Fungal protein from corn waste effluents : a model study

The purpose of this investigation was to study the microbiological aspects of the production of microbial protein ('single cell protein'; SCP) from corn waste effluents with simultaneous reduction of the COD of these effluents.For practical reasons the corn waste water itself was not used in the experiments but a model was chosen, consisting of tap water to which corn steep liquor (CSL) and carbon sources (as a rule glucose) were added.A fungus was chosen as a model organism because of the composition of the waste stream (high content of starch) and the low COD level of such a stream (ca. 6000 mg/l). The costs of separation of the biomass increase at decreasing concentrations of micro-organisms. Fungi have an advantage over bacteria and yeasts in that fungal mycelium can be separated by relatively simple and cheap filtration techniques.Chapter 1 gives a survey of literature on SCP and on the production of such proteins from wastes. In particular, attention was given to the use of fungi for this purpose. The production of algae and the use of hydrocarbons as substrate were mentioned only incidentally.The fungus imperfectus <em>Trichoderma viride</em> was chosen as the model organism in the investigations. This choice was based on the following data from literature: a. relatively high specific growth rate; b. high (crude) protein content; c. low optimum pH; d. ability to utilize many (macromolecular) compounds as carbon source; e. no toxic properties of the biomass observed in feed trials.In chapter 2 a summary is presented of the materials and methods used in the investigations.Chapter 3 deals with batch experiments in which factors were examined affecting growth rate, yield constant, crude protein content, exhaustion of the medium (COD reduction and nitrogen uptake) etc. As a rule the media contained a COD level of ca. 2500 mg/l in order to ensure a sufficient oxygen supply.The highest COD reduction rate in CSL glucose media was observed at an initial C/N ratio of 12.6 and an initial pH between 3.5 and 5.5 (Fig. 3.1). The highest COD reduction observed was about 95 % (Fig. 3.4). The yield constant <em>Y</em> (mg biomass produced per mg COD reduced), decreased from about 0. 5 after initial growth to about 0.4 at exhaustion of the medium (Fig. 3.5). The maximum specific growth rate (μ _{<font size="-1">max</font>} ) observed was 0.28 hr <sup><font size="-1">-1</font></SUP>.The crude protein content of the fungus varied from about 20 to 60 % (Fig. 3.8.). This content decreased with increasing initial C/N ratio of the medium and with increasing incubation time (Figs. 3.8 and 3.9). More than 80% of the nitrogen present in CSL was taken up from the medium; this figure appeared to be independent of the pH and also of the C/N ratio of the medium if C/N>9 (Fig. 3. 10). At lower C/N ratios a higher percentage of the nitrogen remained in the medium. The nitrogen content of the culture filtrate increased after exhaustion of the medium, but more rapidly at low C/N ratios of the fresh medium than at high C/N ratios.Decrease of the C/N ratio in the waste stream to an optimum value can better be achieved by the addition of urea or (and) ammonium salts than by addition of CSL, in order to avoid increase of the COD remaining in the stream (Fig. 3.12 and Table 3.2).The maximum observed specific respiration rate of the fungus was ca. 7 mmol O _{<font size="-1">2</font>} /g dry biomass.hr (Fig. 3.11).The morphology of the fungus depended on a large number of factors such as composition of the growth medium, nature and size of the inoculum, growth conditions and age of the culture.In chapter 4 the utilization is described of ethanol, lactic acid and acetic acid as carbon sources for <em>T. viride</em> . These compounds are present in the waste stream besides glucose and starch.The highest yield constant ( <em>Y</em> ) was found with ethanol (higher than with glucose) and the lowest with lactic acid (Table 4.1). The <em>Y</em> values found with <em>T. viride</em> in the present study were about the same as those reported for <em>Candida utilis</em> by HERNANDEZ and JOHNSON in 1967 (Table 4.3). At an initial pH of 4.5, acetic acid appeared to be taken up simultaneously with glucose but before ethanol, while lactic acid was utilized after ethanol (Table 4.2). Lactic acid was consumed very slowly; the remaining COD in culture filtrates from CSL media must probably be attributed to a large extent to lactic acid originating from CSL. Acetic acid caused a strong growth rate inhibiting effect even at very low concentrations (Fig. 4.3). In the presence of acetic acid in concentrations above 7.5 mM and at pH values below 4.5, growth of the fungus in the pellet form was observed. Concentrations of ethanol above ca. 80 mM (0.5 %; v/v) had also a toxic effect (Fig. 4. 1). A growth rate inhibiting effect of lactic acid could hardly be found, at least at concentrations below 25 mM (Fig. 4.2).Considering the concentrations of ethanol, acetic acid and lactic acid in corn waste effluents, the growth rate inhibiting effects of these carbon sources can be expected to be nihil, in practice.Chapter 5 describes the continuous production of mycelium from CSL media, both in a single-stage system and in a single-stream dual-stage system. Culture vessels were used with a capacity of 3 litres.Although it was impossible to determine the maximum specific growth rate (μ _{<font size="-1">max</font>} ) exactly because of the effect of wall growth, the value of μ _{<font size="-1">max</font>} was probably not markedly different from that found with batch cultures. The biomass concentrations were calculated from the nitrogen balances because of the badly reproducible sampling of the fermentor contents. At an equal COD level of the fresh medium, the maximum yield of biomass in continuous cultures was lower than that obtained in batch cultures; also the maximum COD reduction observed was somewhat lower than in batch cultures (Fig. 5.1 <sup><font size="-1">a-c</font></SUP>). At dilution rates lower than 0.05 hr <sup><font size="-1">-1</font></SUP>autolysis of mycelium was observed, accompanied by an increased COD of the culture filtrate. At C/N ratios between 10 and 15 the yield constant ( <em>Y</em> ) varied from about 0.51 at <em>D</em> = 0.24 hr <sup><font size="-1">-1</font></SUP>to about 0.36 at <em>D</em> = 0.05 hr <sup><font size="-1">-1</font></SUP>; the crude protein content varied from 40 to 60 %. The maximum nitrogen recovery from media with a ratio C/N>10 was about the same as observed in batch culture. The maintenance coefficient calculated with the equation of PIRT (1965) was 25-30 mg glucose/g biomass.hr (Fig. 5.2). Higher COD levels of the fresh medium did not result in a higher percentage of COD reduction at a dilution rate of 0.05 hr <sup><font size="-1">-1</font></SUP>.Single-stream dual-stage cultures (Fig. 5.3) were not markedly advantageous to a single-stage system (Table 5.1). In a numerical approach it was shown theoretically that indeed such a dual-stage system is not necessarily advantageous to a single-stage system, and if so, that the second stage must not necessarily have a larger volume than the first one (Tables 5.3; 5.4; 5.5).The results obtained experimentally in this study may have been highly affected by the small process scale used, especially because of the considerable wall growth.Chapter 6 deals with the amino acid composition of the fungal biomass and with the nature of the nitrogen compounds in the culture filtrate.About 80% of the nitrogen present in the biomass was found to be amino acid nitrogen (Table 6.2). In each of the biomass analyses, the proteins were deficient in the sulphur-containing amino acids (methionine and cysteine), independently whether CSL media or mineral media had been used.Approximately 90% of the nitrogen present in the culture filtrates of CSL media was amino acid nitrogen (Table 6.2) which was for more than 95 % present as peptides, and for less than 5 % as free amino acids. These peptides were partly present already in the CSL media and were partly excreted by the fungus. Various amino acids, in particular proline, cysteine and histidine, are present in CSL-containing media in much larger amounts than those utilized by the fungus (cf. Tables 2.2 and 6.2). Amino acid nitrogen, largely present as peptides, was also excreted when the fungus was grown on a mineral medium with glucose and urea (Table 6.2).The last chapter describes the regulation of the synthesis of starch-hydrolysing enzymes by <em>T. viride</em> . The specific growth rate on starch (one of the predominant carbon compounds in corn waste effluents) was only slightly lower than that on glucose; maltose was consumed very slowly, accompanied by formation of a slime.The amylolytic enzyme system appeared to be completely extracellular; it consists mainly of one or more enzymes of the glucoamylase type. Because no maltose was detected by thin layer chromatography, α-amylase seems to play only a minor role. The high ratio between saccharifying and dextrinizing activity of culture filtrates (SA/DA about 1.6), compared to that of a relatively pure α-amylase preparation obtained from <em>Aspergillus oryzae</em> (SA/DA = 0.15), <em></em> points also to the presence of mainly glucoamylases (see also Fig. 7.2).The optimum pH and the heat stability of the amylolytic enzymes of <em>T. viride</em> were about similar to those of amylolytic enzymes of other fungi (Table 7.2; Figs. 7.3; 7.4; 7.5).The synthesis of amylolytic enzymes required the presence of starch or dextrins as inducer (Table 7.7; Fig. 7.6). Several readily utilizable carbon sources such as glucose, glutamate and other organic nitrogen compounds were shown to exert catabolite repression (Figs. 7.6; 7.7; 7.8).Enzyme synthesis occurred both in the exponential and in the stationary growth phase (Fig. 7.1). Growth on starch resulted in an initially high dextrinizing activity; subsequently, the saccharifying activity increased and became predominant in the course of exponential growth (Fig. 7.6.1). In dextrin DE-30 cultures the saccharifying activity was predominant from the very beginning (Fig. 7.6 IV).It may be concluded that the amylolytic enzyme system of <em>T.</em><em>viride</em> consists of at least two different types of enzymes (the one more saccharifying, the other more dextrinizing), the synthesis of each being regulated specifically.Since maltose, the main final low-molecular product of α-amylase activity, is hardly utilized by the strain of <em>T.</em><em>viride</em> used in this study, the production of α-amylase has to be carefully regulated by the fungus.