Carlos M. Regalado

Non-linear optimization of biotechnological processes by stochastic algorithms: Application to the maximization of the production rate of ethanol, glycerol and carbohydrates by Saccharomyces cerevisiae

A non-linear optimization, based on an stochastic multi-start search algorithm, has been applied to the maximization of the production rates of ethanol, glycerol and carbohydrates by Saccharomyces cerevisiae. This optimization is applied to two alternative (non-linear) model representations of the same system, namely the Michaelis-Menten and the generalized mass action forms. We find a complete agreement between the results obtained using both representations. This is, maximization of the ethanol production rate requires modulation of up to six enzymes, while modification of only one enzyme is sufficient to obtain a significant improvement in the production rate of glycerol and carbohydrates. When the results are compared with those previously obtained using an indirect linear optimization method (Torres, N.V., Voit, E.O., González-Alcón, C., Rodríguez, F. 1997. An integrated optimization method for biochemical systems. Description of method and application to ethanol, glycerol and carbohydrate production in S. cerevisiae. Biotechnol. Bioeng. 55(5), 758-772.), we find close agreement between both optimization techniques. Qualitatively, both optimization approaches render the same profile of enzymes to be modulated, while quantitatively, discrepancies arise when the objective function is the maximization of the ethanol production rate. Reasons for such discrepancies and an evaluation of the advantages of each method (linear vs non-linear) are given.

USEFUL SOIL-WATER REPELLENCY INDICES : LINEAR CORRELATIONS

Water repellency (WR) has been classically characterized at fixed (usually oven-dry) soil water content (&thgr;g) in terms of the soil water contact angle (CA), &agr;. However, &agr; has been previously reported to depend upon &thgr;g in a nonlinear fashion, such that WR increases from a wettable state close to saturation (&thgr;g-min) up to a maximum, &agr;max, decreasing afterward either monotonically or rising again to a second local or absolute &agr; maximum nearby the dried soil state. Hence, a CA versus water content (&agr;-&thgr;g) curve may be described in terms of different WR parameters, such as &thgr;g-min, &thgr;g-max, &agr;max, or the integrated area below the &agr;-&thgr;g curve, S. Based on previous &agr;-&thgr;g measurements carried out with the molarity of an ethanol droplet (MED) test, both in mineral and volcanic soils from different world regions, including cultivated and natural forest soils, and textures ranging from clay-loam to sandy, we confirm here the usefulness of the integrated area below the &agr;-&thgr;g curve (S) as a WR describing index for a large variety of &agr;-&thgr;g curve shapes. We found a simple relationship between S and the soil water content at which WR is triggered, &thgr;g-min, such that S = 16.903 &thgr;g-min (R2 = 0.946), which provides an easy method for the rapid characterization of the overall WR degree of soils. S was also linearly correlated with the soil organic matter (SOM) content (R2 = 0.817) for 1 g (100 g)−1 < SOM < 88 g (100 g)−1, such that the best estimate of S was that obtained by combining linearly both &thgr;g-min and the SOM content (R2 = 0.990). Linear correlations were also found between &thgr;g-max, that is, the soil water content at which &agr; is maximum, and S (R2 = 0.834) or the SOM content (R2 = 0.705), and consequently between &thgr;g-max and &thgr;g-min (R2 = 0.830). In addition, both &thgr;g-min and &thgr;g-max were found to depend linearly upon the soil water content at −33 kPa and −1500 kPa, respectively. Finally, a mean soil WR may be defined as the ratio S/&thgr;g-min. We found that the maximum CA, &agr;max, and the mean soil WR S/&thgr;g-min were positively correlated (R2 = 0.780), such that a particular soil with high (low) values of maximum CA is expected to exhibit a high (low) WR degree on average across the whole water regimen from −33 kPa down to oven-dry moisture. Such an estimate of the mean WR index S/&thgr;g-min was further improved if both &agr;max and the SOM content were available (R2 = 0.825).ABBREVIATIONS CA: contact angle; IRDI: Integrative Repellency Dynamic Index; MED: molarity of an ethanol droplet; SOM: soil organic matter; WDPT: water drop penetration time; WR: water repellency.

Quality Assessment of a Metabolic Model and Systems Analysis of Citric Acid Production by Aspergillus Niger

Citric acid is a chemical of foremost importance in the food, cosmetic and chemical industry. Its annual production amounts up to 350,000 tons, with the carbohydrate fermentation by means of the filamentous fungus Aspergillus niger (ascomycetes group) being one of the leading processes for its production worldwide. A great deal of work has been done on the biochemistry of this process1 and, consequently, information is available on most of its biochemical aspects. However, in spite of the amount of information on the system and its economic importance, no attempt has been made to explain its regulation on a quantitative basis.

Hydrological fluxes of a crest laurel subtropical forest in the Garajonay National Park

A study was carried out to quantify the relative importance of the hydrological fluxes involved in crest areas of a laurel subtropical forest ecosystem in the Garajonay National Park (Canary Islands). From a two-year time series of micrometeorological measurements, the forest water inputs and potential evapotranspiration were quantified. Different models were adapted for this type of vegetation to describe physical processes such as potential evapotranspiration, canopy interception losses and the impaction of fog water droplets onto cylindrical elements (needle-like leaves). Conventional precipitation shows seasonality, while impaction models predict that fog precipitation acts as an additional water supply distributed along the year, representing around 20-45% of total yearly rainfall. Due to the prevailing climatic conditions and the stomatal control exhibited by this vegetation, the forest water demand is constrained by reduced yearly evapotranspiration values, which are satisfied mainly by rainfall and additionally by fog interception.