333
Biotechnol. Prog. IQQO, 6,333-340
Defining an Optimal Carbon Source/Methionine Feed Strategy for Growth and Cephalosporin C Formation by Cephalosporium acremonium Steven M. Vicik,' Anthony J. Fedor, and Randall W. Swartz Biotechnology Engineering Center, Tufts University, Medford, Massachusetts 02155
The effect of the method of methionine addition, growth-limiting carbon source (glucose vs sucrose), and culture growth rate on cephalosporin C production was investigated in a Cephalosporium acremonium defined medium fed batch fermentation. Batch addition of methionine, a t a concentration of 3 g/L, prior to the start of a fed sucrose fermentation was found to interfere with the ability of the culture to utilize this sugar, thus limiting growth and decreasing cephalosporin C production. Batch methionine addition had no effect on glucose-limited cultures. Concurrent exponential feeding of methionine with sucrose improved both culture growth and productivity. Under the control of identical carbon source limiting feed profiles, sucrose was observed to support greater cephalosporin C production than glucose. Optimal cephalosporin C production in a C. acremonium defined medium fed batch fermentation was obtained through controlling culture growth during the rapid growth phase a t a relatively low level with respect to p- ( p 0.036 h-l) until achieving a desired cell mass with a concurrent sucrose and methionine feed, followed by maintaining relatively vigorous growth ( p 0.01 h-l) with sucrose for the duration of the fermentation.
-
Introduction The development of a fermentation strategy that improves cephalosporin C production from Cephalosporium acremonium must be based on the interactions of many regulatory phenomena. Over the last 25 years, numerous studies have described productivity increases associated with the influence of various medium components and related these to the activities of enzymes in the cephalosporinC synthetic pathway as well as on overall cephalosporin C production. Carbon source repression/ inhibition and methionine induction have been observed to be two of the most important factors affecting antibiotic synthesis. In addition, it has been observed that the control of various physical parameters, such as specific growth rate (Matsumura et al., 1978) or specific sugar uptake rate (Scheidegger et al., 1988),exerts a strong influence on cephalosporin C production. Carbon source catabolite regulation of antibiotic biosynthesis has been reported in many antibiotic fermentations including that of C. acremonium for the production of cephalosporin C [for a review, please refer to Demain et al. (1983)l. In the cephalosporin C fermentation, glucose has been reported to support rapid growth but interfere with antibiotic biosynthesis (Matsumura et al., 1978; Kennel and Demain, 1978). Glucose, and other rapidly utilized carbon sources such as glycerol, reduce cephalosporin C synthesis rates in C. acremonium C-10 by repressing the ring expansion enzyme desacetoxycephalosporin C synthetase (Behmer and Demain, 1983; Zanca and Martin, 1983) and through the inhibition of 6-(L-cr-aminoadipyl)-L-cysteinyl-Dvaline (ACV) synthetase (Zhang et al., 1989), the first enzyme in the cephalosporin C biosynthetic pathway. Production of cephalosporin C by many strains is known to be stimulated by the addition of the amino acid me-
-
thionine during the growth phase of C. acremonium fermentations (Matsumura et al., 1978). Although methionine can supply sulfur to the cephalosporin C biosynthetic pathway, this is not the sole mechanism of its stimulatory effect, since norleucine, a non-sulfur analogue, also has a similar effect (Drew and Demain, 1973). Methionine has also been reported to stimulate the morphological differentiation of C. acremonium cultures (Matsumura et al., 1980) and the induction of isopenicillin N synthetase (cyclase), desacetoxycephalosporin C synthetase (expandase) (Sawada et al., 1980), and 6 - ( ~ - a aminoadipyl)-L-cystehyl-D-valine (ACV) synthetase (Zhang et al., 1987), three enzymes in the synthetic pathway. Proper control of physical parameters such as specific growth rate and/or specific sugar uptake rate is an important consideration in maximizing cephalosporin C yields. Smith (1985) has observed a linear relationship between specific growth rate and specific cephalosporin C production rate in chemostats where soya bean oil was growth limiting using a high-yielding C. acremonium mutant. However, several other studies have shown an inverse correlation between specific growth rate and the specific rate of antibiotic production. Kennel and Demain (1978) observed a roughly inverse correlation between C. acremonium CW19 growth rate and antibiotic production, with sucrose-grown cultures having the greatest specific rate of antibiotic production but the lowest growth rate of the five sugars tested. Matsumura et al. (1978) determined that a relatively slower growth rate during the rapid growth phase of a C. acremonium M8650 culture maximized the specific antibiotic production rate. Recently, Scheidegger et a. (1988) demonstrated that maximal cephalosporin C production rates could be obtained if the specific glucose uptake rate, and subsequently the specific growth rate, were controlled a t a
8756-7938/90/3006-0333$02.50/0 0 1990 American Chemical Society and American Institute of Chemical Engineers
334
relatively low level. Fed batch control of C . acremonium growth through glucose addition, used by Matsumura et al. and Scheideggeret al. in the studies described above and typical or industrial processes (Scheidegger et al., 1988), is at least in part effective by reducing carbon catabolite repression. Maximizing production in an antibiotic fermentation involves not only maintaining high specific rates of product formation but also maintaining high cell mass. While methionine induces cephalosporin C synthesis in batch defined medium cultures in which glucose and sucrose served as the major carbon sources, it also inhibits the growth of such cultures (Sawada et al., 1980; Zhang et al., 1987). Cephalosporin C production might be improved if methioninerelated growth inhibition was alleviated while maintaining its stimulatory effect on cephalosporin C synthesis. In addition, most studies t h a t have employed carbon limitation to control growth have used rapidly utilized carbon sources such as glucose. While Kennel and Demain (1978) observed that sucrose supported higher specific rates of P-lactam production in batch cultures, no studies have reported its use as a growth-limiting nutrient. This work was undertaken to investigate the use of sucrose as a growth-limiting nutrient, the effect of methionine on sucrose-fed cultures, and the effect of culture growth rate during the rapid growth phase and for the duration of a C. acremonium fermentation on cephalosporin C production. The results of this investigation will define an improved carbon source/methionine feed strategy useful in industrial processes.
Experimental Met hods Media and Culture Conditions. C. acremonium C-10 was maintained on agar slants (Shen et al., 1986). Cells from a slant were suspended in 5 mL of sterile water and added to 50 mL of complex seed medium. This firststage seed medium contained 6.0% glucose, 0.4% (NH4)2S04,1.2% CaC03, 0.003% ZnS04.7H20, 0.5% soybean meal, 1.5% cottonseed flour (Pharmamedia), and 3 70 corn steep liquor. Glucose was autoclaved separately from the other medium components, whose pH was adjusted to 6.4 with 50% NaOH prior to sterilization. Cultures were grown for 2-3 days in a rotary incubator/ shaker at 25 "C and 200 rpm. The composition of the second- and third-stage seed media are identical with the defined fermentation medium described by Shen et al. (1986) with the exception that 3.0% (NH4)2S04and 1% Mg3(PO4)2-8H20were substituted for 1.2 % asparagine. The second-stage seed contained 50 mL of chemically defined medium and 10% inoculum and was grown for 2-3 days in a rotary incubator/ shaker a t 25 "C and 200 rpm. The contents of a secondstage seed were added to 450 mL of the aforementioned chemically defined medium contained in a 2.8-L Fernbach flask. Third-stage seeds were grown for 3 days in a rotary incubator/shaker a t 25 "C and 200 rpm. Two 500-mL third-stage seed cultures were used to inoculate a New Brunswick Microferm 14-L fermentor (-8-L working volume). The batch fermentation medium contained 0.5% (NH4)~S04,1.0% Mg3(P04)~~H20,0.007% CaCl~2HzO,O.O0075 % CuSOc5Hz0,0.003 % ZnSOc7Hz0, 0.003 7; MnS04sH20, 0.035% MgS04.7Hz0, 0.165% 0.30% NazS04-10H20, 0.016% Fe(NH4)2(S04)~-6H20, KzHP04, 0.16% KH2P04, 0.3% DL-methionine, 2.7 72 glucose, and 3.6% sucrose. The sugars were autoclaved separately as a 1.5-L solution. The remaining medium components were added to 5.4 L of water contained in the fermentation jar. The pH of this salt solution was adjusted to 7.3 with 50% NaOH prior to sterilization.
Biotechnol. Prog., 1990, Vol. 6, No. 5
The fermentor was operated a t 25 "C with an air flow rate of 1 w m (volume of air per volume of culture broth per minute) and 5 psig of back pressure. The dissolved oxygen level was maintained a t a minimum of 50% of saturation through agitation control; pH was controlled at 6.5 through the addition of 30% ammonium hydroxide. For fed batch fermentations, stock sugar or sugar and methionine solutions, a t a glucose or sucrose concentration of 375 g/L, were fed over the course of the fermentation with a Chrontrol Model L6004 microprocessor/controller controlling the feed rate. When methionine was also fed, 24 g of methionine (as per the batch fermentation medium) was added to the appropriate sugar solution volume to be fed over the growth phase of the fermentation. The feed addition rate was adjusted every 2 h during the rapid growth phase and every 3 h during the "productive" phase (details of the feed profile development will be presented in a subsequent section). Assays: Cell Mass. Duplicate 3-mL whole broth samples were vacuum-filtered on Millipore AP20 preweighed prefilters. The cakes were washed with 10 mL of distilled water, 10 mL of 2 M HC1, and 10 mL of distilled water and dried overnight a t 95 "C. Duplicates were averaged to obtain DCW (dry cell weight). Glucose/Sucrose. Glucose concentration was measured by using a YSI Model 2000 glucose and L-lactate analyzer. Glucose and sucrose concentrations were determined enzymatically with the Boehringer Mannheim sucrose/glucose test kit. Fructose. Fermentation broth fructose concentrations were measured enzymatically with the Boehringer Mannheim glucose/fructose test kit. Cephalosporins. HPLC analysis was used to measure cephalosporin concentrations. Equipment and operating conditions were identical with those of Kupka and Shen (1983) with the exception that the mobile phase consisted of 0.02 M ammonium acetate (pH 5.7) and 2% acetonitrile (Dr. Yong-Qiang Shen, BIOPURE Inc., personal communication). Methionine. Culture broth methionine concentrations were measured by HPLC. Methionine was eluted from an Alltech 4.6 mm X 150 mm 5-pm C-18 column with a 10% acetonitrile and 0.1% trifluoroaceticacid mobile phase (Dorothy Phillips, Millipore-Waters Corp., personal communication) pumped a t a flow rate of 1 mL/min. Invertase. Yeast invertase was purchased from Boehringer Mannheim Biochemicals Inc. The effect of methionine inhibition on invertase activity was quantified by using a modified version of the procedure outlined by Tracey (1963).
Feed Profile Development The feed profile was designed to control growth at a constant specific rate through the addition of a growthlimiting carbon source. Culture growth rate was reduced from a specific growth rate of 0.035-0.042 h-l during the rapid growth phase to 0-0.015 h-l during the productive phase at a cell density of -30 g DCW/L. The strategy used in developing the feed profile was based on observations made in early investigations of secondary metabolite production. In batch cultures, there appeared to be two distinct phases of growth, with each having a characteristic growth rate. The first phase had a rapid growth rate. This was followed by an antibiotic production phase, which had a relatively slower growth rate (Demain et al., 1983). These phases have classically been referred to as t h e tropophase and idiophase, respectively. However, these phase distinctions must not be taken literally, because in a medium supporting a slow
Biotechnol. Prog., 1990,Vol. 6, No. 5
335
growth rate during the "tropophase", antibiotic can be produced (Demain et al., 1983). From a process perspective, growth to a desired cell mass followed by a decrease in growth rate for the duration of the fermentation is advantageous. In most antibiotic fermentations, including that of C. acremonium for the production of cephalosporinC, the oxygen transfer capacity limits of the fermentor can be readily exceeded. Oxygen limitation has been shown to severely reduce cephalosporin C yield by reducing the activity of desacetoxycephalosporin C synthetase and desacetylcephalosporin C o-acetyltransferase (Scheidegger et al., 1988). The feed strategy employed is effective in avoiding oxygen limitation during the fermentation and is typical of a fed batch industrial process. A Chrontrol Model L6004 controller was used to control feed rate. This controller permits a desired feed rate profile to be programmed prior to the start of a fermentation. In this study, the feed rate was adjusted every 2 h during the rapid growth phase and every 3 h during the productive phase of a fermentation. The feed profile was developed by solving the mass balance equations for the system over a given 2- or 3-h time interval and by subsequently using these results to determine the feed rate for the following interval. The amount of sugar to be fed over a given time interval was determined by solving the mass balance equations for the system. The balance on cell mass can be written as d(xV)/dt = p x V and can be integrated when p is constant to yield
(1)
x2V2= (xlV1)ept (2) The substrate balance for a fermentor can be described by
change in sugar mass = amount of sugar fed amount of sugar consumed (3) Employinga constant concentrationsugar feed, the amount of sugar fed can be written as F(~)SF, where SFis the sugar concentration in the feed and F(t) is the time-dependent feed rate. At specific growth rates less than pmar,sugar is utilized for both growth and maintenance. Under such conditions, substrate utilization with time can be described by
+
substrate utilization = [d(xV)/dt](l/Y,) m(xV) (4) where Yg is the true growth yield and m is the maintenance coefficient. By using eq 4, eq 3 can be written as d(SV)/dt = F(t)SF - [d(xV)/dtI(l/Y,) - m(xV) (5) Under sugar limitation, the mass of sugar in the fermentor will be approximately constant and approach zero. Under these conditions, eq 5 can be simplified to F(t)SFdt = d(xV)/Y, + m(xV)dt (6) By using eq 1, d(xV)/p can be substituted for xV dt, resulting in F(t)S,dt = d(xV)/Y,
+ m d(xV)/p
(7) If we only consider what occurs during a relatively short time interval, we can assume that the sugar feed rate is approximately constant. With this assumption and simplification of eq 7, the amount of substrate fed during
Table I. Maximum SDecific Growth Rate Data maximum specific growth rate, h-1 tswe of fermentation overall (glucose/ sucrose) 0.059 f 0.005 sucrose 0.047 0.008 ~
~
~~
a given time interval can be described by
+
FSFAt = ( l / Y g m/p)A(xV) (8) From eq 8 and 2, it is evident that the feed rate for a given time interval necessary to control growth at a constant specific growth rate over the course of a fermentation can be determined if the initial cell mas is known as well as Yg and m. Batch fermentations, using the defined fermentation medium described under Experimental Methods, were used to determine a true growth yield of 0.442 f 0.056 g DCW (g of sugar)-l h-l and an initial cell mass of 19.5 g DCW/ fermentor (The value of the true growth yield was later confirmed by using the equation 1 / Y = l / Y g m/p. A true growth yield of 0.456 f 0.040 g DCW/g of sugar was calculated by plotting 1/Y vs l / p at several specific growth rates.) At present, there is no published value for the maintenance demand of C. acremonium, so a value of 0.022 g of sugar (g DCW)-l h-l, the maintenance ratio of Penicillium chrysogenum (Righelato et al., 1968),was assumed. The culture growth rate could be effectively controlled at a constant specific rate after a lag period of -24 h by using this approach.
-
+
Results Comparison of Cephalosporin C Production for Cultures Grown under Sucrose and Glucose Limitation. Initial experiments were designed to investigate the use of sucrose as a growth-limiting nutrient and to compare cephalosporin C production of cultures grown under sucrose limitation with those grown under glucose limitation. In order to effectively control culture growth rate at a relatively slow rate, it was necessary to determine the maximum specific growth rate of C. acremonium on glucose and sucrose. The overall maximum specific growth rate, considered to be a measure of the maximum specific growth rate of C-10 on glucose, was measured by using the batch defined medium outlined under Experimental Methods. It has been previously reported (Matsumura et al., 1978) that glucose is used preferentially to sucrose in a defined medium similar to the one employed in this study. However, we observed that 10-20% of the sucrose was utilized prior to glucose exhaustion in these batch fermentations. The maximum specific growth rate of sucrose grown cultures was measured in similar medium excluding glucose and methionine. The results are presented in Table I. On the basis of the results of these batch fermentations, we elected to compare cephalosporin C production between glucose- and sucroselimited cultures grown for the first 80 h of the fermentation at the relatively slow rate of 0.035 h-l. Comparisonsof culture growth, nutrient utilization, and the specific rate of cephalosporin C formation were made between cultures grown under the control of identical (on a mass and volume feed basis) glucose and sucrose limiting feed profiles. While the specific rate of cephalosporin C formation was relatively similar for such cultures, the changes in dry cell weight (Figure la) and residual sugar (Figure lb) concentrations over the course of glucose- and sucrose-fed fermentations were significantly different. Methionine has been observed to inhibit growth in C. acremonium cultures in which glucose and sucrose served as
Biotechnol. Prog., 1990, Vol. 6, No. 5 a
401
V
'
0
20
40
60
80
100
120
b 2o
1
Time (hours)
0
20
40
60
80
100
120
Figure 2. Sucrose, methionine, and In DCW profiles in an exponentially fed sucrose and methionine fermentation (0.035 h-1 rapid growth phase feed profile).
C
31
0
20
40
60
80
100
120
Time (hours)
Figure 1. Comparisonof the effects of batch methionineaddition in glucose- and sucrose-fed fermentations on (a) culture growth, (b)residual sugar levels, and (c) residual methionine levels (0.035 h-l rapid growth phase feed profile). the major carbon sources (Sawada et al., 1980; Zhang et al., 1987). In the aforementioned glucose and sucrose fermentations, 3 g of DL-methionine/L was included in the initial batch. Examination of the growth curves and residual sugar and methionine levels (Figure 1) in these fermentations demonstrates that sucrose-fed cultures were not effectively carbon source limited, thereby reducing cell mass early in such fermentations, while methionine was used a t a similar rate irrespective of sugar fed. In glucosefed fermentations, glucose was efficiently utilized, methionine was slowly utilized over the first -60 h, and growth rate was effectively controlled through sugar limitation. However, similar results were not obtained for sucrose-fed fermentations as residual sucrose levels increased over the first -55 h, indicating that growth rate was not effectively controlled through sucrose limitation. The time a t which residual methionine was exhausted approximated the time at which sucrose became effectively utilized. This result suggests that high levels of methionine interfere with sucrose utilization. Effect of Concurrent Exponential Feeding of Methionine with Sucrose. In order to effectively control growth through sucrose limitation, it was necessary to develop an alternate method of methionine addition.
Concurrent exponential feeding of sucrose with methionine allowed for efficient use of this amino acid, thus preventing an accumulation of exogenous methionine (00.1 g/L). This permitted effectivegrowth rate control after a 1-day lag phase through uninhibited sucrose utilization (Figure 2). A comparison of both the specific cephalosporin C production rate and volumetric cephalosporin C production between batched and exponentially fed methionine cultures in which sucrose was fed is presented in Figure 3. It is evident that the specific rate of cephalosporin C production in fermentations employing the sucrose and methionine concurrent feed strategy was approximately 1.5 times greater than that of fermentations in which methionine was added to the batch. However, due to the greater cell mass over the first -65 h of the former fermentations, the overall cephalosporin C production in the fed methionine fermentations was approximately twice that of fermentations in which methionine was added to the batch. The Effect of Methionine on the Enzyme 8-Fructosidase (Invertase). It was hypothesized that the significant difference in glucose and sucrose utilization in the presence of methionine was due to the effects of methionine on the enzyme invertase, which catalyzes the hydrolysis of sucrose to glucose and fructose. There are several potential modes of action. Methionine could either inhibit the activity of the enzyme, repress its synthesis, or interfere with its secretion. Inhibition was assessed through an in vitro study of the effects of methionine on the activity of yeast invertase at various methionine and sucrose concentrations. The results of this investigation are presented in Figure 4. It is evident that methionine acts as a competitive inhibitor of yeast invertase activity. Comparison of Cephalosporin C Production between Cultures Grown under Glucose and Sucrose Limitation in which Methionine Was Concurrently Fed. Employing the concurrent methioninelsugar feed strategy with glucose-fed cultures only marginally improved specific cephalosporin C production over cultures in which methionine was added to the batch. A comparison of specific cephalosporin C production rates between glucoseand sucrose-limited cultures in which methionine was
Biotechnol. Pmg., 1990, Vol. 6, No. 5
337 a
h L
3 0
2
40
3.01
/
-
0
E
E
30 -
w
9
20
-
10
-
0 a
s O l -104.
0
I
20
'
I
40
'
I
60
'
I
80
.
I
100
,
3.0g/lmet 0.0 0.00
t
I
0.10
0.30
0.20
120
*
0"
1/S (grams sucrosell1ter)A-1
b
Figure 4. Effect of methionine on the activity of invertase (@fructosidase). 40 i
A L
Q
-
31
E
Glucose Sucrose
6
3
E
0
20
40
60
80
100
120
Time (hours) Figure 3. Effect of method of methionine addition (batched vs fed) on (a) the specific rate of cephalosporin C production and (b) cephalosporin C production in fed sucrwe fermentations (0.035 h-1 rapid growth phase feed profile).
concurrently fed is presented in Figure 5. I t is evident that the maximum specific cephalosporin C production rates are approximately 1.5 times higher in the sucroselimited cultures. It was also determined that volumetric cephalosporin C production was 1.5 times larger in the sucrose-fed fermentations. Effect of Tropophase Growth Rates on Cephalosporin C Production. We next investigated the effect of culture growth rate during the rapid growth phase of a fermentation. In these experiments,sucrose limitation with concurrent methionine feeding was employed to control growth and to maximize cephalosporin C production. The specific growth rate for the previously described experiments, in which we attempted to control growth at the relatively slow specific rate of 0.035 h-1, was measured to be 0.036 f 0.004 h-l. We also investigated cephalosporin C production in cultures grown at the relatively rapid rate of 0.041 f 0.005 h-1. A comparison of the specific cephalosporin C production rate of these sucroselimited methionine-fed cultures is presented in Figure 6. The maximum specific cephalosporin C production rate of cultures grown at the relatively slower rate is approximately 1.5 times larger than for the more rapidly grown
-6 0
a
0
0
20
40
60
60
100
120
Time (hours)
Figure 5. Comparison of the specific rates of cephalosporin C production between glucose- and sucrose-limited fermentations in which methionine was concurrently fed with the sugar (0.035 h-l rapid growth phase feed profile).
-
cultures. However, overall cephalosporinC production per fermentation was only 10% greater for the slowly grown cultures due to the greater average cell mass over the course of the high tropophase growth rate fermentations. Effect of Idiophase G r o w t h R a t e s on Cephalosporin C Production. The feed profile employed to control culture growth rate was designed such that the specific growth rate was reduced from a high growth phase growth rate to a relatively low growth rate at a projected cell mass of 30 g/L to alleviate the possibility of oxygen limitation during the latter course of a fermentation. The latter phase of the fermentation in which growth is relatively slow has classically been referred t o as the idiophase or the productive phase (Demain et al., 1983). We investigated cephalosporin C production during this phase at several specific growth rates between 0 and 0.015 h-l. While the effect of varying idiophase growth rate on the specific rate of cephalosporin C production was not as significant as varying the tropophase growth rate under the fermentation
Biotechnol. Prog., 1990,Vol. 6,No. 5 40
30
-
1
3
r
p=0.035hour-I p0.041 hour-I
5
0
-
0
E
ew
3
20
p=O.OOI 5 hour-1 ~ = 0 . 0 0 8hour-1
.-c
L
0
n
-m 0
10
c n 0
E
Ew
0
v
P 0 7-
n
a .10
I
20
40
60
80
100
120
Time (hours)
0
20
40
60
80
100
120
Time (hours)
Figure 6. Effect of specific growth rate during the rapid growth phase on the specific rate of cephalosporin C production.
Figure 7. Effect of specific growth rate during the productive phase on the specific rate of cephalosporin C production.
conditions employed, the general trend observed during this phase is presented in Figure 7. A slower rate of decay from the maximum specific rate of cephalosporin C production is observed for cultures grown at a relatively faster productive phase growth rate, which promotes slightly higher overall specific cephalosporin C production.
It was observed that sucrose-limited cultures (achieved by concurrent methionine feeding) produce cephalosporin C at higher specific rates than comparable glucoselimited ones. Kennel and Demain (1978) observed that while sucrose supported lower growth rates in batch cultures than glucose, it yielded significantly greater antibiotic production levels. They suggested that this behavior could be attributed t o glucose repression/ inhibition of antibiotic formation and noted that there was roughly an inverse correlation between specific growth rate and the specific rate of antibiotic production. The data from this investigation suggest that there is another factor influencing cephalosporin C synthesis. Sucrose- and glucose-limited cultures grown under the control of identical feed profiles had statistically indistinguishable growth rates. In addition, while glucose represses the ring expansion enzyme desacetoxycephalosporin C synthetase (Behmer and Demain, 1983; Zanca and Martin, 1983) and inhibits 6-(L-cY-aminoadipyl)-L-cysteinyl-Dvaline (ACV) synthetase (Zhang et al., 1989), we do not believe the differences in cephalosporin C production in glucose- and sucrose-limited C-10 cultures can solely be attributed to carbon catabolite repression due to larger residual glucose levels in the glucose-fed culture broth. Although feedback control of sugar addition was not employed in this study, glucose and sucrose measurements made every -12 h over the course of a fermentation demonstrated that growth rate control through carbon source limitation was achieved after the first -24 h. Glucose-limited fermentations typically had a residual glucose level of
