intercalated in the interlayers spaces of smectites, usually undergoing extensive ... the effect of heating on the stability of protein-clay ... acetate (pH 5.0) and citrate, dithionite, bicarbonate. (CDB) solutions (McKeague, 1978). The Na-saturated clay minerals were prepared by .... According to these authors, the inflection point.
Clay Minerals (1995) 30, 325-336
PHYSICOCHEMICAL PROPERTIES OF PROTEIN-SMECTITE AND PROTEIN-AL(OH)x-SMECTITE COMPLEXES A. V I O L A N T E ,
A. DE C R I S T O F A R O ,
M.A. RAt
AND L. G I A N F R E D A
Dipartimento di Scienze Chimico-Agrarie, Universith di Napoli, "Federico H", 80055 Portici, Napoli, Italy (Received 22 June 1994; revised 2 May 1995)
A B S TRACT: Proteins (catalase, albumin, pepsin and lysozyme with different molecular weights and isoelectric points) were differently adsorbed at pH 7.0 on the clay fraction of three raw Nasaturated smectites (Crook and Uri montmorillonites and one hectorite). The adsorption isotherms of proteins on clay minerals showed typical Langmuir characteristics. Lysozyme was adsorbed under the effect of electrostatic interactions between the opposite charges of clay surfaces and protein molecules, whereas catalase and albumin were adsorbed under the effect of non-electrostatic forces. Pepsin was held in relatively high amounts only on the surfaces of hectorite. Proteins were intercalated in the interlayers spaces of smectites, usually undergoing extensive unfolding. Proteinsmectite complexes showed different behaviour to heating treatment. Some complexes remained practically unchanged after heating at 200~ Presence of 'wrecks' of interlayered materials was found after heating at 500~ for two hours. The amounts of proteins adsorbed on the external and interlamellar surfaces of clay minerals, partially coated with OH-A1 species, were much lower than those fixed on the clean clays. Only lysozyme was intercalated in chlorite-like complexes.
In the last 40 years, adsorption of proteins on clay minerals has received the attention of many researches with diverse interests. The capacity of proteins to be sorbed is influenced by pH of the system, surface area of the clay minerals (kaolinite, illite, montmorillonite), isoelectric point (iep) of proteins, cation exchange capacity (CEC) of minerals, nature of cation saturating the clays, and temperature (McLaren et aL, 1958; Armstrong & Chesters, 1964; Harter & Stotzky, 1971, 1973; Harter, 1975; Theng, 1979). However, in spite of the fact that the adsorption and interlayering of proteins in expansible clay minerals has been studied in detail (McLaren et al., t958; Harter & Stotzky, 1973; Harter, 1975; Larsson & Siffert, 1983; Norde, 1986; Fusi et al:, 1989; Quiquampoix & Ratcliffe, 1992), the mechanism for the intercalation of proteins in the interlamellar spaces of smectites has received different interpretations (McLaren et al., 1958; Harter & Stotzky, 1973; Theng, 1979; Larsson & Siffert, 1983; Norde, 1986; Fusi et al., 1989; Quiquampoix & Ratcliffe, 1992). Furthermore, no studies have been carded out on the effect of heating on the stability of protein-clay mineral complexes in order to obtain more
information on the nature, interlayering and properties of these complexes. X-ray diffraction (XRD) data of protein-clay complexes have usually been obtained at room temperature or, only in a few studies, after preheating at I I0~ (Harter & Stotzky, 1973; Fusi et al., 1989). In acidic soil environments, A1 ions are released to soil solutions through chemical and biological weathering reactions of clay minerals and undergo a series of complicated reactions. Soluble monomers and polymers of AI as well as short-range ordered A1 hydroxides and oxyhydroxides may be retained by expansible clay minerals on their interlayer surfaces modifying the physicochemical properties and the mineralogy of minerals (Huang & Violante, 1986; Barnhisel & Bertsch, 1989). The adsorption of proteins on clay minerals partially coated with hydrolytic products of A1 has received scant attention (Gianfreda et al., 1991, 1992; Naidja et al., 1995). The purpose of our research was to study: (1) the adsorption of proteins (catalase, albumin, pepsin and lysozyme) of different molecular weight (MW) and isoelectric points on each of three smectites (two montmorillonites and one hectorite); (2) the
9 1995 The Mineralogical Society
A. Violante et al.
326
physicochemical properties of the resultant proteinclay mineral complexes; and (3) the effect of hydroxy A1 species (OH-A1) coating the surfaces of hectorite, on the sorption and interlayering of proteins. MATERIALS
AND METHODS
Clay minerals and chemicals The 100,000 (their DC1 sample). Our findings did not confirm the results of other authors on the intercalation of catalase by smectite (Harter & Stotzky, 1973; Fusi et al., 1989). Hatter & Stotzky (1973) found that catalase did not intercalate the Ca-, Al-, La- or Th-Wyoming montmorillonite, even though the ratio of adsorbed protein-to-clay exceeded 1:5. On the contrary, Fusi et al. (1989) found that catalase first covered the external surfaces and then penetrated the inteflayers of a Ca-saturated Upton montmofillonite. In their study, the basal spacings shifted from 15.7 ,~ (when < 300 mg g-1 were sorbed) to 29.4 ,~ at maximum adsorption of catalase ( ~ 6 7 0 mg g - i of clay). In our experiments the basal spacings of the catalasesmectite complexes tended to increase gradually with the amount adsorbed (Fig. 5). Intercalation of catalase or albumin, which have large molecules and are also negatively charged at pH 7.0, by diffusion into the interlayers of smectites appears particularly difficult. Our results, as well as those of Harter & Stotzky (1973) and of Fusi et al. (1989) could be explained by the mechanism proposed by Larsson & Siffert (1983) for the incorporation of protein molecules into the inter-
Physicochemical properties of smectite complexes CATALASE
331
ALBUMIN
3s.3~
36.6
~.s
~
29
PEPSIN
~7.~ 12.5
LYSOZYME
HECTORITE
~TE '14.9
FIG. 3. X-ray powder diffraction patterns, at 20~ of protein-smectite complexes obtained by adding catalase, albumin, pepsin or lysozyme to hectorite, or to the Uri or Crook montmorillonite at an initial ratio of 1:1 (wt:wt).
layer space of smectites. These authors claimed that the process of radial diffusion into the interlayer space is questionable and demonstrated that the opening of the clay layers, when saturated with Na cations, is easier than a diffusional introduction of bulky molecules into the interlayer spacing. The intercalation of the protein molecules (clay-complex formation) consists of a 'rebuilding' of the clay crystals by an association process of individual protein-covered clay layers. With Na as the exchangeable cation, the dispersion energy necessary to disperse or separate
the clay crystals into small crystallites or individual layers is low and the protein can easily be adsorbed at the surface of the individual clay layers. On the other hand, when a smectite is saturated with highly charged cations (Ca, A1, La, Th) or is coated with OH-A1 or OH-Fe species, incorporation cannot take place easily because the opening of the interlayer spacing becomes more difficult.
Pepsin- and lysozyme-clay mineral complexes. Pepsin filled only hectodte intedayers (Table 2; Figs. 3 and 4) and hectodte-peps!n complexes showed the lowest d-spacing (19.7 A at 20~ of
A. Violante et al.
332 CATALASE - HECTORITE
A L B U M I N - HECTORITE
l&O 20.0
20"C
200( ;
i PEPSIN - HECTORITE
~.7
! LY~YME
- HECTORITE
2~.~ ~ - 2 0 0 ~
FIG. 4. X-ray powder diffraction patterns of catalase, albumin, pepsin or lysozyme-hectorite complexes at 20~ and after heating at 200, 300 or 500~ Initial protein:hectorite ratio 1:1 (wt:wt).
all the complexes studied. The d-basal spacing of the purified hectorite increased from 12.9 to 17,2 .A by adding 1.0 g of protein per g of clay (data not shown). Lysozyme-smectite complexes showed dspacings at 20~ ranging from 22.0 ,~ (Crook montmorillonite) to 23.5 ,& (hectorite). Protein molecules were very well intercalated in the interlayer spaces of the Crook and Uri montmorillonites, as appeared evident from the sharp peaks at 20~ (Fig. 3) and at 200~ (not shown). These
last complexes showed an additional strong peak at 14.8-14.9 A, which could be attributed to unexpanded layers (Larsson & Siffert, 1983). Conversely, the lysozyme-hectorite complex showed only a broad peak centred at 23.5 ,~. After heating from 20 to 300~ all the complexes showed a relatively small decrease in d-spacings (22.5-18.1 A; Table 1). In particular the lysozymehectorite complex remained practically unchanged after heating up to 300~ (Fig. 4). These findings seem to demonstrate that tysozyme molecules are
Physicochemical properties of smectite complexes particularly stable under heating, probably because they are well protected in the interlayers of smectites. Much more lysozyme than albumin was adsorbed on the smectites (Fig. 1), but the d-spacings of lysozyme-smectite complexes were smaller than those of albumin-smectite complexes (Table 1; Fig. 3), evidently because of the greater size of albumin molecules. Armstrong & Chesters (1964) found that a complex containing 930 mg of lysozyme sorbed per ~ of montmorillonite showed a basal spacing of 35 A (A-value of 25 ,~) and attributed this finding to the formation of a single-layer complex with the globular lysozyme molecules (25 A x 32 ellipsoid), having the shorter axis perpendicular to the silicate layer. Larsson & Siffert (1983) also showed that at lysozyme/montmorillonite weight ratios I> 0.52, a basal spacing of only 36-38 A was observed. In our experiments, by adding up to 1.5 g of protein per gram of clay, the maximum amounts of lysozyme adsorbed ranged from 532 mg g-1 of the Uri montmorillonite to 918 mg g-I of the Crook montmorillonite (Fig. 1), but the d-spacings remained substantially similar, 22.0-23.5 A with a Ad-spacing of ~ 10-11 .~ (Table 2), which may be explained by the unfoldi~ of the lysozyme molecules on the clay surfaces in response to strong attractive electrostatic interactions below the iep of the protein. Unpublished data collected by Violante et al. (1995; unpublished data) show a d-spacing of 30-35 A at pH of 11.1 and a lysozyme:Uri montmorillonite weight ratio > 0,8, According to Quiquampoix (1987), unfolding of protein molecules is less extensive at pH values near the iep of protein.
333
CATALASE - MT. CROOK O
16.4 A
g-1
34.1 ~ . ~ 4 2 5
,
,
2
5
mg g-t
m? g-' 10
15
FIG. 5. X-ray powder diffraction patterns, at 20~ of catalase-Crook montmorillonite complexes, containing different amounts of adsorbed protein.
TABLE 3. Basal spacings (.A) of chlorite-like minerals and chlorite-like-mineral-protein complexes obtained by adding 1.0 g of protein per g of clay, at 20~ and after heating at 200~ 300~ or 500~ T~ 20 200 300 500
Ch1-1.5 15,2 14.1 13.7 10.1
20
+Lysozyme
Chl-3
+Pepsin
+Lysozyme
16.6 16,4 14.1 13.4
15.0 14.1 13.9 12.3
15.3 14.6 14.1 t3,4
15.6 14.8 13.9 13.6
Chl-l.5 = chlorite-like complex containing 1.5 mEq of AI per g of hectorite. Chl~3 = chlorite-like complex containing 3 mEq of AI per g of hectorite.
334
A. Violante et al. 380
In our study, only lysozyme molecules were partially intercalated in hectorite interlayers coated by OH-A1 species (chlorite 1.5 and chlorite 3), evidently because this protein has molecules which are much smaller than pepsin, albumin and catalase (Table 3). In fact, in the presence of lysozyme, the chlorite 1.5 showed, at 20~ a d-spacing of 16.6 vs. 15.2 A (A-value of 1.4 A) and at 200~ d-spacings of 16.4 A vs. 14.1 ,~ (A-value of 2.3 A; Table 3), whereas chlorite 3 showed much lower Avalues (0.6 A at 20~ and 0.7 ,~ at 200~ These results indicate that the intercalation of lysozyme decreased sharply by increasing the amounts of the OH-A1 species coating the hectorite surfaces. Our results seem to confirm the findings of Naidja et al. (1995) that tyrosinase did not penetrate into the interlayers of a montmorillonite coated with relatively large quantities of non-crystalline A1 oxides (7.5 or 15 mEq of A1 g-1 of clay), but was intercalated only in an Al(OH)x-montmorillonite complex containing 3 mEq of A1 g-] of clay.
470
695
Thermal analyses o f albumin-clay mineral complexes
The DTA curves of albumin (Fig. 6b) differed from those of a physical mixture of albumin and the Uri montmorillonite (1:1 wt:wt; Fig. 6c) and of the albumin-montmorillonite complex, containing 397 mg of protein adsorbed per g of clay (Fig. 6d). The thermal decomposition of albumin was nearly complete by 650~ Two exotherms appeared at ~350 ~ (very strong) and 520~ (Fig. 6b). The DTA pattern of the physical mixture of protein and montmorillonite showed very broad exotherms or shoulders at 300-450~ and at 470-650~ (Fig. 5c). The pattern of the clayprotein complex showed a huge exotherm centred at 380~ with shoulders at 470~ (strong) and 695~ (Fig. 6d), indicating that the thermal decomposition of protein molecules shifted towards higher temperatures. A comparison of curves b and d in Fig. 6 shows that in the complex a much greater percentage of the protein was decomposed at temperatures ranging from 400 to 650~ or even at higher temperatures (695~ This behaviour could be also attributed to the presence of intercalated molecules. Similar results were obtained by Schnitzer & Kodama (1977) and Buondonno et al. (1989) who studied the nature and physicochemical properties of organo-clay mineral complexes.
690
180 '
I
100
I
300
!
'
I
700 500 TEMPERATURE*C
FIG. 6. Differential thermal analysis curves of (a) Ur montmorillonite, (b) albumin, (c) a physical mixture o: albumin and montmorillonite (1:1, wt:wt) and (d) al albumin-Uri montmorillonite complex, containing 397 mg of protein g-l clay.
CONCLUSIONS Our study shows that catalase, albumin, pepsin, an( lysozyme were adsorbed differently at pH 7.0 ol the Crook and Uri montmorillonites and on ; hectorite. Lysozyme was very easily adsorbed on th( surfaces of smectites as a result of electrostati( interactions between the opposite charges of cla,. surfaces and protein molecules. It was wel
Physicochemical properties of smectite complexes
intercalated in the interlamellar spaces of smectites, but its molecules suffered extensive unfolding. Catalase and albumin, which were negatively charged at pH 7.0, were probably adsorbed by the effect of non-electrostatic forces. These proteins, which also have large molecules, were intercalated at pH 7.0, in spite of their net negative charge. Intercalation of the proteins probably did not occur by a radial diffusion into the interlayers of smectites, but by the opening of the clay layers, which occurs easily with Na as the exchangeable cation, because the amount of dispersion energy necessary to separate the clay crystals into small crystallites or individual layers is small (Larsson & Siffert, 1983). Catalase molecules were intercalated undergoing extensive unfolding. Pepsin was very poorly sorbed on the Crook and Uri montmorillonites. Relatively large amounts of pepsin were sorbed on the surfaces of raw and purified hectorite. Protein-smcctite complexes showed different behaviour under heating. Proteins were poorly adsorbed on Al(OH)x-smectite complexes. Only lysozyme molecules were interlayered in smectites, partially coated by OH-A1 species. Intercalation of the lysozyme decreased by increasing the quantities of the OH-A1 species coating the surfaces of clay minerals. The OH-A1 species present in the interlayers inhibited the interstratification of protein molecules by steric factors or because the opening of the interlayer spacing of smectite was made difficult by the clay surfaces being coated with hydrolytic products of A1.
ACKNOWLEDGMENT This research was supported by the National Research Council of Italy, Special Project RAISA, Subproject no. 2, Paper no. 2131, Contribution no.103 from Dipartimento di Scienze Chimico Agrarie (DISCA). REFERENCES ARMSTRONGD.E. & CHES~RS G. (1964) Properties of protein-bentonite complexes as influenced by equilibration conditions, Soil Sci. 98, 39-52. BARNHISELR.I. & BERTSCHP.M. (1989) Chlorites and hydroxy interlayered vermiculite and smectite. Pp. 729-778 in: Minerals in Soil Environments (J.B. Dixon & S.B. Weed, editors), 2nd edn,. Soil Science Society of America, Madison, Wisconsin.
335
BARROUG A,, LEMAITRE J. & ROUXHET P.G. (1989) Lysozyme on apatites: A model of protein adsorption controlled by electrostatic interactions. Colloids Surf. 37, 339-355. BORCHARDT G. (1989) Smectites. Pp. 675-727 in: Minerals in Soil Environments (J.B. Dixon & S.B. Weed, editors), 2nd edn. Soil Science Society of America, Madison, Wisconsin. BUONDONNO A., FELLECA D. & VIOLANTE A. (1989) Properties of organo-mineral complexes formed by different addition sequences of hydroxy-Al, montmorillonite, and tannic acid. Clays Clay Miner. 37, 235-242. FusI P., RISTORTG.G., CALAMAIL. & STOTZKYG. (1989) Adsorption and binding of protein on 'clean' (homoionic) and 'dirty' (coated with Fe oxyhydroxides) montmorillonite, illite and kaolinite. Soil Biol. Biochem. 21, 911-920. GIANFREDA L., RAG M.A. & VIOLANTE A. (199l) Invertase ([3-fructosidase): Effects of montmorillonite, Al-hydroxide and Al(OH)x-montmorillonite complex on activity and kinetic properties. Soil Biol. Biochem, 23, 581-587, GIANFREDA L., RAG M.A. & VIOLANTE A. (1992) Adsorption, activity and kinetic properties of urease on montmorillonite, alluminium hydroxide and Al(OH)x-montmorillonite complexes. Soil Biol. Biochem. 24, 51-58. GILES C.H., SMITHD. & HtIITSONA. (1974) A general treatment and classification of the solute adsorption isotherm. I. Theoretical. J. Colloid Interface Sci. 47, 755-765. HARTER R.D. (1975) Effect of exchange cations and solution ionic strength on formation and stability of smectite-protein complexes. Soil Sci. 120, 174-181. HARTERR.D. & STOTZKYG. (1971) Formation of clayprotein complexes. Soil Sci. Soc. Am. Proc. 35, 383-389. HARTER R.D. & STOTZKYG. (1973) X-ray diffraction, electron microscopy, electrophoretic mobility, and pH of some stable smectite-protein complexes. Soil Sci. Soc. Am. Proc. 37, 116-123. HUANGP.M. & VIOLANTEA. (1986) Influence of organic acids on crystallization and surface properties of precipitation products of aluminum. Pp. 159-221 in: Interactions of Soil Minerals with Natural Organics and Microbes (P.M. Huang & M. Schnitzer, editors),
Soil Science Society of America, Spec. Pub. 17, Madison, Wisconsin. LARSSON N. & SIFFERT B. (1983) Formation of lysozyme-containing crystals of montmorillonite. J. Coll. Interf. Sci. 93, 424-431. MCKEAGUEJ.A. (1978) Manual on Soil Sampling and Methods of Analysis. Pp. 212. Canadian Society of Soil Science Ottawa. MCLAREr~A.D., PETERSONG.H. & BARSHaDI. (1958) The adsorption and reactions of enzymes and proteins on
336
A. Violante et al.
clay minerals: IV. Kaolinite and montmorillonite. Soil Sci. Soc. Am. Proc. 22, 239-244. NAIDJA A., VIOLANTE A. & HUANG P.M. (1995) Adsorption of tyrosinase on montmorillonite as influenced by hydroxyaluminum coatings. Clays Clay Miner. 43 (in press). NORDEW. (1986) Adsorption of proteins from solutions at the solid-liquid interface. Adv. Coll. Interf. Sci. 25, 267-340. NORDE W. & LYKLEMA J. (1978) The adsorption of human plasma albumin and bovine pancreas ribonuclease at negatively charged polystyrene surfaces: I. Adsorption isotherms. Effects of charge, ionic strength, and temperature. J. Coll. Interf Sci. 66, 257-265. QUIQUAMPOIXH. (1987) A stepwise approch to the
understanding of extracellular enzyme activity in soil: I. Effect of electrostatic interactions on the conformation of a I3-D-glucosidase adsorbed on different mineral surfaces. Biochimie, 69, 753-763. QUIQUAMPOIXH. & RATCLIFFER.G. (1992) A 31p NMR study of the adsorption of bovine serum albumin on montmorillonite using phosphate and the paramagnetic cation Mn2+: Modification of conformation with pH. J. Coll. Interf. Sci. 148, 343-352. SCHNITZER M. & KODAMA H. (1977) Reactions of minerals with soil humic substances. Pp. 741-770 in: Minerals in Soil Environments (J.B. Dixon & S.B.Weed, editors). Soil Science Society of America, Madison, Wisconsin. THENG B.K.G. (1979) Formation and Properties of Clay-Polymer Complexes. Elsevier, New York.
