Optimization of dextran syntesis and acidic hydrolisis by surface response analysis

Optimization of dextran syntesis and acidic hydrolisis by surface response analysis

D.P. GUIMARÃES, F.A.A. COSTA, M.I. RODRIGUES, and F. MAUGERI

Food Engineering Department, (FEA), UNICAMP, 13083-970, Campinas - SP, Brazil, [e-mail]

(Received: January 19, 1999; Accepted: March 26, 1999)

Abstract - The influence of some variables in the in vitro synthesis of dextran by dextransucrase from Leusconostoc mesenteroides NRRL B512F, as well as in the acidic hydrolysis of the dextran produced, were studied in order to maximize the production of clinical dextran (dextran 70 and dextran 40). The experiments were conducted using a factorial design and surface response analysis.

Keywords: Optimization, dextran, acidic hydrolysis.

INTRODUCTION

Dextran is a microbial biopolimer whose molecular structure is composed exclusively of monomeric 2-D-glucopyranosil units, linked mainly by (a - 1,6) glucosidic bonds. Applications of dextran depend on its molecular weight. Two dextran products are available in most countries for clinical purposes: dextran 70 with a MW of ca. 70000 daltons (75000±25000) and dextran 40 with a MW of ca. 40000 daltons (40000±10000). The in vitro synthesis of dextran is accomplished by dextransucrase (EC 2.4.1.5, a (1,6)-D-glucan 1,6-a-D-glucosil transferase) from Leusconostoc mesenteroides NRRL B512F. The in vitro process produces dextrans with MW of ca. 4000000 daltons as the main product, but also olygosaccharides and fructose, having sucrose as the substrate. Santos (1996) and Pereira et al. (1998) showed that molecular weight and yield depend on process variables such as temperature and sucrose and acceptor concentrations. Therefore, the main goal of this work was to optimize the clinical dextran yield, starting with the maximization of dextran synthesis, minimization of oligodextran production, and optimization of the hydrolytic process, since dextran hydrolysis is markedly influenced by pH and temperature. The experimental design chosen was a factorial design for both processes, and optimization was accomplished by surface response analysis.

MATERIAL AND METHODS

Dextransucrase Production

Leuconostoc mesenteroides NRRL B-512F was grown in medium containing 50 g/l sucrose, 20 g/l yeast extract, 20 g/l MgSO₄, 0.01 g/l NaCl, 0.01 g/l FeSO₄, 0.01 g/l Mn SO₄, 0.02 g/l CaCl₂ and 20 g/l K₂HPO₄. The enzyme was obtained in a fed-batch culture. Temperature was maintained at 27ºC, and aeration at 0.5 l/min. After removing cells by centrifugation (10000 rpm, 4ºC during 15 min), the supernatant was concentrated by ultrafiltration. Fifty percent Polyethyleneglycol 1500 was added to the concentrated enzyme solution and then centrifuged at 5000 rpm, 5ºC, for 10 minutes (Paul et al., 1986). The precipitate was resuspended in a buffer solution containing 20 mM sodium acetate (pH = 5.2) and stored at -15ºC.

In Vitro Synthesis of Dextran

Experiments were carried out within a range of temperatures and substrate concentrations. Temperatures ranged from 20 to 36°C and sucrose concentration from 30 to 100g/l. Polimerizations were performed in a glass reactor containing sucrose solution and 40 UDS/ml dextransucrase. Samples were taken every 30 minutes in order to monitor the reaction by measuring the fructose released in the medium, using the DNS method (Miller, 1959).

Molecular Weight Distribution

Molecular weight distribution was measured by gel permeation chromatography using a Varian liquid chromatograph with three GPC columns in series (TSK 5000 PWXL, TSK 4000 PWXL and TSK 2500 PWXL) and the Waters' Millennium software. Yields were determined by Equations 1 (global yield) and 2 (specific yield).

Y_global = (% total dextran area)/0.474 (1)

Y_specific = (% of area of D70 or D40) * _Yglobal (2)

where 0.474 is the theoretical yield of dextran from sucrose, D70 is Dextran 70 and D40 is dextran 40.

Acidic Hydrolysis

Dextran hydrolysis was carried out according to a factorial design and monitored for 7 hours. Temperatures changed from 73 to 87°C and pH from 1.0 to 1.8, and pH values were regulated by adding sulfuric acid to the dextran solution. Experiments were performed in a stirred flask, plunged in still bath, provided with a condenser and thermometer. Samples were taken every hour and molecular weight measured by GPC chromatography.

The yield of each dextran type (D40 and D70) was measured by the percentage of respective area in the chromatograph divided by the percentage of initial dextran in the solution (Equations 3 and 4).

Y_D40 = (% of area from MW 30000 to MW 50000) x x 100 / (% of Dextran) (3)

Y_D70 = (% of area from MW 50000 to MW 100000) x x 100 / (% of Dextran) (4)

Sugar And Enzymatic Activity Analysis

Sugars were measured by the dinitrosalycilic method (DNS). Dextransucrase activity was determined by measuring the initial reaction rate of fructose release from sucrose. The standard unit was defined as the amount of enzyme that catalyzes the formation of 1 mmol of D-fructose per min at 30ºC, in the presence of 100 g of sucrose/L.

RESULTS AND DISCUSSION

Dextran Synthesis

Experimental yields for dextran synthesis reactions, using different temperature and substrate conditions, determined by the factorial design, are shown in Table ¹.

Figures ¹ and ² show the surface responses for high and low molecular weight dextrans, respectively. Tables ² and 3 show the ANOVA (Analysis of Variance) for each model, obtained from the factorial design. These models generated the surfaces in Figure ¹ and ².

Figure 1: High-molecular-weight dextran yields as a function of temperature and sucrose concentration in enzymatic synthesis

Figure 2: Low-molecular-weight dextran yields as a function of temperature and sucrose concentration in enzymatic synthesis

F 0.95; 5;5 = 5.05

F 0.99; 5;5 = 10.97

F 0.95; 5;5 = 5.05

F 0.90; 5;5 = 3.45

The maximum experimental yield for dextran production was 32% at 23°C and 60g/l of sucrose. Figure ¹ shows that the yield for high MW dextran decreases as temperature increases. Moreover, in Figure ², it can be seen that the yield for low MW dextran (MW<10000 daltons) increases as temperature increases. On the other hand, sucrose has little effect on low and high MW dextran yields, except at high temperatures where an increase in the yield of high MW dextrans can be seen.

The surfaces in Figures ¹ and ² are generated by mathematical models obtained with nonlinear regression of experimental data. According to Barros et al. (1995), the mathematical model is predictable as long as the F test value is at least four times greater than the listed value. According to Table ², the F value (31.59) is six times greater than the listed value (5.05) at 95% confidence. In this case, the model represented by Equation 4 can be used to predict yield values for high MW dextrans at 95% confidence. Therefore, in Figure ¹, it can be seen that the yield can be as high as 45% for a low temperature (23°C) and a high sucrose concentration (107g/l). These conditions are suitable for experimental work since, beside high yields, they can also offer high productivities.

Optimization Of Dextran Hydrolysis

Samples from the hydrolytic reactions were analyzed for compositions of Dextran 40 and Dextran 70. Experimental yields for Dextran 40 and Dextran 70, according to the factorial design, are shown in Tables ⁴ and ⁵.

It can be seen in Table ⁶ that pH and also interaction pH×T are the most significant variables for the production of Dextran 40. Table ⁷ shows that only interaction pH×T, mainly at the end of the reaction, is important for Dextran 70.

*Significant effects at 95% confidence

*Significant effects at 95% confidence

Figures ³ and ⁴ show the surface responses for Dextran 40 and Dextran 70 after 6 hours of reaction time, generated by the models (equations 7 and 8, respectively). Tables ⁸ and ⁹ show the ANOVA for both quadratic models. Since F values in these tables are higher than those listed, both models can be employed in predicting dextran yields.

Figure 3: Dextran 40 in the hydrolytic reaction as a function of pH and temperature

F 0.95; 5;5 = 5.05

F 0.90; 5;5 = 3.45

F 0.95; 5;5 = 5.05

F 0.99; 5;5 = 10.97

It can be observed in Figure ³ and Figure ⁴ that Dextran 40 and Dextran 70 yields have nearly the same behavior, as seen by changes in pH and temperature, after 6 hours of reaction. Temperature effects are important mainly at higher pH values, and the lower the pH, the higher the dextran yield.

Figure 4: Dextran 70 yields in the hydrolytic reaction as a function of pH and temperature

Based on the surface responses and all reaction times, it was estimated that the best reaction conditions were near 75ºC and pH 1.0. In order to verify the yields at any time of reaction, under different hydrolytic conditions, the fitted models were adjusted to pH 1.0 and temperatures from 72ºC to 80ºC. The predicted yields were then calculated. The results are shown in Figures ⁵ to ¹⁰. It can be seen that there is an oscillatory behavior with maximun yields after 6 hours of reaction for all conditions, except for the temperature of 80ºC. It can also be observed that the highest yield is achieved at T=72ºC and pH = 1.0 after 6 hours of reaction. Under these conditions, Y_D40 is 44% and Y_D70 is 56%, which means that all the high-molecular-weight dextran was transformed into clinical dextran.

Figure 5: Predicted dextran yields as a function of reaction time for T = 72ºC and pH = 1.0

Figure 6: Predicted dextran yields as a function of reaction time for T = 74ºC and pH = 1.0

Figure 7: Predicted dextran yields as a function of reaction time for T = 75ºC and pH = 1.0

Figure 8: Predicted dextran yields as a function of reaction time for T = 76ºC and pH = 1.0

Figure 9: Predicted dextran yields as a function of reaction time for T = 78ºC and pH = 1.0

Figure 10: Predicted dextran yields as a function of reaction time for T = 80ºC and pH = 1.0

Tables 6 and 7 show the estimates of effects for variables T, pH and the interaction between them, pH×T, as a function of reaction time for Dextran 40 (Table ⁶) and Dextran 70 (Table ⁷). The symbol * represents values with statistical significance.

CONCLUSIONS

In the dextran synthesis experiments the maximum experimental yield was 32% under reaction conditions of 23ºC and a sucrose concentration of 60g/l. The factorial design allows the prediction of optimal conditions, different from the experimental ones, which are a temperature of 23ºC and a sucrose concentration of 107 g/l.

It was also observed that an increase in reaction temperature results in a decrease in the molecular weight of the dextran formed, resulting in an increase in olygossacharides concentration.

In the hydrolytic step of this work, it was verified that pH is a very important parameter to control in dextran hydrolysis. Temperature has a less meaningful effect; however it is important to optimize hydrolytic conditions. The best predicted hydrolytic conditions were shown to be near 72ºC and pH = 1.0, which resulted in yields of approximately 100% conversion of crude dextran to clinical dextran.

REFERENCES

Barros Neto, B.; Scarminio, I. S.; Bruns, R. E. (1995) Planejamento e Otimização de Experimentos, Campinas, Editora da UNICAMP.

Miller, G. L. Use of Dinitrosalicylic Acid Reagent for Determination of Reducing Sugar; Anal. Chem.; 31; p. 426-428; 1959.

Santos, V. M. (1996) Estudo das condições de hidrólise ácida para obtenção de dextrana clínica, Campinas, FEA - UNICAMP.

Paul, F.; Oriol, E.; Auriol, D.; Monsan, P. (1986). Carbohyd. Res. 149: 433 - 441.

Pereira, A. M.; Costa, F. A. A.; Rodrigues, M. I.; Maugeri, F. (1998) In vitro synthesis of oligosaccharides by acceptor reaction of dextransucrase from Leuconostoc mesenteroides, Biotechnology Letters, 20 (4): 397 - 401.

• Barros Neto, B.; Scarminio, I. S.; Bruns, R. E. (1995) Planejamento e Otimizaçăo de Experimentos, Campinas, Editora da UNICAMP.
• Miller, G. L. Use of Dinitrosalicylic Acid Reagent for Determination of Reducing Sugar; Anal. Chem.; 31; p. 426-428; 1959.
• Santos, V. M. (1996) Estudo das condiçőes de hidrólise ácida para obtençăo de dextrana clínica, Campinas, FEA - UNICAMP.
• Paul, F.; Oriol, E.; Auriol, D.; Monsan, P. (1986). Carbohyd. Res 149: 433 - 441.

Table 1 - Central composite design arrangement and responses for high and low-molecular-weight dextran yields

Y = 13.13 – 6.07 T + 5.7 T² - 3.04 S + 2.67 S² - 3.6 S T (5)

Y = 35.19 + 8.69 T – 5.24 T ² - 6.25 S + 0.74 S² - 3.72 S T (6)

Table 2: Analysis of variance in the second-order Y model for high-molecular-weight dextrans

Table 3: Analysis of variance in the second-order Y model for low-molecular-weight dextrans

Table 4 : Experimental yields for Dextran 40 as a function of reaction time under hydrolytic conditions determined by the factorial design

Table 5 : Experimental yields for Dextran 70 as a function of reaction time under hydrolytic conditions determined by the factorial design

Table 6 : Estimates of effects for temperature(T), pH, and pH×T for Dextran 40 as a function of reaction time

Table 7 : Estimates of effects for temperature(T), pH, and pHxT for Dextran 70 as a function of reaction time

Y = 10.072 – 7.689 pH + 3.186 pH² + 7.722 pH T (7)

Table 8 - Analysis of variance in the second-order Y model for Dextran 40

Table 9 - Analysis of variance in the second-order Y model for Dextran 70

Y = 18.661 - 2.898 T - 8.560 pH + 9.385 pH T (8)

Publication Dates

• Publication in this collection
  15 Sept 1999
• Date of issue
  June 1999

History

• Accepted
  26 Mar 1999
• Received
  19 Jan 1999