. The column flow rate was 6-7 ml/h. Gel exclusion chromatography was performed using Sephadex G-75 suspended in Na-phosphate buffer 10 mmol/l, pH 7.8 with added NaCl 600 mmol/l packed into a glass column (3.3 x 70 cm) and eluted continuously with the same buffer at 15 cm hydrostatic pressure. The column elution rate was 12 ml/h. Three to four millilitres of the concentrated RNase samples were applied to the column and 2.5 ml fracuons were collected. Isolation of the erythrocyte RNase The final erythrocyte supematant was treated with water saturated with ammonia sulphate at 4°C. To one volume of erythrocyte supematant, 1.22 volumes of saturated ammonia sulphate was added upon stirring to obtain approximately 2.1 mol/l solution, and the mixture was left to stand for 12 h, and the sediment formed containing the was centrifuged at O”, 8 000 X g for 20 min. The supematant haemoglobin was rejected. The sediment, rich in bound-inactive RNase was dissolved in 20 mOsm Na-phosphate buffer, pH 7.4 (1:3 v/v) and used for further study. To liberate the bound RNase the dissolved sediment was treated with sulphosalicylic acid (SSA) 1 mol/l to obtain pH 2.0 (approximately 0.1 vol of 1 mol/l SSA to 1 vol of the solution). The mixture was neutralised by microtitration with NaOH 1 mol/l, and centrifuged for 3 min at 15 000 x g at room temperature. The supernatant exhibiting easily detectable RNase activity was used for further study. The solution of released RNase was dialysed against Na-phosphate buffer 10 mmol/l, pH 6.0, and applied to the CM cellulose C 52 column as described above. The RNase was retained in the column. The retained RNase was eluted with Na-phosphate buffer 10 mmol/l, pH 6.0, with linearily increasing NaCl concentrations (O-600 mmol/l) [12]. RNase was eluted from the column as a single peak of activity at NaCl 370 mmol/l (Fig. 1). The fractions displaying RNase activity were collected and concentrated to one tenth of the initial volume with an Aquacide II dehydrator (Calbiochem).
Effluent
volume
ml
Fig. 1. Chromatography of human erythrocyte RNase on CM cellulose C-52 at pH 6.0. Heavy solid line and circles represent RNase activity measured at pH 6.5. Thin solid line represents protein concentration measured as light absorption at 280 nm. Slanted broken line represents NaCl concentration gradient.
Concentrated RNase solution was divided into 3- to 4-ml portions which were applied to the Sephadex G-75 column. RNase eluted from the column as one peak of activity with a small shoulder closely following the main peak (Fig. 2). The fractions from the main peak area and the declining slope of the shoulder were collected separately and designated as RNase E, and E,, respectively. These preparations of erythrocyte RNase were subjected to further study (Table I). Properties of isolated erythrocyte RNase Assuming no interactions between the Sephadex gel matrix and the RNase molecules the calculated molecular masses of RNase E, and E, and 19ooO and 15 000, respectively. 1.5 i
250
300
350 Effluent
LCQ volume
ml
Fig. 2. Chromatography of the erythrocyte RNase on Sephadex G-75 column. Heavy solid line and circles represent RNase activity. The arrows show elution volumes of human albumin and hog heart cytochrome c used as molecular mass standards.
23 TABLE Summary
I of ribonuclease
Procedure
Erythrocyte haemolysate (NH,),SO, precipitation and SSA treatment Concentration with Aquacide II CM cellulose column chromatography Sephadex G-75 column chromatography RNase E, h RNase E, ’
isolation
procedure
from human
erythrocytes
Total RNase activity (mu)
Purification ratio
Recovery (W
1
27500 a
loo =
24500
160
89
21ooo
_
76
16500
2400
52
9600 3300
2ooo 3ooo
20 7
a Calculated from RNase activity obtained h Mean of the four runs of the procedure.
upon haemolysate
treatment
with SSA.
Both erythrocyte RNases showed the same maximum of enzyme activity at pH 6.5 with an ability to degrade yeast RNA in a pH range 4.1-8.0. Heating of the erythrocyte RNases to 95°C at pH 6.5 caused 50% decrease in enzyme activity in approximately 10 min. Treatment of the isolated erythrocyte RNases with HCl, H,SO, and HClO, (100 mmol/l) at room temperature for 10 min did not cause any significant inactivation of the enzyme, however, combined heating 95°C and acid treatment produced approximately 80% inactivation in l-2 min for both RNase fractions. Therefore, performed study of the properties of RNases E, and E, showed no difference between these two fractions. It is conceivable that they are two molecular forms of the same enzyme differing in the molecule fragments which do not contribute to the enzymic properties of the protein [13]. The presence of EDTA 1 and 3 mmol/l in the RNase reaction media and pCMB 1 mmol/l had no effect on enzyme activity. The Cu2+ and Mg*+ ions at concentration of 300 mmol/l decreased the erythrocyte RNase activity of 78 and 65%, respectively. The inhibitory action of Cu*+ and Mg*+ ions was abolished upon EDTA 3 mmol/l treatment. Zn*’ and Mn*’ at concentrations of 300 mmol/l had no effect on enzyme activity. The most readily degraded homopolynucleotide was the poly-C (Table II). Poly-U was also degraded, however, at only two-thirds of the poly-C rate. No degradation of poly-A, poly-G, poly-AG nor Ca-bis-p-nitrophenylphosphate was observed. RNase activity in human etythrocyte
cytosol fraction
RNase activity was detected only in seven out of twenty haemolysates. The observed free RNase activity was in the range of 0.6-2.0 mU (Table III). The same
24
TABLE
II
Erythrocyte
cytosolic
Polynucleotide Relative reaction rate a
RNase
ability
to degrade
of synthetic
homopolynucleotides
Poly-c
Poly-u
Poly-A
Poly-G
Poly-AG
1
0.73
0
0
0
a Relative reaction rate was calculated as a ratio of the amount of nucleotide liberated respective polynucleotide to the amount of cytydylic acid liberated from poly-C.
in 1 min from the
haemolysates treated with SSA all showed some RNase activity. These activities are presented in Table III. The level of bound erythrocyte RNase activity in human erythrocyte was close to that in healthy and leukemic human lymphocytes [9]. No correlation between the sex and age of blood donors and erythrocyte RNase activity was observed.
TABLE
III
Free and bound Subject
ribonuclease Sex
in cytosol Age
fraction
Free RNase (mU/mg protein)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 a Calculated b Calculated
m m f f m m f m m m m m E” f f f m f m per mg protein per mg protein
26 18 42 36 31 29 23 37 31 40 23 22 39 50 35 29 32 23 27 34
of healthy
1.5 0 0
0 1.4 2.0 0 0 0.5 1.1 0 0.6 0 0.7 0 0 0 0 0 0
human
erythrocytes
Total RNase after SSA treatment (mU/mg protein) a
600 720 330 610 690 860 1420 250 1190 920 270 780 1710 1620 350 620 250 410 320 220
remaining after SSA treatment. in haemolysate before SSA treatment.
Bound RNase (mU/mg protein) 2.4 2.9 1.9 2.4 1.7 1.5 2.4 5.8 5.4 1.8 1.2 9.2 2.6 7.3 1.8 1.7 5.8 5.2 1.8 7.3
b
25
The erythrocyte protein fraction obtained by precipitation with ammonia sulphate 2.1 mol/l, rich in bound-inactive RNase, displayed a concentration of bound RNase approximately ten times higher than the whole haemolysate. Discussion The results obtained in this study show that erythrocytes contain RNase. The properties of the erythrocyte RNase show that this is the poly-C avid, thermostable RNase similar to those isolated from human granulocytes [14,15] and lymphocytes [9]. The general properties of the human erythrocytes RNase are similar to those of spleen acid RNase [16]. The erythroid precursor cells, by comparison with the mature erythrocytes, displayed easily detectable RNase activity [2-6, 18-211, which decreases and finally disappears while the reticulocyte transforms into a mature erythrocyte [3-6,211. RNase activity in matured erythrocytes has been noticed by Rost [21]. Later Kraft and Shortman [17] reported the presence of RNase and RNase inhibitor in rat erythrocytes. The inhibitor bound RNase was also found in rat reticulocytes [4], heme being found to be a weak RNase inhibitor [2]. In a more extensive study Burka [3,4] described rabbit erythrocyte RNases which existed both in free cytosolic and inhibitor bound forms. The RNase inhibitor complex was dissociated upon treatment with 4 mol/l urea [3,4]. The membrane-bound RNase in rabbit erythroid cells was also found [3]. The results of our study of human erythrocyte RNase are in accord with the quoted data on the appearance of RNase in rat and rabbit erythrocytes. Approximately, two-thirds of the human erythrocyte specimens studied did not display any measurable RNase activity (Table I). All erythrocyte specimens possessed inactive protein bound RNase, liberation of this enzyme requiring the use of procedures producing denaturation of the inhibitory protein which is less stable than the RNase during treatment with acids and high temperature. Neuwelt et al [16], studying physicochemical and immunological properties of the RNase from human cells and body fluids, suggested the existence of two general forms of this enzyme: alkaline RNase, similar to the pancreatic RNase with predominant activity against poly-C and maximum activity at pH 7.8; and acid RNase of the spleen type with predominant activity against poly-C and maximum activity at pH 6.6. Because the erythrocytes consist of more than 60% of the cellular mass of the unperfused spleen, and the procedure of the acid extraction of ribonucleases from spleen homogenates most probably liberates the erythrocyte RNase along with RNases from the spleen specific constituents, the ‘spleen RNase’ should consist of a mixture of the erythrocyte RNase with very similar RNases originating from spleen lymphoid cells and phagocytes. Therefore, to describe the properties of human RNases, well characterized neutrophil RNase [14,15] or lymphocyte RNase [9] should be used as a primer, rather than spleen RNase, which probably originates from several cellular species. The function of intracellular RNase consists of degradation of some fragments of newly synthesized ribonucleic acid molecules and degradation of RNA no longer
26
necessary for the cell [l]. RNases of lysosomal type also participate in digestion of RNA engulfed into the phagosome. Since mature erythrocytes do not metabolize RNA nor synthesize protein, no specific function of the erythrocyte free and bound RNase could be suggested. One may surmise that binding of RNase to an inhibitor is a step in the RNase removal process from the cell. A similar process was suggested for plasma alkaline RNase which is taken up by liver cells, irreversibly bound to the RNase inhibitory protein and digested by proteolytic enzymes [22]. Therefore, RNase described in this paper may be the residual enzyme deprived of any specific function in the cell. On the other hand, the activity of the free and inhibitor-bound erythrocyte RNase may reflect some phenomena still not studied and related to maturation and duration time of the erythrocyte in the blood. Acknowledgement
The authors are greatly indebted to Miss Anna Okrajni for her excellent technical assistance. References 1 Levy CC. Roles of RNases in cellular regulatory mechanisms. Life Sci 1985; 17: 311-316. 2 Farkas W, Marks FA. Partial purification and properties of a ribonuclease from rabbit reticulocytes. J Biol Chem 1968; 243: 6464-6473. 3 Burka ER. Erythroid cell RNase: activation by urea and localization to the cell membrane. J Clin Invest 1971; 50: 60-68. 4 Burka ER. RNase activity in erythroid cell lysates. J Clin Invest 1969; 48: 1724-1730. 5 Hulea SA, Denton MJ, Amstein HVR. Ribonuclease activity during erythroid cell maturation. FEBS Lett 1975; 51: 346-350. 6 Hulea SA, Arstein HRV. Intracellular distribution of RNase during erythroid cell maturation. Biochem Sot Tram 1975; 3: 911-916. 7 Dodge JT, Mitchel C, Hanhan DJ. The preparation and chemical characteristics of hemoglobin-free ghosts of human erythrocytes. Arch Biochem Biophys 1963; 100: 119-130. 8 Anfinsen CB, Redfields RE, Choate WL. Page J, Carrol WR. Studies on the gross structure, cross linkage and terminal sequences of ribonuclease. J Biol Chem 1954; 207: 201-210. 9 Celihski A, Naskalski J, Sznajd J. Ribonuclease deficiency in lymphocytes of patients with chronic lymphocytic leukemia (CLL). Folia Haemat (Leipzig) 1982; 109: 201-212. 10 Sischeimer RL, Koemer JF. Purification of venom phosphodiesterase. J Biol Chem 1952; 198: 293-296. 11 Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurement with the Folin phenol reagent. J Biol Chem 1951; 193: 265-275. 12 Naskalski J. A comparison of the isoenzymic patterns of urinary ribonucleases in healthy subjects and patients with chronic granulocytic leukemia. Polish Med J 1972; 11: 827-834. 13 Beitema JJ, Gaestra W, Scheffer AJ, Weeling GW. Carbohydrate in pancreatic ribonucleases. Eur J Biochem 1976; 63: 441-448. 14 Akagi K, Yamanaka M, Murai K. Omahe T. Purification and properties of acid ribonucleases in human serum and leukocytes. Cancer Res 1978; 38: 2163-2167. 15 Sznajd J, Naskalski JW. Ribonuclease from human granulocytes. Biochim Biophys Acta 1973; 302: 282-292. 16 Neuwelt EA, Boguski MJ, Frank JJ, Procter-Appich K, Levy CC. Possible sites of origin of human plasma ribonucleases as evidenced by isolation and partial characterization of ribonucleases from several human tissues. Cancer Res 1978; 38: 88-93.
21 17 Kraft N, Shortman K. A suggested control function for the animal tissue ribonuclease-ribonuclease inhibitor system, based on studies of isolated cells and phytohaemagglutinin-transformed lymphocytes. Biochim Biophys Acta 1970; 217: 164-175. 18 Goto S, Mizuno D. Relationship between RNase and its corresponding inhibitor in rat reticulocytes. Arch Bicchem Biophys 1971; 145: 71-77. 19 Izaak G, Karsai A, Eylon E, Hershla CH. Ribonucleic acid production and breakdown in synchronised erythroid cohortes from actinomycin treated animals. J Lab Clin Med 1971; 77: 923-930. 20 Rowley PT, Barnes F. A reticulocyte RNase activity declining with cell maturation. Fed Proc 1966; 25: 645-656. 21 Rost G, Hahn Ch. Eigenschaften und Verkommen eines Ribonuklease Hemmestoffes in stromafrien Hemolysat rater Blutkorperchen. Acta Biol Med (FRG) 1959; 3: 276-283. 22 Bartholenys J, Peter-Joris C, Baudhin P. Hepatic nucleases. Extrahepatic origin and association of neutral liver ribonuclease with lysosomes. Eur J Biochem 1975; 60: 385-393.