- Email: [email protected]
Effect of pH on Stability of Anthrax Lethal Factor: Correlation between Denaturation and Activity
Biochemical and Biophysical Research Communications 284, 568 –573 (2001) doi:10.1006/bbrc.2001.5018, available online at http://www.idealibrary.com on
Effect of pH on Stability of Anthrax Lethal Factor: Correlation between Denaturation and Activity Pankaj Gupta, Samer Singh, Ashutosh Tiwari, Rajiv Bhat, 1,2 and Rakesh Bhatnagar 2 Centre for Biotechnology, Jawaharlal Nehru University, New Delhi 110067, India
Received May 3, 2001
Anthrax is caused by Gram positive bacterium Bacillus anthracis. Pathogenesis is result of production of three protein components, protective antigen (PA), lethal factor (LF), and edema factor (EF). PA in combination with LF (lethal toxin) is lethal to animals, while PA in combination with EF (edema toxin), causes edema. PA, LF, and EF are very thermolabile. Differential scanning calorimetry (DSC) was used to unravel the energetics of LF denaturation as a function of pH ranging from 7.8 to 5.5. Transition temperature (T m) of LF was found to be ⬇42°C and onset of denaturation occurs at ⬇30°C. The ratio of calorimetric to van’t Hoff ’s enthalpy was nearly equal to unity at pH 7.0, indicative of presence of single structural domain in LF at pH 7.0, unlike PA which has been structurally observed to consist of 4 domains. It was found by cytotoxicity studies using J774A.1 macrophage like cells that LF was most stable at pH ⬃6.5. This paper reports for the first time the denaturation of LF at different pH values at 37°C and tries to establish a correlation between denaturation and loss of LF activity at different pH values. © 2001 Academic Press Key Words: lethal factor; thermostability; anthrax; protective antigen.
Anthrax toxin, a key virulence factor produced by Bacillus anthracis is composed of the proteins protective antigen (PA), edema factor (EF), and lethal factor (LF) (1–3). PA in combination with EF forms the edema toxin, causing edema in animals, and in conjunction with LF forms the lethal toxin, which is lethal to animals. PA is an 83-kDa protein and binds to specific cell surface receptors, after which it is cleaved by a protease having catalytic properties of furin (4, 5). The cleavage results in release of a 20-kDa fragment from NH 2-terminus while the remaining 63-kDa 1 Presently on sabbatical at Baker’s Laboratory, Department of Chemistry, Cornell University, Ithaca, NY. 2 To whom correspondence may be addressed. Fax: (91) 116165886/6198234/6169962. E-mail: [email protected].
0006-291X/01 $35.00 Copyright © 2001 by Academic Press All rights of reproduction in any form reserved.
COOH-terminal fragment remains bound to the receptor having an exposed site for the high affinity binding of LF or EF. The resulting complex is then internalized via receptor mediated endocytosis and delivered to the cytosol (6, 7). Lethal toxin leads to an increase in the permeability of Na ⫹ and K ⫹ ions after its internalization and is followed by the hydrolysis of ATP (7). This follows an influx of calcium, inhibition of macromolecular synthesis, and leakage of Lactate dehydrogenase (LDH), culminating in cell death (8 –11). Inositol triphospate signaling (12) and MAPK kinase signaling cascade (13) are also involved in anthrax lethal toxin cytotoxicity. Despite several studies carried out so far, the exact mechanism of the action of lethal toxin leading to cell death remains unknown. Recently, the crystal structure of PA in conjunction with other studies has revealed that after proteolytic cleavage it can form a membrane-inserting heptamer that translocates the toxic enzymes EF and LF into the lumen of acidic intracellular compartment the endosome (14 –16). It is suggested that there is a pH-dependent membrane insertion involving the formation of Porin-like membrane spanning -barrel (14). It has been proven that pH sensitive loop of PA (residue 342–355) is required for expression of anthrax lethal toxin activity (17). Anthrax toxin protein PA, LF, and EF are highly thermolabile and have to be kept frozen at ⫺70°C. They lose their activity within one to two weeks when kept at 4°C. PA and LF lose their activity completely when incubated at 37°C for 48 h (18, 19). Earlier studies carried out on PA have shown that high concentration of certain cosolvent additives like MgSO 4, sodium citrate, trehalose, Xylitol, and sorbitol could protect the loss of activity to varying extents with nearly 82% activity retained after incubation in 3M MgSO 4 at 37°C for 48 h (18). Since PA is the major component of vaccine against anthrax and LF is the major virulence factor of anthrax toxin complex, it is necessary to preserve the conformational stability of these proteins. In order to rationally design strategies for thermal stabilization of toxin components, it is of utmost importance to analyze the thermal denaturation behavior of the
568
Vol. 284, No. 3, 2001
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
components. Differential scanning calorimetry (DSC) can be used to unravel the energetics of the protein denaturation. The principle and operation of the instrument has been described elsewhere (20 –22). It provides us with not only the information about the transition temperature but also sensitive thermodynamic parameters like the enthalpy change on denaturation (⌬H) and heat capacity change (⌬Cp) upon denaturation. This paper reports on the DSC studies carried out on LF as a function of pH ranging from 5.5 to 7.8 in order to unravel the pH dependence of stability of LF. In order to correlate the thermodynamics of denaturation with biological activity, activity of LF as a function of pH was also determined by incubating the protein samples at several pH values for several days. This will enable us to analyze the correlation between thermodynamics and the kinetics of LF denaturation or destabilization, if any, and rationally design the strategy to minimize thermal inactivation of LF. MATERIALS AND METHODS Reagents and supplies. Escherichia coli SG13009 cells and NiNTA agarose were obtained from Qiagen (Germany). Resource-Q column was purchased from Amersham Pharmacia Biotech (Sweden). Cell culture plastic wares were obtained from Corning (USA). Fetal calf serum (FCS), RPMI 1640, Dulbecco’s modified Eagle medium (DMEM), Hank’s balanced salt solution (HBSS), trypsin, 3-(4,5-dimethylthiazol-2-yl),-5-diphenyltetrazolium bromide (MTT), Isopropyl-thio--D-galactopyranoside (IPTG), Phenylmethylsulfonyl fluoride (PMSF), bovine serum albumin (BSA), and other chemicals were purchased from Sigma Chemical Co. (USA). J774A.1, a macrophage like cell line was obtained from ATCC (American Type Culture Collection, USA). Media components for bacterial growth were purchased from Hi-Media Laboratories (India). Purification of LF. Plasmid pPG-LF1 (23) carrying E. coli SG13009 (pREP4) was grown in Luria broth containing 100 g of ampicillin per ml and 25 g of kanamycin per ml, with constant shaking at the rate of 250 rpm at 37°C (Infors HT). Culture was induced when Abs 600 (absorbance at 600 nm) reached ⬇1, with 1 mM IPTG and was further grown at 37°C and 250 rpm for 5 h. Cells were harvested by centrifugation at 6000g for 10 min. The cell pellet from 5 l. Culture was suspended in 225 ml of sonication buffer (50 mM Na phosphate, pH 7.8, 300 mM of NaCl). It was sonicated at 4°C (1 min bursts/1 min cooling/200 –300 Watts) for 5 cycles. PMSF was added to a final concentration of 1 mM. The lysate was centrifuged at 16,000g for 30 min. The supernatant was loaded onto a Ni-NTA resin column, which was already equilibrated with Sonication buffer. The column was then washed with 10 column volumes of 50 mM Na Phosphate buffer pH 7.0 with 300 mM NaCl, followed by eluting the protein in elution buffer (50 mM Na phosphate buffer pH 7.0, 250 mM imidazole chloride, 300 mM NaCl, 10% glycerol). One-milliliter fractions were collected and analyzed on SDS–PAGE. Fractions containing high concentration of LF were pooled and extensively dialyzed against T10E5 (Tris 10 mM, EDTA 5 mM) pH 8.0 buffer. The dialyzed samples were loaded on Resource-Q (Amersham Pharmacia) anion exchange column previously equilibrated with T10E5 buffer. Protein was eluted in 100 ml of linear gradient of 0 –500 mM NaCl in T10E5 buffer. One-milliliter fractions were collected and analyzed on SDS–PAGE. Fractions containing more than (98% pure LF were pooled and extensively dialyzed against 10 mM Hepes buffer containing 50 mM NaCl and stored frozen at ⫺70°C in ali-
FIG. 1. DSC profile of LF at pH 7.8. Scan rate 60°C/h. (A) Raw data after buffer– buffer baseline subtraction and concentration normalization (B) after curve fitting. quots for further use. Protein estimation was done using the protein determination dye (USB) based on the Bradford dye-binding procedure (24). Differential scanning calorimetry of LF. DSC experiments were carried out using model MC-2 differential scanning calorimetry from Microcal Inc. (MA). All DSC experiments were performed within a week of purifying the Protein. The protein was concentrated using Centricons (Amicon Inc.) to a concentration of 2 mg/ml and dialyzed extensively against different buffers (10 mM MES pH 5.5, 10 mM Mops pH 6.5, 10 mM Hepes pH 7.0, and 10 mM Tris pH 7.8). Any particulate matter in the protein sample was removed by centrifugation at 4°C for 30 min at 20,000g. Samples were degassed for 30 min at ambient temperature in a vacuum station connected to a pump and having magnetic stirring and then loaded with a gas tight Hamilton syringe in the calorimeter cell. The cells were pressurized with 1.5 atm, pressure of N 2, to avoid the formation of bubbles at high temperature. Initially, the dialysate buffer was loaded in both reference and sample cells, and after a baseline scan had been conducted over temperature to be used for the protein, the contents of the sample cell were gently removed and refilled with the dialyzed
569
Vol. 284, No. 3, 2001
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS the cells to a final concentration of 0.5 mg/ml. The cells were further incubated for 30 min at 37°C to allow uptake and oxidation of the dye by viable cells. The media was decanted and precipitate was dissolved in 100 l/well in 0.5% w/v SDS, 25 mM HCl in 90% isopropyl alcohol and vortexed to dissolve the precipitate. The absorbance was read at 540 nm using a microplate reader (Nunc. GMBH) and percent viability was plotted against time of incubation. Activity assay was carried out in triplicate with the average values plotted.
RESULTS AND DISCUSSIONS Figures 1–5 present the DSC data for the thermal denaturation of the LF as function of pH ranging from pH 7.8 to 5.5. At pH 7.0, the transition temperature was found to be 42°C with the onset of denaturation at ⬃30°C (Fig. 2 and Table 1). This is in accordance with the activity data wherein it was observed that LF loses its activity when incubated at 37°C for 48 h (19). The noise observed in the posttransition zone at pH 7.0 indicates aggregate formation. The calorimetric en-
FIG. 2. DSC profile of LF at pH 7.0. Scan rate 60°C/h. (A) Raw data after buffer– buffer baseline subtraction and concentration normalization (B) data after curve fitting.
protein solution. Results were obtained at a scan rate of 60°C/h. However, at pH 5.5 and 7.0, a 30°C/h scan rate was also used. The dependence of the molar heat capacity on temperature was analyzed using the ORIGIN software (Microcal Inc., USA). The experiments at each condition of pH were done in triplicate. Cytolytic activity assay of LF. Lethal factor was incubated at 37°C in the buffers MES (10 mM pH 5.5), Mops (10 mM pH 6.5), Hepes (10 mM pH 7.0), and Tris (10 mM pH 7.8) at a concentration of 2 mg/ml. Samples were removed after 6, 12, 24, 36, 48, 60, and 72 h, and activity of LF (1 g/ml) along with PA (1 g/ml) was determined using macrophage cell lysis assay (25). For this purpose, J774A.1 cell line was maintained in RPMI 1640 medium containing 10% of heat-inactivated FCS. The cell suspension was plated at 150 l/well in 96-well microtitre plate. Cells were allowed to adhere to surface of well by incubating it at 37°C for 16 h in a CO 2 incubator (95% humidity and 5% CO 2). After incubation, the medium and detached cells were gently aspirated out and replaced with 100 l/well of RPMI containing 1.0 g/ml of LF incubated at different pH values for different time intervals, along with 1.0 g/ml of PA and incubated for 3 h at 37°C in CO 2 incubator (95% humidity and 5% CO 2). After 3 h, MTT dye dissolved in RPMI medium was added to
FIG. 3. DSC profile of LF at pH 7.0. Scan rate 30°C/h. (A) Raw data after buffer– buffer baseline subtraction and concentration normalization (B) after curve fitting.
570
Vol. 284, No. 3, 2001
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
exotherm at 90°C (Fig. 1). The transitions observed at all the pH values were found irreversible after cooling and reheating the samples in the DSC experiments. When pH is lowered to 6.5 (Fig. 3), there was a decrease in the T m value to 39.7°C and the H cal to 75 kcal-mol ⫺1. The onset temperature of denaturation was observed to be marginally higher than at pH 7.0. The aggregation however occurred at a much lower temperature of 55°C than at pH 7.0. When the pH was further lowered to 5.5, the T m decreased to 39.3°C and 38.2°C at 60°C/h and 30°C/h scan rates, respectively (Table 1). The ⌬H cal values decreased to 82 and 62 kcal-mol ⫺1, respectively. The scan rate dependence was due to the irreversible nature of thermal transition and was more prominent at pH 5.5 compared to pH 7.0. From the DSC transition at pH 5.5, it is obvious that the protein is partially denatured at this pH, as evident from very low calorimetric enthalpy value. Figure 6 shows a plot
FIG. 4. DSC profile of LF at pH 6.5. Scan rate 60°C/h. (A) Raw data after buffer– buffer baseline subtraction and concentration normalization (B) data after curve fitting.
thalpy of denaturation (⌬H cal) was found to be 127 kcal-mol ⫺1 at pH 7.0 and the ratio of the calorimetric to the van’t Hoff enthalpy is nearly equal to unity, i.e., ⌬H cal/⌬H V.H. ⬇ 1.0, indicative of the presence of one structural domain in the protein molecule at pH 7.0 unlike the protective antigen (PA), another component of the anthrax lethal toxin, which has been structurally observed to consist of 4 folding domains (14). No significant temperature scan rate dependence was observed for LF denaturation as evident by a similar T m and ⌬H cal values observed at the scan rates of 60°C/h and 30°C/h at pH 7.0. As the pH was changed from 7.0 to 7.8, there was no appreciable change in the thermodynamic parameters observed (Fig. 1, Table 1). This indicates no or very little change in the protein conformation between the two pH values. It was further observed that there is sudden aggregation of the protein at pH 7.8, unlike at pH 7.0, indicated by strong
FIG. 5. DSC profile of LF at pH 5.5. After curve fitting (A) scan rate 60°C/h (B) scan rate 30°C/h.
571
Vol. 284, No. 3, 2001
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS TABLE 1
Thermodynamic Parameters for the Denaturation of LF pH
T m (°C)
⌬H cal
⌬H VH
7.8 7.0 7.0 (SR 30) 6.5 5.5 5.5 (SR 30)
41.8 42.1 41.5 39.7 39.3 38.2
106 127 105 75 82 62
97 97 104 135 140 153
Note. SR is scan rate per hour; it is 60°C/h unless specified in the table above. The error in T m measurement is ⫾0.2°C and in ⌬H value is ⫾5%.
of T m of LF versus pH of the buffer. It can be seen that the protein is more stable at pH 7.0 and 7.8 compared to at pH 6.5 and 5.5. There is a shift in the stability as we move from pH 6.5 to 7.0. To establish any possible correlation of the thermodynamic data as a function of pH by DSC with the activity data of LF at the same pH values, LF was incubated at pH 5.5, 6.5, 7.0, and 7.8 for varied time period up to 72 h at 37°C and the activity monitored by usual J774A.1 (macrophage like cell line) cytotoxicity assay. Figure 7 presents the results of the activity studies. It can be seen that there is steep decrease in the LF cytotoxicity up to 36 h of incubation followed by a gradual decrease reaching (90% viability (10% activity) at 72 h, pH 7.0. There was marginal difference between the data at pH 7.0 and 7.8 similar to what was observed in the DSC scans. LF loses activity rather slowly at pH 6.5, for example, at 48 h of incubation, 55% of the activity (cell lysis) is retained, compared to only 20% at pH 7.0; but at 72 h the activity studied for all the pH values appears to be same, i.e., protein becomes inactive and is unable to lyse the cells. Interestingly, at pH 5.5 no cell lysis was seen, i.e., percent cell viability obtained was 95% indicative of complete loss of biological function. This result correlates very well with the DSC results showing considerably diminished H cal values and a small transition.
FIG. 7. Biological activity of LF (after incubations in buffers of different pH values) on J774A.1 cells along with PA (1 g/ml of each).
The activity data thus supports the DSC data obtained. A decrease in T m and ⌬H cal values is likely to occur as a result of a decrease in the electrostatic interactions in the protein as the pH is lowered as the pI (isoelectric point) of LF is 6.01 (calculated) (26). It has been observed that proteins are maximally stable near their pIs (27). Any change in pH away from their pI leads to accumulation of a net ⫹ve or a ⫺ve charge which can influence the balance of electrostatic interactions and lead to lowering in the free energy of stabilization followed by partial or complete denaturation. Our results are in accordance with this hypothesis. Similar results have been obtained with several proteins studied as a function of pH (28). Further studies on the conformational changes monitored at these pH values by far- and near-UV circular dichroism (CD) spectroscopy may throw light on the differences in the secondary and tertiary structure as a function of pH and its relationship with the biological activity of LF. It is also interesting to note that the cooperativity of protein in folding is very good at pH 7.0 and 7.8 based on the value ⌬H cal/⌬H V.H ⬇1 while it is poor at pH 6.5 and 5.5. This indicates partial denaturation and subsequent association of the polypeptide chain at these pH values. It is first report that establishes the correlation between inactivation of LF at 37°C with thermodynamics of denaturation. It is established that LF is more stable at pH 6.5 in pH range 5.5–7.8. Further studies would throw light on the difference in secondary and tertiary structure of LF as a function of pH, which can help us in understanding the folding of LF. ACKNOWLEDGMENT This work was supported and funded by Department of Biotechnology, Government of India, New Delhi.
REFERENCES
FIG. 6. Transition temperature (T m) of LF at different pH values at scan rate 60°C/h. 572
1. Beall, F. A., Taylor, M. J., and Thorne, C. B. (1962) Rapid lethal effect in rats of a third component found upon fractionating the toxin of Bacillus anthracis. J. Bacteriol. 63, 1274 –1280. 2. Leppla, S. H., Ivins, B. E., and Ezzel, J. W. (1985) Anthrax toxin.
Vol. 284, No. 3, 2001
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
In Microbiology (Levine, L., Bonventre, P. F., Morello, J. A., Schlessinger, S., Silver S. D., and Wu, H. C., Eds.), pp. 63– 64, American Society of Microbiology, Washington, DC. Freidlander, A. M., (1986) Macrophages are sensitive to anthrax toxin through an acid dependent process. J. Biol. Chem. 261, 7123–7126. Leppla, S. H., Freidlander, A. M, and Cora, E. M. (1988) Proteolytic activation of anthrax protective antigen bound to cell receptors. In Bacterial Protein Toxins (Fehrenbach, F., Alouf, E., Falmagene, P., Goebel, W., Jeljaszewics, J., Jurgen, D., and Rappuoli, R., Eds.), pp. 111–112, Gustav Fisher. Klimpel, K. R., Molly, S. S., Thomas, G., and Leppla, S. H. (1992) Anthrax toxin protective antigen is activated by a cell surface protease with the sequence specificity and catalytic properties of furin. Proc. Natl. Acad. Sci. USA 89, 10277–10281. Gordon, V. M., Young, W. W., Lechler, S. M., Gray, M. C., Leppla, S. H., and Hewlett, E. L. (1989) Adenylate cyclase toxins from B. anthracis and Bordetella pertussis. Different processes for interaction with and entry into target cells. J. Biol. Chem. 264, 14792–14796. Hanna, P. C., Kochi, S., and Collier, R. J. (1992) Biochemical and physiological changes induced by anthrax lethal toxin in J774A.1 macrophage like cells. Mol. Biol. Cell 3, 1269 –1277. Bhatnagar, R., Singh, Y., Leppla, S. H., and Friedlander, A. M. (1989) Calcium is required for the expression of anthrax lethal toxin activity in the macrophage like cell line J774A.1. Infect. Immun. 57, 2107–2114. Singh, Y., Chaudhary, V. K., and Leppla, S. H. (1989) A deleted variant of B. anthracis protective antigen is non-toxic and blocks anthrax toxin action in vivo. J. Biol. Chem. 264, 19103–19107. Friedlander, A. M., Bhatnagar, R., Leppla, S. H., and Singh, Y. (1993) Characterization of macrophage sensitivity and resistance to anthrax lethal toxin. Infect. Immun. 61, 245–252. Bhatnagar, R., and Friedlander, A. M. (1994) Protein synthesis is required for expression of anthrax lethal toxin cytotoxicity. Infect. Immun. 62, 2958 –2962. Bhatnagar, R., Ahuja, N., Goila, R., Batra, S., Waheed, S. M., and Gupta, P. (1999) Activation of Phospholipase C and Protein kinase C is required for expression of anthrax toxin cytotoxicity in J774A.1 cells. Cell. Signal. 11, 111–116. Duesbery, N. S., Craig, P. W., Leppla, S. H., Gordon, V. M., Klimpel, K. R., Copeland, T. D., Ahn, N. G., Oskarsson, M. K., Fukasawa, K., Paull, K. D., and Wonde, G. F. V. (1998) Proteo-
14.
15.
16.
17.
18.
19.
20. 21. 22. 23.
24.
25.
26.
27. 28.
573
lytic inactivation of MAP-Kinase-Kinase by anthrax Lethal Factor. Science 280, 734 –737. Petosa, C., Klimpel, K. R., Leppla, S. H., and Leddington, R. C. (1997) Crystal structure of anthrax toxin Protective antigen. Nature 385, 833– 838. Milne, J. C., Furlong, D., Hanna, P. C., Wall, J. S., and Collier, R. J. (1994) Anthrax protective antigen forms oligomers during intoxication of mammalian cells. J. Biol. Chem. 269, 20607– 20612. Saelinger, C. B. (1990) Toxin structure and function. In Trafficking of Bacterial Toxins (Saelinger, C. B., Ed.), pp. 1–14, CRC Press. Batra, S., Gupta, P., Chauhan, V., Singh, A., and Bhatnagar, R. (2001) Trp 346 and Leu 352 residues in protective antigen are required for expression of anthrax lethal toxin activity. Biochem. Biophys. Res. Commun. 281, 186 –192. Radha, C., Salotra, P., Bhat, R., and Bhatnagar, R. (1996) Thermostabilization of protective antigen the binding component of anthrax lethal toxin. J. Biotechnol. 50, 235–242. Fish, D. C., Mahlandt, B. G., Dobbs, J. P., and Lincoln, R. E. (1968) Purification and Properties of in vitro produced Anthrax toxin components. J. Bacteriol. 95, 907–918. Krishnan, K. S., and Brandts, J. F. (1978) Scanning calorimetry. Methods Enzymol. 49, 3–14. Privalov, P. L. (1980) Scanning microcalorimeters for studying macromolecules. Pure Appl. Chem. 52, 479 – 497. Sturtevant, J. M. (1987) Biochemical applications of differential scanning calorimetry. Annu. Rev. Phys. Chem. 38, 463– 488. Gupta, P., Batra, S., Chopra, A. P., Singh, Y., and Bhatnagar, R. (1998) Expression and purification of recombinant lethal factor of Bacillus anthracis. Infect. Immun. 66, 862– 865. Bradford, M. (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilising the principle of protein-dye binding. Anal. Biochem. 72, 248 –254. Gupta, P., Waheed, S. M., and Bhatnagar, R. (1999) Expression and purification of recombinant protective antigen of Bacillus anthracis. Protein Expr. Purif. 16,369 –376. Bragg, T. S., and Robertson, D. L., (1989) Nucleotide sequence and analysis of the lethal factor gene (lef) from Bacillus anthracis. Gene 81, 45–54. Stigter, D., and Dill, K. A. (1990) Charge effects on folded and unfolded proteins. Biochemistry 29, 1262–1271. Privalov, P. L. (1979) Stability of proteins. Small globular proteins. Adv. Prot. Chem. 33, 167–241.
Effect of pH on Stability of Anthrax Lethal Factor: Correlation between Denaturation and Activity Pankaj Gupta, Samer Singh, Ashutosh Tiwari, Rajiv Bhat, 1,2 and Rakesh Bhatnagar 2 Centre for Biotechnology, Jawaharlal Nehru University, New Delhi 110067, India
Received May 3, 2001
Anthrax is caused by Gram positive bacterium Bacillus anthracis. Pathogenesis is result of production of three protein components, protective antigen (PA), lethal factor (LF), and edema factor (EF). PA in combination with LF (lethal toxin) is lethal to animals, while PA in combination with EF (edema toxin), causes edema. PA, LF, and EF are very thermolabile. Differential scanning calorimetry (DSC) was used to unravel the energetics of LF denaturation as a function of pH ranging from 7.8 to 5.5. Transition temperature (T m) of LF was found to be ⬇42°C and onset of denaturation occurs at ⬇30°C. The ratio of calorimetric to van’t Hoff ’s enthalpy was nearly equal to unity at pH 7.0, indicative of presence of single structural domain in LF at pH 7.0, unlike PA which has been structurally observed to consist of 4 domains. It was found by cytotoxicity studies using J774A.1 macrophage like cells that LF was most stable at pH ⬃6.5. This paper reports for the first time the denaturation of LF at different pH values at 37°C and tries to establish a correlation between denaturation and loss of LF activity at different pH values. © 2001 Academic Press Key Words: lethal factor; thermostability; anthrax; protective antigen.
Anthrax toxin, a key virulence factor produced by Bacillus anthracis is composed of the proteins protective antigen (PA), edema factor (EF), and lethal factor (LF) (1–3). PA in combination with EF forms the edema toxin, causing edema in animals, and in conjunction with LF forms the lethal toxin, which is lethal to animals. PA is an 83-kDa protein and binds to specific cell surface receptors, after which it is cleaved by a protease having catalytic properties of furin (4, 5). The cleavage results in release of a 20-kDa fragment from NH 2-terminus while the remaining 63-kDa 1 Presently on sabbatical at Baker’s Laboratory, Department of Chemistry, Cornell University, Ithaca, NY. 2 To whom correspondence may be addressed. Fax: (91) 116165886/6198234/6169962. E-mail: [email protected].
0006-291X/01 $35.00 Copyright © 2001 by Academic Press All rights of reproduction in any form reserved.
COOH-terminal fragment remains bound to the receptor having an exposed site for the high affinity binding of LF or EF. The resulting complex is then internalized via receptor mediated endocytosis and delivered to the cytosol (6, 7). Lethal toxin leads to an increase in the permeability of Na ⫹ and K ⫹ ions after its internalization and is followed by the hydrolysis of ATP (7). This follows an influx of calcium, inhibition of macromolecular synthesis, and leakage of Lactate dehydrogenase (LDH), culminating in cell death (8 –11). Inositol triphospate signaling (12) and MAPK kinase signaling cascade (13) are also involved in anthrax lethal toxin cytotoxicity. Despite several studies carried out so far, the exact mechanism of the action of lethal toxin leading to cell death remains unknown. Recently, the crystal structure of PA in conjunction with other studies has revealed that after proteolytic cleavage it can form a membrane-inserting heptamer that translocates the toxic enzymes EF and LF into the lumen of acidic intracellular compartment the endosome (14 –16). It is suggested that there is a pH-dependent membrane insertion involving the formation of Porin-like membrane spanning -barrel (14). It has been proven that pH sensitive loop of PA (residue 342–355) is required for expression of anthrax lethal toxin activity (17). Anthrax toxin protein PA, LF, and EF are highly thermolabile and have to be kept frozen at ⫺70°C. They lose their activity within one to two weeks when kept at 4°C. PA and LF lose their activity completely when incubated at 37°C for 48 h (18, 19). Earlier studies carried out on PA have shown that high concentration of certain cosolvent additives like MgSO 4, sodium citrate, trehalose, Xylitol, and sorbitol could protect the loss of activity to varying extents with nearly 82% activity retained after incubation in 3M MgSO 4 at 37°C for 48 h (18). Since PA is the major component of vaccine against anthrax and LF is the major virulence factor of anthrax toxin complex, it is necessary to preserve the conformational stability of these proteins. In order to rationally design strategies for thermal stabilization of toxin components, it is of utmost importance to analyze the thermal denaturation behavior of the
568
Vol. 284, No. 3, 2001
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
components. Differential scanning calorimetry (DSC) can be used to unravel the energetics of the protein denaturation. The principle and operation of the instrument has been described elsewhere (20 –22). It provides us with not only the information about the transition temperature but also sensitive thermodynamic parameters like the enthalpy change on denaturation (⌬H) and heat capacity change (⌬Cp) upon denaturation. This paper reports on the DSC studies carried out on LF as a function of pH ranging from 5.5 to 7.8 in order to unravel the pH dependence of stability of LF. In order to correlate the thermodynamics of denaturation with biological activity, activity of LF as a function of pH was also determined by incubating the protein samples at several pH values for several days. This will enable us to analyze the correlation between thermodynamics and the kinetics of LF denaturation or destabilization, if any, and rationally design the strategy to minimize thermal inactivation of LF. MATERIALS AND METHODS Reagents and supplies. Escherichia coli SG13009 cells and NiNTA agarose were obtained from Qiagen (Germany). Resource-Q column was purchased from Amersham Pharmacia Biotech (Sweden). Cell culture plastic wares were obtained from Corning (USA). Fetal calf serum (FCS), RPMI 1640, Dulbecco’s modified Eagle medium (DMEM), Hank’s balanced salt solution (HBSS), trypsin, 3-(4,5-dimethylthiazol-2-yl),-5-diphenyltetrazolium bromide (MTT), Isopropyl-thio--D-galactopyranoside (IPTG), Phenylmethylsulfonyl fluoride (PMSF), bovine serum albumin (BSA), and other chemicals were purchased from Sigma Chemical Co. (USA). J774A.1, a macrophage like cell line was obtained from ATCC (American Type Culture Collection, USA). Media components for bacterial growth were purchased from Hi-Media Laboratories (India). Purification of LF. Plasmid pPG-LF1 (23) carrying E. coli SG13009 (pREP4) was grown in Luria broth containing 100 g of ampicillin per ml and 25 g of kanamycin per ml, with constant shaking at the rate of 250 rpm at 37°C (Infors HT). Culture was induced when Abs 600 (absorbance at 600 nm) reached ⬇1, with 1 mM IPTG and was further grown at 37°C and 250 rpm for 5 h. Cells were harvested by centrifugation at 6000g for 10 min. The cell pellet from 5 l. Culture was suspended in 225 ml of sonication buffer (50 mM Na phosphate, pH 7.8, 300 mM of NaCl). It was sonicated at 4°C (1 min bursts/1 min cooling/200 –300 Watts) for 5 cycles. PMSF was added to a final concentration of 1 mM. The lysate was centrifuged at 16,000g for 30 min. The supernatant was loaded onto a Ni-NTA resin column, which was already equilibrated with Sonication buffer. The column was then washed with 10 column volumes of 50 mM Na Phosphate buffer pH 7.0 with 300 mM NaCl, followed by eluting the protein in elution buffer (50 mM Na phosphate buffer pH 7.0, 250 mM imidazole chloride, 300 mM NaCl, 10% glycerol). One-milliliter fractions were collected and analyzed on SDS–PAGE. Fractions containing high concentration of LF were pooled and extensively dialyzed against T10E5 (Tris 10 mM, EDTA 5 mM) pH 8.0 buffer. The dialyzed samples were loaded on Resource-Q (Amersham Pharmacia) anion exchange column previously equilibrated with T10E5 buffer. Protein was eluted in 100 ml of linear gradient of 0 –500 mM NaCl in T10E5 buffer. One-milliliter fractions were collected and analyzed on SDS–PAGE. Fractions containing more than (98% pure LF were pooled and extensively dialyzed against 10 mM Hepes buffer containing 50 mM NaCl and stored frozen at ⫺70°C in ali-
FIG. 1. DSC profile of LF at pH 7.8. Scan rate 60°C/h. (A) Raw data after buffer– buffer baseline subtraction and concentration normalization (B) after curve fitting. quots for further use. Protein estimation was done using the protein determination dye (USB) based on the Bradford dye-binding procedure (24). Differential scanning calorimetry of LF. DSC experiments were carried out using model MC-2 differential scanning calorimetry from Microcal Inc. (MA). All DSC experiments were performed within a week of purifying the Protein. The protein was concentrated using Centricons (Amicon Inc.) to a concentration of 2 mg/ml and dialyzed extensively against different buffers (10 mM MES pH 5.5, 10 mM Mops pH 6.5, 10 mM Hepes pH 7.0, and 10 mM Tris pH 7.8). Any particulate matter in the protein sample was removed by centrifugation at 4°C for 30 min at 20,000g. Samples were degassed for 30 min at ambient temperature in a vacuum station connected to a pump and having magnetic stirring and then loaded with a gas tight Hamilton syringe in the calorimeter cell. The cells were pressurized with 1.5 atm, pressure of N 2, to avoid the formation of bubbles at high temperature. Initially, the dialysate buffer was loaded in both reference and sample cells, and after a baseline scan had been conducted over temperature to be used for the protein, the contents of the sample cell were gently removed and refilled with the dialyzed
569
Vol. 284, No. 3, 2001
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS the cells to a final concentration of 0.5 mg/ml. The cells were further incubated for 30 min at 37°C to allow uptake and oxidation of the dye by viable cells. The media was decanted and precipitate was dissolved in 100 l/well in 0.5% w/v SDS, 25 mM HCl in 90% isopropyl alcohol and vortexed to dissolve the precipitate. The absorbance was read at 540 nm using a microplate reader (Nunc. GMBH) and percent viability was plotted against time of incubation. Activity assay was carried out in triplicate with the average values plotted.
RESULTS AND DISCUSSIONS Figures 1–5 present the DSC data for the thermal denaturation of the LF as function of pH ranging from pH 7.8 to 5.5. At pH 7.0, the transition temperature was found to be 42°C with the onset of denaturation at ⬃30°C (Fig. 2 and Table 1). This is in accordance with the activity data wherein it was observed that LF loses its activity when incubated at 37°C for 48 h (19). The noise observed in the posttransition zone at pH 7.0 indicates aggregate formation. The calorimetric en-
FIG. 2. DSC profile of LF at pH 7.0. Scan rate 60°C/h. (A) Raw data after buffer– buffer baseline subtraction and concentration normalization (B) data after curve fitting.
protein solution. Results were obtained at a scan rate of 60°C/h. However, at pH 5.5 and 7.0, a 30°C/h scan rate was also used. The dependence of the molar heat capacity on temperature was analyzed using the ORIGIN software (Microcal Inc., USA). The experiments at each condition of pH were done in triplicate. Cytolytic activity assay of LF. Lethal factor was incubated at 37°C in the buffers MES (10 mM pH 5.5), Mops (10 mM pH 6.5), Hepes (10 mM pH 7.0), and Tris (10 mM pH 7.8) at a concentration of 2 mg/ml. Samples were removed after 6, 12, 24, 36, 48, 60, and 72 h, and activity of LF (1 g/ml) along with PA (1 g/ml) was determined using macrophage cell lysis assay (25). For this purpose, J774A.1 cell line was maintained in RPMI 1640 medium containing 10% of heat-inactivated FCS. The cell suspension was plated at 150 l/well in 96-well microtitre plate. Cells were allowed to adhere to surface of well by incubating it at 37°C for 16 h in a CO 2 incubator (95% humidity and 5% CO 2). After incubation, the medium and detached cells were gently aspirated out and replaced with 100 l/well of RPMI containing 1.0 g/ml of LF incubated at different pH values for different time intervals, along with 1.0 g/ml of PA and incubated for 3 h at 37°C in CO 2 incubator (95% humidity and 5% CO 2). After 3 h, MTT dye dissolved in RPMI medium was added to
FIG. 3. DSC profile of LF at pH 7.0. Scan rate 30°C/h. (A) Raw data after buffer– buffer baseline subtraction and concentration normalization (B) after curve fitting.
570
Vol. 284, No. 3, 2001
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
exotherm at 90°C (Fig. 1). The transitions observed at all the pH values were found irreversible after cooling and reheating the samples in the DSC experiments. When pH is lowered to 6.5 (Fig. 3), there was a decrease in the T m value to 39.7°C and the H cal to 75 kcal-mol ⫺1. The onset temperature of denaturation was observed to be marginally higher than at pH 7.0. The aggregation however occurred at a much lower temperature of 55°C than at pH 7.0. When the pH was further lowered to 5.5, the T m decreased to 39.3°C and 38.2°C at 60°C/h and 30°C/h scan rates, respectively (Table 1). The ⌬H cal values decreased to 82 and 62 kcal-mol ⫺1, respectively. The scan rate dependence was due to the irreversible nature of thermal transition and was more prominent at pH 5.5 compared to pH 7.0. From the DSC transition at pH 5.5, it is obvious that the protein is partially denatured at this pH, as evident from very low calorimetric enthalpy value. Figure 6 shows a plot
FIG. 4. DSC profile of LF at pH 6.5. Scan rate 60°C/h. (A) Raw data after buffer– buffer baseline subtraction and concentration normalization (B) data after curve fitting.
thalpy of denaturation (⌬H cal) was found to be 127 kcal-mol ⫺1 at pH 7.0 and the ratio of the calorimetric to the van’t Hoff enthalpy is nearly equal to unity, i.e., ⌬H cal/⌬H V.H. ⬇ 1.0, indicative of the presence of one structural domain in the protein molecule at pH 7.0 unlike the protective antigen (PA), another component of the anthrax lethal toxin, which has been structurally observed to consist of 4 folding domains (14). No significant temperature scan rate dependence was observed for LF denaturation as evident by a similar T m and ⌬H cal values observed at the scan rates of 60°C/h and 30°C/h at pH 7.0. As the pH was changed from 7.0 to 7.8, there was no appreciable change in the thermodynamic parameters observed (Fig. 1, Table 1). This indicates no or very little change in the protein conformation between the two pH values. It was further observed that there is sudden aggregation of the protein at pH 7.8, unlike at pH 7.0, indicated by strong
FIG. 5. DSC profile of LF at pH 5.5. After curve fitting (A) scan rate 60°C/h (B) scan rate 30°C/h.
571
Vol. 284, No. 3, 2001
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS TABLE 1
Thermodynamic Parameters for the Denaturation of LF pH
T m (°C)
⌬H cal
⌬H VH
7.8 7.0 7.0 (SR 30) 6.5 5.5 5.5 (SR 30)
41.8 42.1 41.5 39.7 39.3 38.2
106 127 105 75 82 62
97 97 104 135 140 153
Note. SR is scan rate per hour; it is 60°C/h unless specified in the table above. The error in T m measurement is ⫾0.2°C and in ⌬H value is ⫾5%.
of T m of LF versus pH of the buffer. It can be seen that the protein is more stable at pH 7.0 and 7.8 compared to at pH 6.5 and 5.5. There is a shift in the stability as we move from pH 6.5 to 7.0. To establish any possible correlation of the thermodynamic data as a function of pH by DSC with the activity data of LF at the same pH values, LF was incubated at pH 5.5, 6.5, 7.0, and 7.8 for varied time period up to 72 h at 37°C and the activity monitored by usual J774A.1 (macrophage like cell line) cytotoxicity assay. Figure 7 presents the results of the activity studies. It can be seen that there is steep decrease in the LF cytotoxicity up to 36 h of incubation followed by a gradual decrease reaching (90% viability (10% activity) at 72 h, pH 7.0. There was marginal difference between the data at pH 7.0 and 7.8 similar to what was observed in the DSC scans. LF loses activity rather slowly at pH 6.5, for example, at 48 h of incubation, 55% of the activity (cell lysis) is retained, compared to only 20% at pH 7.0; but at 72 h the activity studied for all the pH values appears to be same, i.e., protein becomes inactive and is unable to lyse the cells. Interestingly, at pH 5.5 no cell lysis was seen, i.e., percent cell viability obtained was 95% indicative of complete loss of biological function. This result correlates very well with the DSC results showing considerably diminished H cal values and a small transition.
FIG. 7. Biological activity of LF (after incubations in buffers of different pH values) on J774A.1 cells along with PA (1 g/ml of each).
The activity data thus supports the DSC data obtained. A decrease in T m and ⌬H cal values is likely to occur as a result of a decrease in the electrostatic interactions in the protein as the pH is lowered as the pI (isoelectric point) of LF is 6.01 (calculated) (26). It has been observed that proteins are maximally stable near their pIs (27). Any change in pH away from their pI leads to accumulation of a net ⫹ve or a ⫺ve charge which can influence the balance of electrostatic interactions and lead to lowering in the free energy of stabilization followed by partial or complete denaturation. Our results are in accordance with this hypothesis. Similar results have been obtained with several proteins studied as a function of pH (28). Further studies on the conformational changes monitored at these pH values by far- and near-UV circular dichroism (CD) spectroscopy may throw light on the differences in the secondary and tertiary structure as a function of pH and its relationship with the biological activity of LF. It is also interesting to note that the cooperativity of protein in folding is very good at pH 7.0 and 7.8 based on the value ⌬H cal/⌬H V.H ⬇1 while it is poor at pH 6.5 and 5.5. This indicates partial denaturation and subsequent association of the polypeptide chain at these pH values. It is first report that establishes the correlation between inactivation of LF at 37°C with thermodynamics of denaturation. It is established that LF is more stable at pH 6.5 in pH range 5.5–7.8. Further studies would throw light on the difference in secondary and tertiary structure of LF as a function of pH, which can help us in understanding the folding of LF. ACKNOWLEDGMENT This work was supported and funded by Department of Biotechnology, Government of India, New Delhi.
REFERENCES
FIG. 6. Transition temperature (T m) of LF at different pH values at scan rate 60°C/h. 572
1. Beall, F. A., Taylor, M. J., and Thorne, C. B. (1962) Rapid lethal effect in rats of a third component found upon fractionating the toxin of Bacillus anthracis. J. Bacteriol. 63, 1274 –1280. 2. Leppla, S. H., Ivins, B. E., and Ezzel, J. W. (1985) Anthrax toxin.
Vol. 284, No. 3, 2001
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
In Microbiology (Levine, L., Bonventre, P. F., Morello, J. A., Schlessinger, S., Silver S. D., and Wu, H. C., Eds.), pp. 63– 64, American Society of Microbiology, Washington, DC. Freidlander, A. M., (1986) Macrophages are sensitive to anthrax toxin through an acid dependent process. J. Biol. Chem. 261, 7123–7126. Leppla, S. H., Freidlander, A. M, and Cora, E. M. (1988) Proteolytic activation of anthrax protective antigen bound to cell receptors. In Bacterial Protein Toxins (Fehrenbach, F., Alouf, E., Falmagene, P., Goebel, W., Jeljaszewics, J., Jurgen, D., and Rappuoli, R., Eds.), pp. 111–112, Gustav Fisher. Klimpel, K. R., Molly, S. S., Thomas, G., and Leppla, S. H. (1992) Anthrax toxin protective antigen is activated by a cell surface protease with the sequence specificity and catalytic properties of furin. Proc. Natl. Acad. Sci. USA 89, 10277–10281. Gordon, V. M., Young, W. W., Lechler, S. M., Gray, M. C., Leppla, S. H., and Hewlett, E. L. (1989) Adenylate cyclase toxins from B. anthracis and Bordetella pertussis. Different processes for interaction with and entry into target cells. J. Biol. Chem. 264, 14792–14796. Hanna, P. C., Kochi, S., and Collier, R. J. (1992) Biochemical and physiological changes induced by anthrax lethal toxin in J774A.1 macrophage like cells. Mol. Biol. Cell 3, 1269 –1277. Bhatnagar, R., Singh, Y., Leppla, S. H., and Friedlander, A. M. (1989) Calcium is required for the expression of anthrax lethal toxin activity in the macrophage like cell line J774A.1. Infect. Immun. 57, 2107–2114. Singh, Y., Chaudhary, V. K., and Leppla, S. H. (1989) A deleted variant of B. anthracis protective antigen is non-toxic and blocks anthrax toxin action in vivo. J. Biol. Chem. 264, 19103–19107. Friedlander, A. M., Bhatnagar, R., Leppla, S. H., and Singh, Y. (1993) Characterization of macrophage sensitivity and resistance to anthrax lethal toxin. Infect. Immun. 61, 245–252. Bhatnagar, R., and Friedlander, A. M. (1994) Protein synthesis is required for expression of anthrax lethal toxin cytotoxicity. Infect. Immun. 62, 2958 –2962. Bhatnagar, R., Ahuja, N., Goila, R., Batra, S., Waheed, S. M., and Gupta, P. (1999) Activation of Phospholipase C and Protein kinase C is required for expression of anthrax toxin cytotoxicity in J774A.1 cells. Cell. Signal. 11, 111–116. Duesbery, N. S., Craig, P. W., Leppla, S. H., Gordon, V. M., Klimpel, K. R., Copeland, T. D., Ahn, N. G., Oskarsson, M. K., Fukasawa, K., Paull, K. D., and Wonde, G. F. V. (1998) Proteo-
14.
15.
16.
17.
18.
19.
20. 21. 22. 23.
24.
25.
26.
27. 28.
573
lytic inactivation of MAP-Kinase-Kinase by anthrax Lethal Factor. Science 280, 734 –737. Petosa, C., Klimpel, K. R., Leppla, S. H., and Leddington, R. C. (1997) Crystal structure of anthrax toxin Protective antigen. Nature 385, 833– 838. Milne, J. C., Furlong, D., Hanna, P. C., Wall, J. S., and Collier, R. J. (1994) Anthrax protective antigen forms oligomers during intoxication of mammalian cells. J. Biol. Chem. 269, 20607– 20612. Saelinger, C. B. (1990) Toxin structure and function. In Trafficking of Bacterial Toxins (Saelinger, C. B., Ed.), pp. 1–14, CRC Press. Batra, S., Gupta, P., Chauhan, V., Singh, A., and Bhatnagar, R. (2001) Trp 346 and Leu 352 residues in protective antigen are required for expression of anthrax lethal toxin activity. Biochem. Biophys. Res. Commun. 281, 186 –192. Radha, C., Salotra, P., Bhat, R., and Bhatnagar, R. (1996) Thermostabilization of protective antigen the binding component of anthrax lethal toxin. J. Biotechnol. 50, 235–242. Fish, D. C., Mahlandt, B. G., Dobbs, J. P., and Lincoln, R. E. (1968) Purification and Properties of in vitro produced Anthrax toxin components. J. Bacteriol. 95, 907–918. Krishnan, K. S., and Brandts, J. F. (1978) Scanning calorimetry. Methods Enzymol. 49, 3–14. Privalov, P. L. (1980) Scanning microcalorimeters for studying macromolecules. Pure Appl. Chem. 52, 479 – 497. Sturtevant, J. M. (1987) Biochemical applications of differential scanning calorimetry. Annu. Rev. Phys. Chem. 38, 463– 488. Gupta, P., Batra, S., Chopra, A. P., Singh, Y., and Bhatnagar, R. (1998) Expression and purification of recombinant lethal factor of Bacillus anthracis. Infect. Immun. 66, 862– 865. Bradford, M. (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilising the principle of protein-dye binding. Anal. Biochem. 72, 248 –254. Gupta, P., Waheed, S. M., and Bhatnagar, R. (1999) Expression and purification of recombinant protective antigen of Bacillus anthracis. Protein Expr. Purif. 16,369 –376. Bragg, T. S., and Robertson, D. L., (1989) Nucleotide sequence and analysis of the lethal factor gene (lef) from Bacillus anthracis. Gene 81, 45–54. Stigter, D., and Dill, K. A. (1990) Charge effects on folded and unfolded proteins. Biochemistry 29, 1262–1271. Privalov, P. L. (1979) Stability of proteins. Small globular proteins. Adv. Prot. Chem. 33, 167–241.