Blotechnol. hog. 1992, 8, 179-186
170
ARTICLES ~
Tyrosinase Reaction/Chitosan Adsorption for Removing Phenols from Wastewater Wei-Qiang Sun,Gregory F. Payne,*Monica S. G. L. Moas,?Jennifer H. ChuJ and Kimberlee K. Wallace Department of Chemical and Biochemical Engineering and Center for Agricultural Biotechnology, University of Maryland, Baltimore County, Baltimore, Maryland 21228
A two-step approach for removing phenols from aqueous solutions was investigated. In the first step, weakly adsorbable phenols are converted to quinones by the enzyme mushroom tyrosinase. The tyrosinase-generated quinones are then chemisorbed onto chitosan, a readily available waste product of the shellfish industry. In the absence of enzyme, quinone was observed to be rapidly adsorbed onto chitosan. Also, the enthalpy for quinone adsorption onto chitosan was observed to be -24.7 kcal/mol, which compares to enthalpies of -7 kcal/mol for adsorption of phenols and quinone onto activated charcoal. With the monophenol reactant cresol, the tyrosinase enzyme was observed to be somewhat stabilized in the presence of chitosan. This stabilization of tyrosinase is presumably due to the rapid adsorption of the reactive quinones onto chitosan. In contrast, tyrosinase was not stabilized by chitosan when the o-diphenol catechol was the reactant. The ability of chitosan to stabilize tyrosinase for monophenols but not for o-diphenols is discussed in terms of the relative rates of phenol oxidation by tyrosinase and quinone chemisorption onto chitosan. When mushroom tyrosinase and chitosan were added simultaneously to dilute, phenol-containing solutions, a nearly complete removal of UV-absorbing material was observed. This observation demonstrates the feasibility of removing phenols from dilute solutions using the tyrosinase reaction/chitosan adsorption approach.
Introduction Phenols represent one of the most important classes of synthetic industrial chemicals and are often observed in effluents from various manufacturing operations (Keith and Telliard, 1979). Further, phenols are common components of pulp and paper wastes and have been observed as groundwater contaminants. Despite the importance of treating phenol-containing wastewaters, current methods are not satisfactory and the search for improved methods continues (Lanouette, 1977). Physical (e.g., adsorption) and chemical (e.g., oxidation) treatment methods become exceedingly expensive when low effluent concentrations must be achieved. For instance, in many wastewater treatment applications,the activated carbon requirements have dramatically increased (Irving-Monshaw, 1990)due to the need to meet more stringent effluent standards. Microbiological treatment approaches have shown great promise for treating phenolic wastes (e.g., see Donaldson et al., 1987,and Worden and Donaldson, 1987). However, for small volume wastes which are generated discontinuously, microbiological treatment has been plagued by instabilitiesresulting from the toxicity of these compounds to the microbial population (Jones et al., 1973;Yang and Humphrey, 1975). ~~
~
~
Present address: Escola Superior de Biotechnologia, Rua Dr. Antonio Bernardino de Almeida, 4200 Porto, Portugal. t Present address: Department of Chemical Engineering, Princeton University, Princeton, NJ 08544-5263. t
87567938/92/3008-0179$03.00/0
The idea of using enzymes for treating phenol-containing wastewaters was first proposed in the 1980s. Early workers suggested that enzymes could be used to convert soluble phenols into insoluble polyphenolic precipitates which could be removed by filtration. Enzymes considered for this phenol precipitation included peroxidases (Klibanov and Morris, 1981;Klibanov et al., 1980,1983;Aitken et al., 19891,laccases (Shuttleworth and Bollag, 1986) and tyrosinases (Atlow et al., 1984). The advantages of this enzymaticapproach are that these enzymes can react with a wide range of phenols even under dilute conditions and that these enzymes are likely to be less sensitive to operational upsets than microbial populations. Despite these advantages, interest in the use of enzymes to precipitate phenols from wastewaters appears to have diminished. The waning interest in enzymatic phenol precipitation appears to result from three rather serious problems. These problems are illustrated by considering the enzymatic and subsequent nonenzymatic reactions which occur in the presence of tyrosinase:
-
tyrosinase
phenol
(fast)
- -
-
slow
o-quinone other intermediates oligomers polyphenols (4) (1) Quinones are rapidly formed from the tyrosinase-catalyzed oxidation of phenols. These quinones are reactive and undergo nonenzymatic conversion to form additional, more stable intermediates. These more stable intermediates
0 1992 American Chemical Society and American Institute of Chemical Engineers
Biotechnoi. Pmg., 1992, Vol. 8, No. 3
180
slowly undergo oligomerizationreactions which ultimately can yield high molecular weight, insoluble polyphenolics. The first problem with the enzymatic phenol precipitation approach is that the oligomerization and precipitation reactions are generally slow, requiring hours to days for completion. The second problem is that unless initial phenol concentrations are high, oligomerization may be limited to the formation of low molecular weight oligomers, which do not precipitate but rather remain in solution. The final problem is that these enzymes can be inactivated by the reactive intermediates generated from their reactions (Atlow et al., 1984; Canovas et al., 1987). Thus, despite the advantage that the enzymes can react with a range of phenolics even under dilute conditions, the previously-proposed approach of using enzymes for phenol precipitation has serious practical limitations. To overcome the limitations noted above, a two-step approach was examined. In the first step, tyrosinase is used to generate reactive intermediates, which are then strongly adsorbed onto specific sorbents. This proposed approach is illustrated by the following: Step 1: Tyrosinase Reaction phenol
-
o-quinone
+ other intermediates
(2)
Step 2: Chemisorption
-
o-quinone + other intermediates + sorbent chemisorbed compounds (3) The first step involves tyrosinase, a copper-containing enzyme capable of catalyzing two reactions (see Butt and Lamb, 1981). The first reaction catalyzed by tyrosinase is a monooxygenation where one atom of molecular oxygen is incorporated into the aromatic structure and the reducing reagent AH2 is oxidized: D
R
O
H
+ 02 +
AH2
a:; +
4
R
H20
+
A
(4)
Although alternative reducing agents have been used to increase reaction rates, the o-diphenol generated from the above reaction can act as the reductant (Vaughan and Butt, 1970; McIntyre and Vaughan, 1975). Thus, no complex, external cofactors are required by this enzyme (see Ito et al., 1991). Tyrosinase can also catalyze a second reaction, the oxidation of o-diphenol to o-quinone:
In the second step of this two-step treatment approach, the reactive intermediate generated from the tyrosinasecatalyzed oxidation (Le., the quinone) is adsorbed. The key to this treatment approach is that the sorbent must be readily available and effective. Although several readily available materials could be examined as reactive sorbents, we studied the polyglucosamine chitosan, which is derived from chitin. Chitin is an abundant natural polymer, and there is considerable interest in developing chitosan-based products because geographic areas with large shellfish industries are experiencing considerable problems disposing of the overabundant, chitin-rich shells. Thus, the sorbent used here is derived from a waste. The rationale for using chitosan as a sorbent is not merely that chitosan is a waste material but rather that there are specific chemical reactions between chitosan and the ty-
25.97
25.79
I
I
I
0
2
4
I
I
I
6
8
10
1 12
Time (min)
Figure 1. Typical thermogram used to determine the temperature rise (AT,) for adsorption. In this experiment, chitosan (12 g) was added to an aqueous solution (100 g) containing 5 mM p-quinone. The temperature rise (ATc)observed in this experiment was 0.036 "C,which compares to 0.0085 O C observed if chitosan was added to distilled water without p-quinone.
rosinase-generated quinones which can be exploited to strongly adsorb these compounds. Although the chemistry has not been completely established, it appears that the amine in the chitosan structure reacts with compounds containing carbonyl groups to form covalent bonds (Milun, 1957; Hall and Yalpani, 1980; Muzzarelli et al., 1982). In a study involving the binding of amines to carbonyl- and quinone-containing organics, Parris (1980) suggested that two reaction paths are possible, a rapid reversible reaction resulting in imine formation and an irreversible oxidative reaction path resulting in the formation of aminoquinones. The goal of this study was to examine the tyrosinase reaction/chitosan adsorption approach for removing phenols from dilute aqueous solutions. Specifically, we examined (i) the strength of quinone-chitosan adsorption, (ii) the effect of chitosan on tyrosinase activities, and (iii) the capabilities of this two-step approach for removing phenols.
Materials and Methods Materials. Mushroom tyrosinase (EC 1.14.18.1) of specific activity 1050 unitslmg (activity determined by supplier) was purchased from Worthington Biochemicals (Freehold, NJ). Chitosan from crabshells and activated charcoal were used as adsorbing surfaces and were purchased from Sigma Chemical Co. (St. Louis, MO). In addition to phenol, we studied p-cresol and pyrocatechol (1,2-benzenediol). These phenols and all other chemicals used in this study were obtained commercially and were of analytical grade. Methods. Adsorption heat (8) was measured using an adiabatic solution calorimeter (Parr Instrument Company, Moline, IL). A typical thermogram is shown in Figure 1, where it can be seen that there is a sharp increase in temperature after adding sorbent. The temperature difference, AT,, between the two baselines is used to calculate the adsorption heat using the equation
Q = S ( A T , - ATo)
(6) where AT0 accounts for the heat of wetting and is the temperature rise observed in a control where sorbent was contacted with distilled water containing no solute. The AT0 was observed to be 0.0085 OC for chitosan and was negligible for activated charcoal. S is the heat capacity
of the system and is given by
S = CPimi+ Cpsms+ S, (7) C,i is the heat capacity of the solution (1.0 cal/g-"C), mi is the mass of the solution (100 g), S, is the heat capacity of the calorimeter (21.4 cal/"C), C,, is the specific heat capacity of the sorbent, and m, is the mass of sorbent. Although an accurate measure of C,, is unavailable, the second term on the right hand side of eq 7 is typically small (Maityet al., 1991). The adsorption enthalpy (AH") was determined from the slope of the adsorption heat versus amount of solute adsorbed. It should be noted that calorimetric studies were done in distilled water (and not buffered solutions) to avoid the large temperature rise observed when chitosan was contacted with the phosphate buffer. Liquid-phase quinone and phenolic concentrations were measured or estimated using ultraviolet spectrophotometry (Gilford Response, Oberlin, OH). When liquid-phase concentrations were measured, the wavelength of maximum absorption was used. For adsorption studies, the amount of solute adsorbed was calculated from the difference between the initial (prior to sorbent addition) and final (after equilibration) solute concentrations (Co and C , respectively) using the equation amount adsorbed = (Co- C)VL (8) where VL is the solution volume. In studies involving enzymatic reaction, mushroom tyrosinase was added to 50 mM phosphate buffer (pH 6.8) containing a single phenolic. In studies involving both enzymatic reaction and chitosan adsorption, mushroom tyrosinase and chitosan were added simultaneously to a 50 mM phosphate buffer (pH 6.8) containing a single phenolic. Reactions were conducted at 25 or 27 "C in bottles containing small volumes of liquid (either 10 or 25 mL), which were agitated either by shaking or by using a stir bar. The reaction kinetics were measured by monitoring changes in dissolved oxygen (DO) using a dissolved oxygen meter (Microelectrodes, Londonderry, NH). Results Quinondhitosan Adsorption. To characterize chitoean adsorption, we performed calorimetric studies in which chitosan was contacted with aqueous solutions containing quinone. Because the o-quinone which is generated from the tyrosinase reaction is unstable and not readily available, we chose to use p-quinone for our adsorption studies. After contact of chitosan with a 5 mM p-quinone solution, Figure 1 shows that the temperature in the calorimeter increased above the initial, preadsorption baseline. Approximately 1 min after chitosan addition, the temperature rise slowed and a second baseline was observed. The temperature rise (AT,) of 0.036 "Cobserved in Figure 1can be converted into an adsorption heat by eq 6. In addition to measuring adsorption heat, we measured the aqueous-phase quinone concentration before and after adsorption to determine the amount of quinone adsorbed (Le., by eq 8). Figure 2 shows that when varying initial quinone concentrations were studied, the adsorption heat increased linearly with the amount of quinone adsorbed. From the slope-of Figure 2, the adsorption enthalpy for quinone adsorption onto chitosan was calculated to be -24.7 kcal/mol. Adsorption enthalpies of this magnitude are generally believed to result from strong, presumably covalent interactions and such adsorption is referred to as chemisorption. In summary, the results in Figures 1and 2 demonstrate that quinone adsorption onto chitosan is rapid (on the order of minutes) and strong.
, /ph 181
Blofechnol. Prog., 1992, Vol. 8, No. 3
2o
c
pyrocatechol-
activated charcoal
t
5 1
I
0
I
I
0.5 1 1.5 Solute adsorbed concenuation ("ole)
c
I
2
1
2.5
Figure 2. Adsorption heat versus amount of solute adsorbed for adsorption onto chitosan and activated charcoal. Solute adsorption was varied by varying the initial concentration of phenol or quinone in the calorimeter. With chitosan, initial p-quinone concentrations were varied between 3 and 10mM. With activated charcoal, initial solute concentrations were varied between 10 and 20 mM. Table I. Adsorption Enthalpies of Phenol, Pyrocatechol, and &Quinone with Activated Charcoal and of &Quinone with Chitosan adsorption enthalpy (AH"),kcal/mol solute activated charcoal chitosan phenol -6.4 pyrocatechol -6.7 p-quinone -7.3 -24.7
For comparison, we examined the adsorption of phenols and quinones onto activated charcoal. As shown in Figure 2 and Table I, considerably less heat was evolved for adsorption onto activated charcoal. The observed enthalpies for phenols and p-quinone adsorption onto activated charcoal were between to be -6 and -7 kcal/mol, which suggests that adsorption onto activated charcoal involved low-energy physical forces (e.g., hydrophobic interactions). It should be noted that since the phenols are not adsorbed onto chitosan (discussed later), adsorption enthalpies for these solutes onto chitosan are not reported in Table I. Effect of Chitosan on Tyrosinase Activities. Stoichiometry. To examine the stoichiometry of the tyrosinase reactions, we used a dissolved oxygen probe to measure oxygen consumption and we set experimental conditions of low initial phenol concentrations. As shown in Figure 3, the dissolved oxygen decreased over time until a constant value was reached. Presumably, the tyrosinase reaction ceased a t these latter times due to the depletion of the phenol reactant. When higher initial phenol concentrations were used, Figure 3 shows that the final dissolved oxygen concentration was lower. Plots of the total oxygen consumed versus the initial phenol concentration are shown in Figure 4. The fact that these plots are linear supports the contention that the phenols were the limiting reactants in these studies. Thus the slopes of Figure 4 were used to determine the stoichiometric coefficients between oxygen and phenol consumption. For the stoichiometric coefficients listed in Table 11, three observations can be made. First, the ratio of oxygen consumed per initial phenol is greater for the monophenol cresol than for the diphenol pyrocatechol. Qualitatively, this difference is expected since monophenols can undergo two oxidative reactions [(4) and (511, while o-diphenols can only undergo the single oxidative reaction (5). A second observation is that the stoichio-
Blotechnol. Prog., 1992, Vol. 8, No. 3
182 I
0.3
I
I
Table 11. Stoichiometric Ratio of Oxygen Consumption per Mole of Phenol Oxidized in the Presence and Absence of Chitosan’
1
Initial cresol concentration
-m-
0.1 mM
--C
0.15 mM
stoichiometric coefficients, ratio of mol of Odmol of phenol w/o w/ coefficients with and phenol chitosan chitosan without chitosan p-cresol 0.94 1.24 1.32 pyrocatechol 0.84 1.15 1.37 a Experimental details are discussed in the legends of Figures 3 and 4. ~~
0.3 0
5
0
10
15
1
I
I
I
I
Time (min)
0.25
’ 14\\\t
1
catechol w/o chitosan
--t
calechol w/ chitosan
I
p-cresol w/o chitosan p-cresol w/ chtosan
--C--
phenol wio chitosan
----e phenol
O”
0.1
I
--c--
-
20
Figure 3. Dissolved oxygen versus reaction time for the tyrosinase-catalyzed oxidation of cresol. Reactions with differing initial cresol concentrations were conducted in buffered solutions (50 mM phosphate) at a pH of 6.8 with 32 units/mL tyrosinase. Chitosan was not added in this experiment.
~~~~~
w/ chilosan
L I
‘B
r
0.2 0 v
7 E
0.15
I
8c
0.1
E
8
0.05
U w/o chitosan w/ chitosan
0
0
I
I
I
I
0.05
0.1
0.15
0.2
Initial cresol concentration (mM) I
0.2
1
I
I
b
Pyrocatechol
B;
0.1
8
Jr
4
-
0.05
w/o chitosan
L
chitosan
--t w/
I
0
.
0.25
I
0.05
I
I
0.1 0.15 0.2 Initial pyrocatechol concentration (mM)
1
0.25
Figure 4. Total oxygen consumption versus initial substrate concentration for the tyrosinase-catalyzedreaction in the presence (w/) and absence (w/o) of chitosan. Reactions were conducted in buffered solutions (50 mM phosphate) at pH 6.8 with either (a) the monophenol cresol or (b) the o-diphenol pyrocatechol. When used, chitosan was added initially a t a level of 5% (w/v). Tyrosinase activities were 32 units/mL for cresol and 20 units/ mL for pyrocatechol experiments.
metric coefficients observed in Table I1 cannot be quantitatively predicted from the reaction stoichiometries shown in reactions (4) and (5). Differences between expected and observed oxygen stoichiometries for tyrosinase-catalyzed reactions are not uncommon and oxygen to phenol ratios similar to those shown in Table I1 have been observed in other studies (Wright and Mason, 1946; Mason and Wright, 1949; Horowitz and Shen, 1952; Daw-
I
0
10
20
I
I
I
31) Time (min)
40
50
60
Figure 5. Dissolved oxygen versus reaction time for the tyrosinase-catalyzed oxidation of pyrocatechol, phenol, and cresol in the presence (w/) and absence (w/o) of chitosan. Reactions were conducted in buffered solution (50 mM phosphate) a t pH 6.8 with 55 units/mL tyrosinase and an initial phenolic concentration of 0.5 mM. When used, chitosan was added initially at a level of 5% (w/v).
son and Tarpley, 1963; Mayer et al., 1966). It should also be noted that, in these other studies, oxygen to phenol ratios were observed to vary with reaction conditions,initial phenol concentration, and the actual products formed from the reaction. It is possible that variation in this stoichiometric coefficient results from differences in the types and degrees of oxidation of the polyphenolic speciesformed from the reactions. The final observation in Table I1 is that more oxygen was consumed per phenol when chitosan was present. Although the reason for this increased oxygen consumption in the presence of chitosan is unknown, it is possible that more oxidized species are adsorbed by chitosan. This possibility is supported by a couple of additional observations. When p-quinone was adsorbed onto chitosan in the absence of tyrosinase, no oxygen consumption was observed. Also, if tyrosinase was added afterp-quinone was adsorbed to chitosan, no oxygen consumption was observed. Thus, it seems unlikely that the increased oxygen consumption in the presence of chitosan could be explained by an oxygen requirement either during or after adsorption. Kinetics: High Enzyme Activities. Figure 5 shows oxygen consumption when relatively high levels of tyrosinase (55unita/mL) and phenols (0.5 mM) were used. When the phenolic substrate was the o-diphenol pyrocatechol, Figure 5 shows the dissolved oxygen decreased rapidly. For the monophenols phenol and cresol, a more complex kinetic pattern was observed. For these monophenols, a low initial rate was followed by a somewhat greater rate. Thislow initial rate is often termed a “lag”and it is believed that this lag results because of the need to generate reductant for reaction (4) (Vaughan and Butt, 1970; Duckworth and Coleman, 1970; Lerch and Ettlinger, 1972; McIntyre and Vaughan, 1975). It is reasoned that when sufficient levels of the diphenol reductant are generated
Bbtechwl. Prog.., 1992, Vol. 8, No. 3
Table 111. Oxygen Consumption Rates for Tyrosinase-CatalyzedOxidation of pCresol and Ratio of These Rates in the Presence and Absence of Chitosan When Different Levels of Tyrosinase Were Used* rate of oxygen ratio of consumption, "0UL.h oxygen consumption W/ rates with and tyrosinase w/o chitosan chitosan without chitosan activities 14 units/mL 0.82 1.11 1.35 55 unite/mL 2.04 3.11 1.52 a Experimental details are discussed in the legend of Figure 5.
by the first reaction, the lag is overcome and the reaction rate increases. However, Figure 5 shows that the rate of reaction for the monophenols always remains considerably less than that for the o-diphenol. In the presence of chitosan it can be seen that the rate of oxygen consumption for the o-diphenol pyrocatechol was not significantly altered. In contrast, the rate of oxygen consumption for the reaction of the monophenols phenol and cresol was enhanced in the presence of chitosan. To quantify this enhancement, we estimated the oxygen consumption rate for cresol over the region where the dissolved oxygen decreased nearly linearly over time (Le., the reaction rates were nearly constant between 0.2 and 0.04 mM dissolved oxygen). This estimate as well as those obtained when a somewhat lower enzyme level were used are reported in Table 111. It can be seen that the ratio of the oxygen consumption rates in the presence and absence of chitosan varies between 1.3 and 1.5. This ratio is similar to the ratio of the oxygen stoichiometries in the presence and absence of chitosan, which was observed to be 1.3 (Table 11). Thus, despite an increase in the rate of oxygen consumption, if differences in oxygen stoichiometries are accounted for, it appears that the reaction of cresol may actually not be enhanced in the presence of chitosan. At worst, the results in Figure 5 and Table I11 demonstrate that chitosan does not inhibit tyrosinase activities. Kinetics LowEnzyme Activities. To further explore the effect of chitosan on tyrosinase performance, we examined the reaction when relatively low levels of tyrosinase (1.4units/mL) and high levels of phenols (0.5 mM) were used. Under these conditions, we expect the suicide inactivation of tyrosinase to be most evident. Suicide inactivation appears to result because the quinones formed from the tyrosinase-catalyzed reaction are capable of reacting with and inactivating the tyrosinase enzyme. For the case of the o-diphenol pyrocatechol, Figure 6 shows the initial oxygen consumption rate to be similar in the presence and absence of chitosan. Over time, however, the reaction of this diphenol stops, despite the presence of both reactants. This cessation in reaction is believed to be due to inactivation of tyrosinase. Figure 6 shows that, in the presence of chitosan, the total amount of oxygen consumed increased (0.20 versus 0.18mM oxygen consumed). However, because of differencesin the oxygen stoichiometries (Table 11))it is possible that the actual oxidation of pyrocatechol was not enhanced in the presence of chitosan. Thus chitosan does not appear to limit tyrosinase inactivation for the case of this o-diphenol. Wealso examined the reaction of the monophenol cresol. In the absence of chitosan, Figure 6 shows that, after a lag, a period of rapid reaction was followed by a period where the reaction slowed and ultimately stopped. Again, it is believed that the cessation in reaction resulted from the inactivation of tyrosinase. In the presence of chitosan, the reaction with cresol was observed to proceed faster and over a longer period. Even when differences in the stoichiometries of oxygen consumption are accounted for,
, h 183
0'3
0.2
e
I
I
50
0
I
Time (min)
100
150
Figure 6. Dissolved oxygen versus reaction time for the tyrosinase-catalyzed oxidation of pyrocatechol and cresol in the presence (w/)and absence (w/o) of chitosan. Reactions were conducted in buffered solution (50 mM phosphate) a t pH 6.8 with 1.4 unita/mL tyrosinase and an initial phenolic concentration of 0.5 mM. When used, chitosan was added initially a t a level of 5% (w/v).
considerably more reaction occurred in the presence of chitosan. Thus, it appears that, for the monophenolic substrate, chitosan was able to reduce tyrosinase inactivation.
Qualitative Kinetic Model. To assist in interpreting the above results, we propose the following qualitative reaction scheme: monophenol
rl
r2
rs
diphenol- o-quinone chemisorbed species (9) where the monophenol is converted to the o-diphenol at rate r l , the o-diphenol is converted to the o-quinone at r2, and the o-quinone is chemisorbed at r3. Under the conditions studied here, Figures 5 and 6 show that the second reaction occurs over the course of seconds to a few minutes, while the first reaction occurs over the course of several minutes to hours. The calorimetric study (Figure 1) suggesb that chemisorption may occur at a rate intermediate between the two enzymatic reactions. From this analysis, it appears that r2 > r3 > rl
(10)
The relative rates proposed above are reasonably supported from oxygen consumption measurements in the presence and absence of chitosan (Figures 5 and 6). If, as proposed above, r2 > r3, then the reaction of the o-dipheno1 would result in a significant transient accumulation of the intermediate o-quinone. In the extreme of r2 >> r3, the initial oxygen consumption rate for the reaction of the o-diphenol should reflect only the second reaction and be independent of the presence or absence of chitosan. Figures 5 and 6 show that, for pyrocatechol oxidation, the initial oxygen consumption rate in the presence and absence of chitosan is similar. Small differencesin oxygen consumption rates and differences in the oxygen consumption stoichiometries in the presence and absence of chitosan (Table 11)could result in differences in the final amount of oxygen consumed for pyrocatechol oxidation. Such differences were observed in Figure 6. If rl is much less than either r2 or r3, then the first oxidation step would be rate limiting such that the monophenols would be chemisorbed as soon as they were oxidized. In this case,the initial oxygen consumption rates for monophenol oxidation should reflect all the reactions. Thus, the initial rate of oxygen consumption for reaction
Biotechnol. Prog., 1992, Vol. 8, No. 3
104
of the monophenols should vary in the presence and absence of chitosan due to differences in the oxygen consumption stoichiometries (Table 11). This difference in initial oxygen consumption rate was observed in Figures 5 and 6 . Since the proposed reaction scheme and relative rates are qualitatively capable of explaining the various observations, we used this qualitative model to suggest a reason why chitosan can stabilize tyrosinase for monophenolic but not for diphenolic substrates. For the monophenolic substrates, the above model suggests that since the first reaction is slow, then low levels of intermediate quinones will accumulate in the presence of chitosan. Since quinones have been reported to cross-link proteins (Leatham et al., 1980) it is possible that these intermediates are a significant cause of suicide inactivation of tyrosinase. Thus it is possible that chitosan stabilizes tyrosinase by limiting the accumulation of quinones. In contrast, since rz is proposed to be greater than r3,when diphenolic substrates are used, quinones will be produced faster than they can be adsorbed. Again, if quinones are a major cause of tyrosinase inactivation, then the inability of chitosan to prevent quinone accumulation with the diphenolic substrates may explain chitosan’s inability to stabilize the enzyme with pyrocatechol. In summary, results from our kinetic studies can be qualitatively explained by the reaction scheme and relative rates proposed above and by the assumption that the accumulation of intermediate quinones accelerates tyrosinase inactivation. To more rigorously verify the proposed model it would be necessary to characterize the reaction by measuring phenolic reaction intermediates and rates and not simply by measuring changes in dissolved oxygen as was done here. However, given the fact that numerous phenolic intermediates (including oligomers) can be generated by the tyrosinase-catalyzed and subsequent nonenzymatic reactions, characterization of the reaction path by identifying and quantifying intermediate phenolic species may prove difficult. Thus, the qualitative model provides a simple means to explain the kinetic data for this complex, if not intractable, reaction system. Repeated Use of Tyrosinase for Cresol Oxidation. To further examine the ability of chitosan to stabilize tyrosinase for monophenolic substrates, we conducted alongterm study with high initial tyrosinase activities (55 units/ mL) and repeated cresol additions. Figure 7a shows results when 0.2 mM cresol was initially added to an oxygensaturated solution containing tyrosinase. After addition of cresol, the dissolved oxygen decreased rapidly until approximately 0.06 mM oxygen remained. From stoichiometric considerations (Table 11), this level of oxygen consumption corresponds to the complete conversion of cresol. When the dissolved oxygen reached 0.06 mM, air was bubbled through the solution for half an hour to obtain oxygen saturation. After air was bubbled through, 0.2 mM cresol was added, and the reaction was allowed to continue for a second cycle. As shown in Figure 7a, the reaction rate decreased with each subsequent reaction cycle until the reaction essentiallyceased during the fourth cycle. Figure 7b shows results from a similar experiment where chitosan was added to the reaction cycle. Because of differences in oxygen stoichiometries, the dissolvedoxygen was allowed to decrease from saturation (0.25 mM) to 0.01 mM to permit complete conversion of cresol in the presence of chitosan. As seen in Figure 7b, the tyrosinase-catalyzed reaction rate was not reduced until the third cycle and the reaction continued through the fifth cycle. For comparison, Table IV lists the reaction rates estimated from the individual cycles for reactions in the presence and absence of chitosan. Again chitosan enhanced the reaction with
I
0.3
I
I
I
I
1
I
a
wto chitosan
0 0
10
20
30
40
50
60
I
I
70
Time (min)
0.3
I
1
I
I
b
~
w/ chitosan
L
0.2
0.1
0
0
10
20
30 40 Time (min)
50
60
70
Figure 7. Dissolved oxygenversus reactiontime for the repeated reaction of cresol with tyrosinase. Reactions were conducted in buffered solution (50 mM phosphate) at pH 6.8 with 55 units/ mL tyrosinase and an initial cresol concentration of 0.2 mM. (a) When chitosan was not added, the dissolved oxygen was allowed to be depleted from 0.25 to 0.06 mM, after which time air was bubbled through the liquid to resaturate the solution and then 0.2 mM cresol was added. (b) When chitosan [ 5 % (w/v)] was added, the dissolved oxygen was allowed to be depleted from 0.25 to 0.01 mM, after which time air was bubbled through the liquid to resaturate the solution and then 0.2 mM cresol was added. Table IV. Oxygen Consumption Rates for Tyrosinase-Catalyzed Oxidation of pCresol for Repeated Reactions in the Presence and Absence of Chitosan. rate of oxygen consumption,mmol/L.h reaction cycles w/o chitosan w/chitosan first cycle 1.92 2.52 second cycle 1.38 2.46 third cycle 1.02 1.92 fourth cycle 0.49 1.32 fifth cycle 0.72 a Data obtained from Figure 7.
cresol, due presumably to the stabilization of tyrosinase by removal of the reactive quinones. Tyrosinase Reaction1Chitosan Adsorption. From the previous results, it can be seen that p-quinone is strongly bound to chitosan, that chitosan does not adversely affect the activities of tyrosinase, and that for monophenolic reactants chitosan may even stabilize tyrosinase. Our next goal was to demonstrate that the twostep, tyrosinase reactionlchitosan adsorption operation was capable of removing phenols from solution. For these studies, we examined solutions containing either a monophenol (phenol or cresol) or the o-diphenol pyrocatechol. Curves A of Figure 8 show the ultraviolet (UV)
&techno/. Rog., 1992, Vol. 8, No. 3
185
Phenol
3
250
310
370 430 WAVELENGTH (nm)
490
550
(not shown), presumably due to oligomerization reactions and polyphenol formation. The UV absorbances of the phenol solutions after incubation for 2 h with tyrosinase are shown in curves B. If, instead of enzyme, chitosan was contacted with the phenolic solutions (curves C), there was no shift but rather a small increase in the UV absorbance as compared to the controls (i.e., curves A). Thus chitosan was unable to catalyze the conversion of the phenols (asexpected), and also, chitosan was unable to adsorb the phenols from these dilute solutions. When the phenol solutions were simultaneously contacted with mushroom tyrosinase and chitosan (curves D), it can be seen that there was nearly a complete reduction in UV absorbances. Thus these results show that the combination of tyrosinase reaction and chitosan adsorption was capable of removing UV-absorbing material (i.e., phenols and quinones) from dilute solutions.
Discussion
r I
250
A
!
-7 0
310
I
370 430 WAVELENGTH (MI) I
I
I
I
490
550
affinity e-mIRT (11) where the adsorption affinity is shown to increase exponentially with increases in the adsorption strength (i.e., AH")(Maityet al., 1991). An increased adsorption affinity is particularly important because increasingly stringent effluent standards mandate that phenols be removed from more dilute wastewaters. In our studies, we observed enthalpies of -24.7 kcal/mol for chitosan and -7 kcal/mol for activated charcoal adsorption. For an order of magnitude comparison, if we assume the proportionality in relation 11is similar for chitosan and activated charcoal adsorption, then Qc
I
I
1
(affinityIchi,= e - ( A € Z o c u t - A € P ~ ) / R T
q&
(affinity)ch I
250
The basis of this study is that by using the enzyme tyrosinase it is possible to convert weakly adsorbable phenols into reactive intermediates (quinones) which can be strongly adsorbed onto sorbents of appropriate surface chemistries. From this and previous studies, there are three potential benefitsto the two-step tyrosinasereaction/ chitosan adsorption approach. The first potential benefit is that because adsorption of quinones onto chitosan is very strong,the two-stepapproach should be more effective for removing traces of phenols from wastewaters. The importance of adsorption strength is illustrated by the proportionality
310
370
430
490
550
WAVELENGTH (nm)
Figure 8. Enzymatic reactiodadsorptionof (a)phenol, (b)cresol, and (c) pyrocatechol. Curve A Phenolic added alone. Curve B Phenolic plus mushroom tyrosinase added. Curve C: Phenolic plus chitosan. Curve D: Phenolic plus mushroom tyrosinase plus chitoean. Within eachset of experiments,a buffered solution (50 mM phosphate) at pH 6.8 containing an initial phenolic concentration of 100 mg/L (1.06 mM phenol; 0.92 mM cresol; 0.91 mM pyrocatechol) was used. When added, mushroom tyrosinase was added to a final activity of 55 units/mL and chitosan was added at a level of 5% (w/v). The reaction mixtures in curves ByC, and D were open to the atmosphere to permit oxygenation and were shaken for 2 h at 27 "C before the UV absorbance was measured.
absorbance of the controls in which neither tyrosinase nor chitosan was added. When mushroom tyrosinase was added to the phenol solutions, there was a shift in the maximum UV absorbance from 270 nm (characteristic of phenols) to approximately 380-400 nm characteristic of the quinones (Duckworth and Coleman, 1970). Over the 2-h period of this experiment, the quinone peak decreases
(12)
From the above equation and the measured 17 kcal/mol enthalpy difference, chitosan adsorption can be estimated to yield a 1012-fold improvement in adsorption affinity over activated charcoal. Obviously, to achieve this 1012fold improvement in affinity, chitosan sorbents with high specific surface areas will need to be developed. Also, the "cost" of such strong adsorption is that desorption of the quinones is likely to be quite difficult. Thus, if chitosan regeneration and reuse are necessary,methods for breaking adsorption would need to be developed. Despite these concerns, the above estimate illustrates that the two-step tyrosinase reactionlchitosan adsorption approach may offer significant potential advantages for removing traces of phenols from wastewaters. The second potential benefit of the two-step tyrosinase reactionlchitosan adsorption approach is that tyrosinase can react with a range of phenolics and is less sensitive to changes in waste stream composition and strength (Atlow et al., 1984). Therefore, this enzymatic approach is likely to be more generic and operationally stable than microbiological treatment approaches. Microbiological treatment of phenolic wastes is plagued by instabilitiesresulting
Bbtechml. Rog., 1992, Vol. 8, No. 3
188
from the toxicity of these compounds to the microbial population (Jones et al., 1973;Yang and Humphrey, 1975). Although laboratory studies have demonstrated considerable benefit for the steady-state operation of microbial reactors, there are often problems in attaining and maintaining steady states in applications where the wastes ’ are generated discontinuously and where the waste strength and composition varies over time. Further, the ability of using tyrosinase on an “as needed basis” provides the flexibility of treating small-scale and/or urgent waste problems (e.g., accidental discharges) without the need for adapting a microbial culture to the waste. The final potential benefit of the two-step tyrosinase reaction/chitosan adsorption is that chitosan is obtained from chitin, a waste product of the shellfish industry. In many places, the chitin-rich wastes from this industry are landfilled at considerable cost. Thus, the tyrosinase reaction/chitosanadsorption approach provides a potential opportunity to convert this waste into a useful product. Concerning the practicality of the tyrosinase reaction/ chitosan adsorption approach, it is important to recognize that the mushroom tyrosinase enzyme used in this study is quite expensive. We chose to use this mushroom tyrosinase to examine the technical feasibility of the tyrosinase reaction/chitosan adsorption approach. For practical applications, a cheaper tyrosinase would need to be available. We are currently examining the extracellular bacterial tyrosinase from Streptomyces (Gardner and Cadman, 1990).
Acknowledgment This work was partially supported by National Science Foundation Grant CTS-8912141.
Literature Cited Aitken, M. D.; Venkatadri, R.; Irvine, R. L. Oxidation of phenolic pollutants by a lignin degrading enzyme from the white-rot fungus Phanerochaete chrysosporium. Water Res. 1989,23, 443. Atlow, S. C.; Bonadonna-Aparo,L.; Klibanov,A. M. Dephenolization of Industrial Wastewater Catalyzed by Polyphenol Oxidase. BiotechnoL Bioeng. 1984,26,599-603. Butt, V. S.;Lamb, C. J. Oxygenases and the Metabolism of Plant Products. In Biochemistry of Plants, Vol. 7,Secondary Plant Products; Conn, E. E., Ed.; Academic Press: New York, 1981; pp 627-667. Canovas, F. G.; Tudela, J.; Madrid, C. M.; Varon, R.; Carmona, F. G.; Lozano, J. A. Kinetic Study on the Suicide Inactivation of Tyrosinase Induced by Catechol. Biochem. Biophys. Acta 1987,912,417-423. Dawson, C. R.; Tarpley, W. B. On the Pathway of the CatecholTyrosinase Reaction. Ann. N.Y. Acad. Sci. 1963,100,937950. Donaldson, T. L.; Strandberg, G. W.; Hewitt, J. D.; Shields, G. R.; Worden, R. M. Biooxidation of Coal Gasification Wastewaters Using Fluidized-BedBioreactors. Enuiron. Prog. 1987, 6,205-211. Duckworth, H. W.; Coleman, J. E. Physicochemicaland Kinetic Properties of Mushroom Tyrosinase. J.Biol. Chem. 1970,425 (7),1613-1625. Gardner, A. R.; Cadman, T. W. Product Deactivation in Recombinant Streptomyces. BiotechnoL Bioeng. 1990,36,243251. Hall, L. D.; Yalpani, M. Formation of Branched-chain, Soluble Polysaccharide from Chitosan. J. Chem. SOC.,Chem. Commun. 1980, 1153-1154. Horowitz, N.H.; Shen, S.-C. Neurospora Tyrosinase. J. Biol. Chem. 1962,197,513-520.
Irving-Monshaw,S. New Zip in Activated Carbon. Chem. Eng. 1990,(Feb.), 43. Ito, N.; Phillips, S. E. V.; Stevens, C.; Ogel, Z. B.; McPherson, M. J.; Keen, J. N.; Yadav, K. D. S.; Knowles, P. F. Novel Thioether Bond Revealed by a 1.7 A Crystal Structure of Galactose Oxidase. Nature 1991,350,87-90. Jones, G. L.; Jansen, F.; Makay, A. J. Substrate Inhibition of the Growth of Bacterium NCIB 8250 by Phenol. J. Gen. Microbiol. 1973,74, 139-148. Keith, L. H.; Telliard, W. A. Priority Pollutants I: A Perspective View. Environ. Sci. Technol. 1979,13, 416-423. Klibanov, A. M.; Morris, E. D. Horseradish Peroxidase for the Removal of Carcinogenic Aromatic Amines from Water. Enzyme Microb. Technol. 1981,3,119-122. Klibanov, A. M.; Alberti, B. N.; Morris, E. D.; Felshin, L. M. Enzymatic Removal of Toxic Phenols and Anilines from Waste Waters. J. Appl. Biochem. 1980,2,414-421. Klibanov, A. M.; Tu, T.-M.; Scott, K. P. Peroxidase-Catalyzed Removal of Phenols from Coal-Conversion Waste Waters. Science 1983,221,259-261. Lanouette, K.H. Treatment of Phenolic Wastes. Chem. Eng. 1977,(Oct.), 99-106. Leatham, G. F.; King, V.; Stahmann, M. A. In Vitro Protein Polymerization by Quinones or Free Radicals Generated by Plant or Fungal Oxidative Enzymes. Phytopathology 1980, 70,1134-1140. Lerch, K.; Ettlinger, L. Purification and Characterization of a Tyrosinase from Streptomyces glaucecens. Eur. J. Biochem. 1972,31,427-437. Maity, N.; Payne, G. F.; Chipchosky,J. L. AdsorptiveSeparations Based on the Differencesin Solute-SorbentHydrogenBonding Strengths. Ind. Eng. Chem. Res. 1991,30,2456-2463. Mason, H. S.; Wright, C. I. The Chemistry of Melanin: V. Oxidation of Dihydroxyphenylalanineby Tyrosinase. J.Biol. Chem. 1949,180,235-247. Mayer,A. M.; Harel, E.; Ben-Shad, R. Assay of CatecholOxidase: A Critical Comparison of Methods. Phytochemistry 1966,5, 783-789. McIntyre, R. J.; Vaughan, P. F. T. Kinetic Studies on the Hydroxylation of p-Coumaric Acid to Caffeic Acid by SpinachBeet Phenolase. Biochem. J. 1975,149,447-461. Milun, A. J. Colorimetric Determination of Primary Amine in Fatty Amine Acetates and Fatty Amines. Anal. Chem. 1957, 29,1502-1504. Muzzarelli, R. A. A.; Tanfani, F.; Emanuelli, M.; Mariotti, S. N-(Carboxymethylidene)Chitosans and N-(Carboxymethyl)Chitosans: Novel Chelating Polyampholytes Obtained from Chitosan Glyoxylate. Carbohydr. Res. 1982,107,199-214. Parris, G. E. Covalent Binding of Aromatic Amines to Humates. 1. Reactions with Carbonyls and Quinones. Environ. Sci. Technol. 1980,14,1099-1106. Shuttleworth, K. L.; Bollag,J.-M. Soluble and ImmobilizedLaccase as Catalysts for the Transformation of Substituted Phenols. Enzyme Microb. Technol. 1986,8,171-177. Vaughan, P. F. T.; Butt, V. S. The Action o-Dihydric Phenols in the Hydroxylation of p-Coumaric Acid by a Phenolase from Leaves of Spinach Beet (Beta vulgaris L.).Biochem. J. 1970, 119,89-94. Worden, R. M.; Donaldson, T. L. Dynamics of a Biological Fixed Film for Phenol Degradation in a Fluidized-Bed Bioreactor. Biotechnol. Bioeng. 1987,30, 398-412. Wright, C. I.; Mason, H. S. The Oxidation of Catechol by Tyrosinase. J. Biol. Chem. 1946,165,45-53. Yang, R. D.; Humphrey, A. E. Dynamic and Steady State Studies of Phenol Biodegradation in Pure and Mixed Cultures. Biotechnol. Bioeng. 1976,17,1211-1235. Accepted February 4, 1992. Registry No. Tyrosinase, 9002-10-2; chitosan, 9012-76-4; pcresol, 106-44-5; catechol, 120-80-9; phenol, 108-95-2; p-quinone, 106-51-4;pyrocatechol, 120-80-9.
