Seminalplasmin, a bovine seminal plasma protein, lyses dividing but not resting mammalian cells

ELSEVIER

Biochi~ic~a et Biophysica A~ta

Biochimicaet BiophysicaActa 1221 (1994) 109-114

Seminalplasmin, a bovine seminal plasma protein, lyses dividing but not resting mammalian cells T. R a m a k r i s h n a

Murti

a, S u s h i l

A . C h a n d a n i a, A s h o k K h a r P u s h p a M . B h a r g a v a b,.

a

G. Laxma Reddy

a

a Centrefor Cellularand MolecularBiology, Hyderabad500 007, India b Indian Institute of Chemical Technology, Hyderabad500 007, India (Received 12 August 1992)

Abstract

Seminalplasmin, an antimicrobial and transcription-inhibitory protein of bovine seminal plasma, is shown to lyse dividing mammalian cells in vitro. It lyses cells in culture such as CHO, Vero, HeLa and L929. It also lyses regenerating rat liver parenchymal cells and cells of two ascitic tumours of rat - the Zajdela ascitic hepatoma and the AK-5. However, it does not lyse resting cells such as adult liver parenchymal cells, erythrocytes, or resting lymphocytes, though it binds to their cell surface. It can be used, therefore, to distinguish cells that are in the division cycle from cells that are in the resting phase. The cell-lytic activity of seminalplasmin is inhibited by Ca 2+.

Key words: Seminal plasmin; Mammalian cell lysis; Calcium ion

1. Introduction

Antibacterial peptides are widely distributed in the animal kingdom [1,2]. Many of these peptides not only possess antibacterial activity but also exhibit lytic activity against eukaryotic cells [3,4]. Cecropins, magainins, melittin and some other lytic peptides probably act by a mechanism that depends on their ability to form amphipathic alpha-helices that insert into membranes. Seminalplasmin (SPLN), a 47-amino-acid residue protein isolated from bovine seminal plasma, has previously been shown to possess potent antibacterial activity; it inhibited the growth of many gram-positive and gram-negative bacteria [5,6]. SPLN has also been shown to have bacteriolytic activity at concentrations that are higher than those required for bactericidal activity [7]. It associates with model membranes of phosphatidyl choline and phosphatidic acid [8] and increases the fluidity of sperm acrosomal and plasma membranes [9]. In this paper, we show that SPLN also lyses several

* Corresponding author. 0167-4889/94/$07.00 © 1994 Elsevier Science B.V. All rights reserved SSDI 0167-4889(93)E0182-Y

types of mammalian cells. It appears that it lyses only cells that are in the division cycle and not resting ceils.

2. Materials and methods

Lipopolysaccharide (LPS) was obtained from Sigma, St. Louis, MO, USA, and [3H]thymidine (15 Ci/mmol) from BARC, Bombay. Natural SPLN (n-SPLN) was prepared according to the method of Reddy and Bhargava [5] and further purified by high-performance liquid chromatography (HPLC) using a C-18 column. Synthetic SPLN (s-SPLN) was prepared using an Applied Biosystems Model 431A peptide synthesizer; all amino-acid derivatives and reagents were purchased from Applied Biosystems. Crude s-SPLN was purified by HPLC using a Waters semipreparative C-18 /~Bondapak column. The chemical identity of s-SPLN with n-SPLN was established by determination of amino-acid composition on a LKB 415i alpha-plus amino-acid analyzer, sequencing on a solid-phase sequenator, and by comparison of biological properties as described earlier for n-SPLN [10,11]. Details of the synthesis will be published elsewhere.

110

T. Ramakrishna Murti et al. / Biochirnica et Biophysica Acta 1221 (1994) 109-114

The peptides, AKLANRLANP and LANRLANP, representing amino acids 19-28 and 21-28 of the Nterminal of SPLN, were synthesized by the solid-phase method [12].

Cells CHO, Vero, HeLa and L-929 cells, and a primary culture of rat skin fibroblasts (RSF), were grown in Dulbecco's minimal essential medium (DMEM) containing 5% foetal calf serum (FCS). Cells were maintained at 37°C in air with 5% CO2 and 95% humidity. Lymphocytes were prepared from mouse spleens by the method of Mishell and Shiigi [13]. They were maintained in RPM1-1640 medium containing 25 mM Hepes (pH 7.2), 5- 10 -5 M /3-mercaptoethanol, 1 mM glutamine, and 5% heat-inactivated FCS with or without lipopolysaccharide (LPS) (25/zg/ml). Rat erythrocytes were prepared from blood drawn from the heart. The cells were washed 4-5 times with and suspended in phosphate-buffered saline (PBS, pH 7.4). The Zajdela ascitic hepatoma (ZAH) [14], a chemically-induced tumour of rat hepatocytes, and the AK-5 histiocytoma [15] were maintained intraperitoneally in adult Wistar rats. The cells were drawn 5 days after transplantation of the tumour, washed 3-4 times either with PBS or with CaZ+-free Eagle's MEM in Earle's balanced salt solution containing 40 mM Hepes and 100 U/ml of penicillin-G at pH 7.4 (EH-MEM), and used immediately. Regenerating rat-liver cells were obtained 20 h after partial (approx. 50%) hepatectomy of adult rat livers. Cell suspensions from normal and regenerating rat liver were prepared enzyrnatically using 0.05% collagenase in Hank's balanced salt solution (HBSS) [16]. The cells were washed with and suspended in Ca2+-free EH-MEM.

Incorporation of [ 3H]thymidine Mouse lymphocytes (1 • 106 cells/ml) were cultured with or without LPS. Stimulation of lymphocytes at the specified time was monitored by studying the incorporation of [3H]thymidine (10/xCi/ml) that was added 8 h prior to each of the time points stated. The cells were precipitated with 10% trichloroacetic acid, and the precipitate washed and dissolved in 0.6 M NaOH for measurement of radioactivity. The incorporation of [3H]thymidine in regenerating rat-liver cells obtained 20 h after partial hepatectomy, and in normal rat-liver cells, was studied by incubating the cells (1.10 6) with [3H]thymidine (10 /~Ci) for 90 min in Ca2+-containing EH-MEM (pH 7.4) in a final volume of 3 ml. After incubation, the cells were processed as for lymphocytes.

Flow cytometry ZAH cells were incubated at 37°C in PBS with the required amount of SPLN. The unlysed cells were pelleted, washed with PBS and stained with ethidium bromide as described by Larsen et al. [17]. The cells were analysed on a Becton Dickinson FACS Star cell sorter.

Synchronization of cultures CHO or RSF cells were plated as described above. After attachment, the medium was changed and hydroxyurea was added to a final concentration of 2 mM. Following incubation for 2 h, the cells were washed and suspended in CaZ+-free DMEM. SPLN was added and lysis monitored using trypan blue as above.

3. Results

Effect of SPLN on dividing and resting cells Measurement of lysis and cell viability Normal or regenerating rat-liver cells, ZAH cells, erythrocytes or lymphocytes, were incubated at a concentration of 1 • 106 cells/ml in a volume of 1-2 ml of the specified medium at 37°C in a shaken water bath for the required time, and lysis was monitored at various time intervals by counting the number of intact cells using a haemocytometer. Viability of ZAH cells was monitored by the trypan blue dye exclusion method. Cultured and AK-5 cells were washed with and suspended in Ca2+-free DMEM, and incubated at a concentration of 1. l0 s cells/ml at 37°C for the required time. The viability of the cultured cells was monitored either directly, using a phase-contrast microscope, or by the trypan blue dye-exclusion method. Seminal plasma, SPLN or the synthetic peptide was added as desired before the commencement of the incubation; other additions were made as stated in the legends to the figures and tables.

The extent of lysis of ZAH cells was studied as a function of the concentration of SPLN (Fig. 1A). With 200/xg/ml of SPLN, virtually all the ZAH cells were lysed by 30 min. SPLN lysed all the cell lines maintained in vitro that were tried (CHO, L-929, HeLa and Vero). AK-5 histiocytoma cells maintained in the ascitic form in rats, and primary cultures of rat skin fibroblasts, were also lysed by SPLN. In these cases, 100% lysis was obtained with 100 /zg/ml of SPLN within 30-60 min; lysis was proportional to the concentration of SPLN in the 0-100 tzg/ml range. Results of a representative experiment with CHO cells are given in Fig. lB. The time-course of lysis of the ZAH ceils in the presence of SPLN is shown in Fig. 2 (A, B). Dialysed seminal plasma also lysed ZAH cells. 1 mg of seminal plasma protein per ml caused extensive lysis after 15 min of incubation (Fig. 2C). SPLN (200/zg/ml) did not lyse resting lymphocytes but lysed lymphocytes that had been triggered into the

T. Ramakrishna Murti et al. / Biochimica et Biophysica Acta 1221 (1994) 109-114

/ A

lO0 8O

B

•

• ,~ ~

500

~

~oo

_s.,~

5

300

~,

20o

.£

i/o

[/0

2 0I 0 "

I

I

]

I

25

50

75

IOO

600

8

~.~

2O

5 Io

Fig. 1. Lysis of the Z A H (A) cells or CHO (B) cells as a function of the concentration of SPLN. The Z A H cells (approx. 1.0' 106 cells/ml) were suspended in Ca2+-free EH-MEM, and the CHO cells (1.105 cells/ml) in Ca2+-free D M E M containing 2% FCS, and the required amount of n-SPLN added. The cells were incubated in a total volume of 2 ml for 30 min in A and 0.2 ml for 30 min in B, and lysis measured as described in the text, n-SPLN was used in A, and s-SPLN in B.

division cycle by LPS (Fig. 3), the peak of lysis (when 21% of the cells were lysed) being obtained just after the peak of incorporation of thymidine into DNA. Similarly, SPLN (200 /zg/ml) lysed 0-3% of cells obtained in suspension from resting adult rat liver, but lysed 30-40% of the cells obtained from regenerating liver (Table 1). T h e r e was no detectable lysis of rat erythrocytes suspended in PBS with or without 0.25 M sucrose by SPLN (100 /~g/ml) in 1.5 h at 37°C. In order to verify the possibility that SPLN binds to resting cells without causing lysis, lymphocytes were treated with SPLN (100/~g/106 cells) and then stimulated with LPS. There was no lysis of resting lymphocytes by SPLN. When the incorporation of thymidine was measured in these cells and compared to incorpo-

A

c

80

/ 20

SPL.N (/Jg/ml)

ioof

111

f

20

.4

v ._2 u l0

40

60 80 Time ( h )

I00

,1o

Fig. 3. Time-course of the effect of SPLN on lymphocytes stimulated by LPS. Mouse lymphocytes (1.106 cells/ml) were cultured for the desired time with or without LPS (25/xg/ml). Aliquots of the culture were extensively washed at the specified time, suspended in Ca 2 +-free EH-MEM, incubated with n-SPLN (200 /xg/ml) for 1 h, and lysis was measured as described in the text. Stimulation of lymphocytes was monitored in a parallel set of cultures by studying the incorporation of [3H]thymidine into trichloroacetic-acid-insoluble material; the incorporation in the control was 1300 c p m / m g protein, o, stimulation of incorporation of [3H]thymidine; o, lysis by SPLN.

ration in cells not treated with SPLN, it was observed that the incorporation of thymidine was consistently lower in these cells compared to that of the control at all time points studied (data not shown). n-SPLN and s-SPLN behaved virtually identically in regard to lysis of the cells studied (e.g., Fig. 2A, B; Table 1). It was consistently observed that when ZAH cells were incubated with lower concentrations of SPLN (50 /xg/ml), only 50-55% of the cells were lysed by 30 min. The ZAH cells that were not lysed by SPLN were shown to be viable by trypan blue staining (data not shown). On further incubation, there was no increase in the extent of lysis (Fig. 2A, B). The ZAH cells that were left unlysed after treatment with various concentrations (up to 100/zg/ml) of SPLN for 15 and 30 min were analysed on a fluorescence activated cell sorter

~3

Table 1 Effect of n- and s-SPLN on isolated parenchymal cells from normal and regenerating rat liver

40

Expt. No.

2

3'0 'o

9%

6% Time

,'o

6'0

Time (min)

% Cells lysed Normal Regenerating liver liver

Incorporation of [3H]thymidine ( c p m / m g protein)

s-SPLN

s-SPLN

n-SPLN

Normal liver

30

0

60

0 0 -

18 26 34 -

11 22 33 38

-

34

,~o

(min)

Fig. 2. Time-course of the lysis of Z A H cells by n- and s-SPLN (A, B), and dialysed bovine seminal plasma (C). In A, the Z A H cells (approx. 1.0.106/ml) were suspended in EH-MEM, n-SPLN added to a final concentration of 50 (e) or 200 ( 0 ) /zg/ml, and the cells incubated in a final volume of 2 ml. In B, the Z A H cells (approx. 1.106 cells/ml) were suspended in PBS, s-SPLN added to a final concentration of 50 or 2 0 0 / z g / m l , and the cells incubated in a total volume of 0.5 ml. In C, the Z A H cells (approx. 1.0. 106/ml) were suspended in Ca2+-free E H - M E M and dialysed seminal plasma was added to the cells to a final concentration of 1 mg of seminal plasma protein per ml; the cells were incubated in a final volume of 2 ml.

1

90 120 2

90

2.6

Regenerating liver

-

-

-

-

-

-

2351

12890

The cells were incubated in PBS for studying lysis, and in Ca2+-con taining E H - M E M for thymidine incorporation. The concentration of SPLN was 2 0 0 / z g / m l .

112

T. Ramakrishna Murti et al. / Biochimica et Biophysica Acta 1221 (1994) 109-114

Table 2 FACS analysis of ethidium bromide-stained Z A H cells that remain unlysed after treatment with SPLN

Table 4 Effect of two synthetic peptides derived from SPLN on the Z A H cells

% Cells After 15 min

Z A H control +25/xgSPLN + 5 0 / x g SPLN +100/zgSPLN

After 30 rain

G~

G 2 +M

S

G1

G 2 +M

S

49.2 53.3 51.9 64.3

30.4 27.0 27.4 17.1

20.4 19.7 20.7 18.6

48.4 51.7 69.7 88.4

30.2 29.0 16.6 6.5

21.4 19.3 13.4 5.1

T h e cells were incubated at 37°C in PBS with SPLN. T h e cells that remained unlysed after incubation for 15 or 30 min were analysed on FACS. 10000 events were scored for each sample.

Peptide

% Cells lysed

concentration (/xg/ml)

30 min

Control

60 min

360 min

0

(/

(I

0

Peptide 1

50 100 200

0 0 0

0 3.5 0

3.5 0 0

Peptide 2

50 100 200

0 0 27.8

0 15.1 35.8

0 20.6 37.0

Peptide 1, Leu-Ala-Asn-Arg-Leu-Ala-Asn-Pro; Peptide 2, Ala-LysLeu-Asn-Arg-Leu-Ala-Asn-Pro. T h e cells were incubated in PBS at 37°C.

(FACS) with a view to examine whether SPLN preferentially lysed cells at any particular stage in the cell cycle (Table 2). The proportion of cells in the G 2 + M stages of the cell cycle was lower in cells treated with SPLN (25 /zg/106 cells) in comparison to the untreated cell population. When the concentration of SPLN was 100 /.zg/106 cells, the overall number of cells left unlysed reduced significantly. The proportion of ceils in the G 2 + M stages of the cell cycle reduced sharply; however, there was also a reduction of cells in the S phase at this concentration, with a concomitant increase in the cells in the G 1 stage. These results indicate that SPLN preferentially lyses dividing cells in the S and G 2 + M phases. Cells in the G 1 phase also seem to undergo lysis, but to a considerably lesser extent. CHO and RSF cells that were synchronized by treatment with hydroxyurea, which arrests cells at the beginning of the S phase, and then grown in DMEMFCS were completely lysed at 50/xg/ml of SPLN (data not shown). Effect of Ca 2 + on cell lysis As Ca 2÷ has been shown to inhibit the bacteriolytic activity of SPLN [7], we studied the effect of Ca 2÷ on the lysis of ZAH cells by SPLN. Table 3 shows that 0.4 mM Ca 2÷ provided partial protection to the ZAH cells against lysis by SPLN, whereas 4 and 10 mM Ca 2÷ totally protected the cells against the lysis.

Table 3 Effect of Ca 2+ on the lysis of Z A H cells by SPLN % Inhibition of lysis by calcium

Cells + SPLN + 0.4 m M CaCI2 + 4 m M CaCI 2 + 10 m M CaCI 2

30 min

60 min

90 min

300 min

0 94.4 100 100

0 69.7 100 100

0 75.0 100 100

0 43.0 87.0 100

T h e cells were incubated in Ca2+-free E H - M E M . T h e concentration of SPLN was 50 # g / m l .

Synthetic peptides of SPLN We synthesized two constituent peptides of SPLN representing amino-acid residues 19-28 and 21-28 of SPLN from the N-terminal end. One of the above peptides (SPLN residues 21-28) showed no lytic activity on ZAH cells. The second peptide (residues 19-28) showed lytic activity but only about half of that obtained with SPLN (Table 4).

4. Discussion

SPLN lysed all the established cell lines tested as well as cells of the two ascitic tumours of rat: one of liver parenchymal cells and the other a histiocytoma; 70-100% of these cells were lysed by 100-200/~g/ml of SPLN in 30-120 min. SPLN, however, did not lyse, even at higher concentrations, resting lymphocytes, erythrocytes, and primary liver cell suspensions from adult rat liver in which the parenchymal cells are almost entirely in the resting state. Stimulation of lymphocytes and partial hepatectomy led to the lysis of 21% and 30-40% of the cells, respectively (Fig. 3 and Table 1). The above-mentioned percentages may represent the cells that entered the cell cycle from the resting (G 0) state. The FACS analysis of the ethidium bromide-stained unlysed ZAH cells incubated with SPLN seemed to suggest that ceils in the S and G 2 + M stages of the cell cycle were more susceptible to lysis than cells in the G 1 stage. This observation is supported by the effect of SPLN on hydroxyurea-synchronized cultured CHO and RSF cells. Cells freed from the hydroxyurea block are known to immediately enter the mitotic phase [18]. It therefore seems that susceptibility to lysis by SPLN is a property of dividing cells and that resting cells acquire this property when they transit into the cell cycle from the G O state. SPLN, in this respect,

T. Ramakrishna Murti et al. / Biochimica et Biophysica Acta 1221 (1994) 109-114

resembles magainins that seem to lyse specifically tumour cells but not normal, non-dividing cells [19]; it could thus be used to distinguish cells in the Go state from cells that are in the division cycle, and to selectively lyse the latter in a mixed population. Markers that distinguish G O cells from the ceils in cycle are already known (inter alia [20,21]). Several peptides such as melittin [22], cecropin [23], magainin [24] and ~-haemolysin [25] have been shown to permeabilize membranes and lyse cells. For all these peptides, it has been proposed that their lytic activity is a consequence of their ability to form amphipathic helices which insert into the membrane and permeabilize it. These peptides have charged residues either flanking the amphipathic regions or in the helical region; these residues are believed to be important for their membrane-destabilising activity. Lysis of susceptible cells by these peptides, however, does not appear to require their entry into the cells and interaction with intracellular components, but is a consequence of their interaction with the cytoplasmic membrane. Earlier studies indicate that SPLN, though unstructured in aqueous solution, adopts an a-helical conformation in a hydrophobic environment [8]. A peptide corresponding to residues 28-40 of SPLN has already been shown to lyse erythrocytes by forming 36-40 ,A diameter lesions in the membrane [26]. A sub-sequence of SPLN representing amino acids 19-28 was shown to have a sequence homology, in respect of the positions of the hydrophobic amino acids, with some of the lytic peptides such as melittin and O-haemolysin [27]. This peptide ( A K L A N R L A N P ) was synthesized chemically and shown to have about half the lytic activity of SPLN (Table 4). It is therefore possible that SPLN may lyse cells by a mechanism similar to that obtained in the case of other lyric peptides mentioned above. It has been earlier reported that SPLN increases the fluidity of bovine sperm plasma and acrosomal membranes [9]. Therefore, the possibility of SPLN also lysing dividing cells through a direct mechanism cannot be ruled out. Although SPLN does not lyse resting cells, it seems to bind to their cell surface. SPLN has been shown to inhibit transcription and growth of a variety of microorganisms such as grampositive and gram-negative bacteria and various kinds of yeasts, by entering the cells [5,6,28]. Ca 2÷ was shown to reverse the inhibition of both transcription in and growth of E. coli, by SPLN [6]. Ca 2÷ has also been shown to inhibit the lysis of E. coli by SPLN at concentrations higher than those required for the inhibition of growth or of transcription [7]. Ca 2÷ appears to act by preventing the entry of SPLN into the cells; the Ca 2÷ effect in whole cells can thus be reversed by increasing the concentration of SPLN, while Ca 2÷ does not reverse the inhibition of transcription by E. coli R N A polymerase in vitro. There is good evidence that

113

Ca 2÷ acts extraceUularly; it does not prevent binding of pore-forming agents to cells but it blocks the lesions [29]. SPLN (also known as caltrin, [10]) has also been shown to inhibit the uptake of calcium in spermatozoa [30]. Ca z÷ inhibited the lysis of Z A H cells by SPLN (Table 3). It would, therefore, seem that the entry of SPLN into the cells may be a prerequisite for its ability to lyse the cells, and that resting cells and G 1 cells may be unable to take up SPLN even though SPLN may bind to their cell surface. The permeability properties of the cytoplasmic membrane are known to undergo a change when cells transit from the resting state into the division cycle. For example, resting cells, unlike dividing ceils, do not take up thymidine [31]. Ca 2÷ could thus prevent the lytic action of SPLN, also by its known ability to repair membrane lesions [29]. As the SPLN content of seminal plasma is approx. 500 /zg/ml and bovine seminal plasma contains approx. 46 mg of protein/ml, 1 mg of seminal plasma p r o t e i n / m l which could completely lyse Z A H cells in 1 h, would correspond to approx. 10 ~ g / m l of SPLN. When pure (natural or synthetic) SPLN was used at this concentration, no detectable lysis of the Z A H or other cells was obtained. Certain other lyric peptides have also been shown to be present in the bovine seminal plasma [32,33]. Therefore, the lytic activity of bovine seminal plasma that is not accounted for by SPLN may be due to the presence of other lytic factors or of phospholipases, or both.

5. Acknowledgements We are grateful to Dr. R. Nagaraj for the synthesis of SPLN and for advice in the preparation of this manuscript, to N. Sitaram for initial synthesis of the smaller peptide, to B.S.N. Murthy for providing n-SPLN and to Mubarak Ali for help in the cell culture experiments.

6. References [1] Boman, H.G. and Hultmark, D. (1987) Annu. Rev. Microbiol. 41, 103-126. [2] Boman, H.G., Faye, I., Gudmundsson, G.H., Lee, J.-Y. and Lidholm, D.-A. (1991) Eur. J. Biochem. 201, 23-31. [3] Bernheimer, A.W. and Rudy, B. (1986) Biochim. Biophys.Acta 864, 123-141. [4] Thompson, S.A., Tachibana, K., Nakanishi, K. and Kubota, I. (1986) Science 233, 341-343. [5] Reddy, E.S.P. and Bhargava, P.M. (1979) Nature 279, 725-728. [6] Shivaji, S., Scheit, K.H. and Bhargava, P.M. (1989) Proteins of Seminal Plasma, John Wiley and Sons, New York. [7] Chitnis, S.N., Prasad, K.S.N. and Bhargava, P.M. (1987) J. Gen. Microbiol. 133, 1265-1271. [8] Galla, H.J., Warncke, M. and Scheit, K.H. (1985) Eur. Biophys. J. 12, 211-216.

114

7'. Ramakrishna Murti et al. / Biochimica et Biophysica Acta 1221 (1994) 109-114

[9] Shivaji, S. (1986) FEBS Lett. 196, 255-258. [10] Lewis, R.V., San Agustin, J.T., Kruggel, W. and Lardy, H.A. (1985) Proc. Natl. Acad. Sci. USA 82, 6490-649t. [11] Sitaram, N., Krishna Kumari, V. and Bhargava, P.M. (1986) FEBS Lett. 201,233-236. [12] Stewart, J.W. and Young J.D. (1984) in Solid Phase Peptide Synthesis, 2nd edn., p. 18, Pierce Chemical Co., Rockford, Illinois, USA. [13] Mishell, B.B. and Shiigi, S.M. (1980) Selected Methods in Cellular Immunology, p. 23, W.H. Freeman and Co., New York. [14] Zajdela, F. (1964) in Colloque Franco-Sovietique Quelques Problemes poses par la Cellule Cancereuse, p. 47, GauthiersVillars, Paris. [15] Khar, A. (1986) J. Natl. Cancer Inst. 76, 871-878. [16] Seglen, O. (1976) Methods Cell Biol. 13, 29-83. [17] Larsen, J.K., Munch-Petersen, B., Christiansen, J. and Jorgensen, K. (1986) Cytometry 7, 54-63. [18] Maurer-Schultz, B., Siebert, M. and Bassukas, I.D. (1988) Exp. Cell Res. 174, 230-243. [19] Cruciani, R.A., Barker, J.L., Zasloff, M., Cheu, H.C. and Colamonici, O. (1991) Proc. Natl. Acad. Sci. USA 88, 3792-3796. [20] Bhargava, P.M. and Chandani, S.A. (1988) Biosci. Rep. 8, 519529. [21] Gewirtz, A.M., Anfossi, G., Venturelli, D., Valpreda, S., Sims, R. and Calabretta, B. (1989) Science 245, 180-183.

[22] Terwilliger, T.C. and Eisenberg, D. (1982) J. Biol. Chem. 257, 6010-6015. [23] Holak, T.A., Engstrom, A., Kraulis, P.J., Lindberg, G., Bennich, H., Jones, T.A., Gronenborn, A.M. and Clore, G.M. (1988) Biochemistry 27, 7620-7629. [24] Zasloff, M. (1987) Proc. Natl. Acad. Sci. USA 84, 5449-5453. [25] Lee, K.H., Fitton, J.E. and Wuthrich, K. (1987) Biochim. Biophys. Acta 911, 144-153. [26] Sitaram, N. and Nagaraj, R. (1990) J. Biol. Chem. 265, 1043810442. [27] Comte, M., Malnoe, A. and Cox, J.A. (1986) Biochem. J. 240, 567-573. [28] Scheit, K.H., Shivaji, S. and Bhargava, P.M. (1985) J. Biochem. 97, 463-471. [29] Pasternak, C.A. (1988) Biosci. Rep. 8, 579-583. [30] San Agustin, J.T., Hughes, P. and Lardy, H.A. (1987) FASEB J. 1, 60-66. [31] Bhargava, P.M., Allin, E.P. and Montagnier, L. (1976)J. Membr. Biol. 26, 1-17. [32] Gladigau, V., Morarescu, A. and Ruhenetroth-Bauer, G. (1973) Blut 26, 127-133. [33] Matousek, J., Richa, J., Louda, F., Stanek, R. and Kysilka, C. (1986) Acta Veterinaria 36, 197-206.