ORIGINAL RESEARCH PAPER

Introduction to bioreactors of shake-flask inocula leadsto development of oxidative stress in Aspergillus niger

Andrew O’Donnell Æ Yantao Bai ÆZhonghu Bai Æ Brian McNeil ÆLinda M. Harvey

Received: 12 December 2006 / Revised: 5 February 2007 / Accepted: 12 February 2007 /Published online: 10 March 2007� Springer Science+Business Media B.V. 2007

Abstract Inoculation of bioreactors with shake-

flask cultures present the organism with an

immediate shift from an environment with little

O2 to one in which O2 is typically at 100%

saturation. The inoculation of such shake-flasks

cultures into bioreactors sparged with 1 vvm air or

1 vvm air/O2 mix i.e. 50% O2 enrichment is an

oxidatively stressful event, as judged by immedi-

ate increases in the intracellular concentrations of

superoxide anion radical (O2�–) (from 4,600 to

11,600 RLU mg DCW–1 and 5,500 to 23,000 RLU

mg DCW–1 respectively) and changes in the

activities of the major antioxidant enzymes super-

oxide dismutase and catalase in all cultures. There

are further effects on metabolic indices,

particularly decreased nutrient consumption in

oxygenated cultures (from 0.16 to 0.12 g starch g

DCW h–1) and decreased protein production,

indicating that inoculation of the bioreactor exerts

a global burden on the cellular metabolic

networks.

Keywords Filamentous fungi � Lag phase �Oxidative stress � Oxygen enrichment

Introduction

Oxidative stress is a natural cellular phenomenon

in which an organism is subject to a change in the

balance between oxidants and antioxidants in

favor of the oxidants (Halliwell and Gutteridge

1999). These oxidants are generally termed reac-

tive O2 species (ROS), and include superoxide

anion radical (O2�–), H2O2, and the hydroxyl

radical (�OH), amongst others (Dalton et al.

1999). ROS can be extremely detrimental to

cellular viability as they are damaging to DNA,

proteins, lipids, and cell membranes (Storz and

Imlay 1999). Nevertheless, ROS generation is an

unavoidable consequence of aerobic life, with a

major proportion of ROS formation within the

cell occurring via electron leakage from the

respiratory chain (Gottlieb 1971; Turrens et al.

1985). Consequently, all aerobic organisms have

developed complex antioxidant defence systems

to combat the deleterious effects of oxidative

stress. These antioxidant systems include enzymes

A. O’Donnell � B. McNeil (&) � L. M. HarveyStrathclyde Fermentation Centre, Department ofBioscience, University of Strathclyde, GlasgowG1 1XW, UKe-mail: b.mcneil@strath.ac.uk

Y. BaiMedway School of Medicine, University ofGreenwich at Medway, Chatham Maritime, KentME4 4TB, UK

Z. BaiDepartment of Technical Support, Ortho-ClinicalDiagnostics (A Johnson & Johnson Company),Cardiff CF14 7YT, UK

123

Biotechnol Lett (2007) 29:895–900

DOI 10.1007/s10529-007-9336-3

such as superoxide dismutase (SOD), catalase

(CAT), and glutathione peroxidase (GPx), and

non-enzymatic defence mechanisms such as

ascorbate, metallothioneins, and glutathione

(Fridovich 1978; Sies 1993). Additionally, recent

investigations have proposed that, in fungal cells,

morphological adaptation and enhancement of

alternative respiratory pathway activity may play

an important role in minimising ROS formation

under conditions of high O2 availability (Kreiner

et al. 2003; Bai et al. 2003).

Oxidative stress is particularly important as

regards industrial exploitation of the filamentous

fungi since it can have pronounced effects on

product formation (Manjula-Rao and Sureshku-

mar 2001; Bai et al. 2004). Since filamentous fungi

are widely used for the production of both native

and recombinant proteins, it is important to

understand how their productivity can be altered

by oxidative stress. Possibly the most widely

employed method for cultivating filamentous

fungi is submerged liquid culture, i.e. within a

bioreactor. Prior to inoculation into the bioreac-

tor, however, the organism is commonly grown in

shake-flasks, aeration in which is achieved simply

via gas-liquid contact brought about by agitation

of the culture (Gupta and Rao 2003). Such flasks

are commonly plugged with cotton wool or foam

plugs, although this may significantly limit oxygen

transfer, with reported oxygen concentrations in

the headspace of the flask as low as 6%, while

CO2 concentrations may reach 15% (Gupta and

Rao 2003). Thus, a ‘‘typical’’ fermentation inoc-

ulum will involve transfer from a shake-flask

environment with reduced O2 levels, elevated

CO2, and, commonly, reduced levels of carbon

substrate, to a highly aerated, agitated bioreactor

where O2 availability is high and abundant carbon

source is available. Based on the findings of

previous studies (Kreiner et al. 2003; Bai et al.

2004), such a steep change may pose a significant

risk of inducing conditions of oxidative stress

within fungal cultures in the lag phase. This could

have important consequences for cultivation of

industrial filamentous fungi as endogenous

sources of oxidative stress have been shown to

restrict growth and productivity of the culture

(Bai et al. 2003).

In order to investigate whether inoculation of

shake-flask cultures of Aspergillus niger into a

bioreactor does lead to the occurrence of signif-

icant oxidative stress, the concentrations of

superoxide anion radical (O2�–) were measured,

as were the activities of two key antioxidant

enzymes, i.e. superoxide dismutase (SOD) and

catalase (CAT), to gain insight into both sides of

the oxidative stress ‘‘balance’’. The metabolic

activities of oxidatively stressed and control

cultures were also determined, as was the con-

centration of total intracellular protein, to deter-

mine effects of oxidative stress on culture growth

and productivity in lag phase. These measure-

ments were carried out for cultures sparged with

air (control) or 50% O2 enriched air (endogenous

oxidative stress).

Materials and methods

Microorganism and culture conditions

A recombinant strain of A. niger B1-D in which

the hen egg white lysozyme (HEWL) cDNA gene

is placed under the control of the Aspergillus

awamori glucoamylase promoter was used

(Archer et al. 1990a). The maintenance of master

culture and preparation of spore suspensions were

as described by Bai et al. (2003). The composition

of the medium used for the batch cultivations was

described by Wongwicharn et al. (1999a). The

batch fermentation was performed in a 15 l (total

volume) stainless steel bioreactor (MBR Bio

Reactor AG, Wetzikon, Switzerland), with a

working volume of 10 l, which was inoculated

with 4% (v/v) of 48-h-old shake-flask culture

grown at 25�C and 200 rpm. Process parameters

were as described by Bai et al. (2003). In the

processes using O2 enriched air, 50% (v/v) O2

enriched air was sparged at 1 vvm.

Analytical techniques

Biomass was estimated based on the method of

Wongwicharn et al. (1999a). Monitoring of the

superoxide radical was carried out by a lucigenin-

derived chemiluminescence (LDC) method (Bai

896 Biotechnol Lett (2007) 29:895–900

123

et al. 2001). The LDCL was measured with a

luminometer (Biocounter M2500, Lumac bv,

Landgraat, Netherlands) at room temperature.

For enzyme assays, fungal mycelium was sepa-

rated immediately by filtration from broth with a

4.25 cm diameter GF/C filter, particle retention

size of approx 1.2 lm, The filter cake was washed

twice with distilled water prior to resuspending in

50 mM, pH 7.0 phosphate buffer. The fungal cells

were subsequently disrupted in a high-pressure

cell disrupter (Model 4000, Constant Systems

Ltd., Warwick, U.K.). Cell free extracts were

separated from cell debris by centrifugation

(9,000 g, 4�C, 30 min), and were used in the assay

of enzymes activities. Superoxide dismutase

(SOD) activity was measured by its ability to

inhibit the reduction of cytochrome c with super-

oxide radical produced by xanthine/xanthine oxi-

dase system, which was described by Grapo et al.

(1978). Catalase activity was determined by the

decomposition of H2O2 at 240 nm within 30 s

(Aebi 1984). Total intracellular protein was

determined using the Bradford method. Before

measurement, the fungal cells were disrupted and

centrifuged as described above. The concentra-

tion of residual sugar in the medium was mea-

sured according to the method of Dubois (1956),

whilst the ammonium ion concentration was

determined according to the method of Arnold

et al. (1999).

Results and discussion

The occurrence of oxidative stress in lag phase

cultures of A. niger B1-D was determined based

on the balance between oxidants and antioxi-

dants, namely O2�–, SOD, and CAT. The effects on

metabolism as regards culture growth, carbon

source uptake, nitrogen source uptake and pro-

tein production were also measured. As shown in

Fig. 1, in cultures sparged with O2-enriched air,

superoxide anion radical (O2�–) concentrations

were higher than in the control cultures. It is of

particular note that superoxide concentrations

increase greatly immediately after inoculation of

shake-flask cultures into the fermenter. This

occurs both in control and oxygen-enriched cul-

ture and thus it is feasible to comment that

inoculation into a fully aerated bioreactor, in

terms of rapidly increasing reactive oxygen spe-

cies (ROS) concentrations in the culture, does

appear to be an oxidatively stressful event as

regards filamentous fungi. O2�– concentrations

again increase sharply approximately 4 h after

inoculation into the bioreactor in the oxygen-

enriched culture. This coincides with a decrease in

the specific activities of numerous antioxidant

Fig. 1 Superoxide anion radical, superoxide dismutase(SOD) and catalase (CAT) activities in lag phase A. nigercultures sparged with 1 vvm air and 1 vvm air/O2 mix(50% O2 enrichment). Superoxide concentrations weremeasured in terms of relative lights units by a luminom-eter. Symbols: d 1 vvm air, � 50% O2 enrichment

Biotechnol Lett (2007) 29:895–900 897

123

enzymes, particularly superoxide dismutase

(SOD) and catalase (CAT), which are also

outlined in Fig. 1. The increased O2�– concentra-

tions also correspond with decreased growth rate

of the organism at 6 h, as detailed in Fig. 2. Since

it has been proposed that decreased concentra-

tions of ROS during growth can be attributed to

dilution of radical species by a high rate of

formation of daughter cells (Nystrom 1998), it

could also be that the decreased growth rate of

the culture at this point during the lag phase

allows accumulation of ROS in the cell (Bai et al.

2004).

Superoxide dismutase (SOD) and catalase

(CAT) also provide some interesting data, as

outlined in Fig. 1. Upon inoculation of the

cultures into the bioreactor, both cultures show

an immediate increase in the activities of each

of these enzymes. However, O2 enriched

batches show elevated SOD and CAT activity

compared to control cultures during the first 5 h

of cultivation. This demonstrates that upon

inoculation of the bioreactor with shake-flask

cultures, O2 enrichment leads to the sharpest

responses. Thus, these cultures are clearly more

oxidatively stressed than control cultures. After

the first 5 h, however, control cultures show

increased SOD and CAT activities as compared

with oxygenated cultures, suggesting that after

the initial inoculation, oxygen enriched cultures

are better at dealing with the oxidative envi-

ronment than control cultures. This is corrobo-

rated by the superoxide data which shows

increased ROS concentrations in control cul-

tures after 6 h compared to oxygenated batches.

Since SOD and CAT are both decreased in

their activities at this point, it would appear

other cellular defence mechanisms are maybe

being employed to deal with the oxidative

environment. A similar result was noted by

Bai et al. (2003), who proposed that within such

cultures a group of alternative respiratory

enzymes, namely alternative NADH dehydro-

genases, were acting to diminish ROS produc-

tion by the electron transport chain under

highly oxidative environments, and it could be

possible that this is also occurring here.

Figure 2 shows that throughout lag phase the

culture sparged with O2-enriched air has a

slightly elevated dry cell weight (DCW) as

compared with the control but results in a

lower final DCW. The DCW of the control

culture is, for example, 6.62 g/l at the end of lag

phase (24 h), whilst that of the O2-enriched

culture is significantly lower at only 4.77 g/l

(t-test, P < 0.05). Moreover, if the specific

growth rates of these cultures are compared, it

can be clearly noted that O2-enriched air

decreases the specific growth rate of the cul-

tures during lag phase. Similarly, as can be seen

from Fig. 2, the total intracellular protein con-

tent of the O2-enriched culture is generally

lower than the control throughout the lag phase.

The decreased growth rate of O2 enriched

cultures during lag phase can be further corre-

lated with the metabolic activity of the culture

as regards carbon source uptake, as these

demonstrate generally reduced C-source uptake

rates during lag phase compared to the control.

Extrapolation of data from Fig. 3 allows

Fig. 2 Dry cell weights and intracellular proteins concen-trations of lag phase A. niger cultures sparged with either1 vvm air or 1 vvm air/O2 mix (50% O2 enrichment).Symbols: d 1 vvm air, � 50% O2 enrichment

898 Biotechnol Lett (2007) 29:895–900

123

comparison of the specific C-source uptake

rates of each culture, and it can be demon-

strated that O2 enrichment decreases C-source

uptake by the organism. The specific carbon

uptake rate for the control culture is approxi-

mately 0.16 g g DCW h–1, for example, whilst

that of the O2-enriched culture is circa

0.12 g g DCW h–1. Figure 3 also demonstrates

that nitrogen source uptake in the control and

O2-enriched cultures are broadly similar. There-

fore, the decreased protein concentrations in

the oxygenated cultures are unlikely to be a

result of decreased N-source uptake, and must

therefore be attributable to another facet of the

cellular physiology.

Conclusions

The inoculation of a bioreactor with shake-flask

cultures results in an immediate and dramatic

increase in the concentrations of superoxide

anion radicals and the corresponding decrease in

the activity of key antioxidant enzymes, namely

SOD and CAT. The decreased activity of these

enzymes therefore predisposes the organism to

the deleterious effects of excessive ROS concen-

trations, i.e. oxidative stress. Thus, during lag

phase (which is widely regarded as an adaptive

period of growth) the organism is highly stressed,

and this is reflected in the metabolic activity of

the culture during this period. Specifically, oxida-

tive stress as a result of inoculation results in

decreased C-source consumption rates, decreased

protein production, and a decreased specific

growth rate of the culture, as proven by the

creation of an increasingly oxidative environment

through oxygen enrichment. Interestingly, oxygen

enriched cultures generally seem more capable of

dealing with highly oxidative environments than

control cultures, following an initial burst of ROS

generation and corresponding alternations in the

activities of the major defensive enzymes.

Since culture performance is tightly linked to

the ability of the organism to adapt to its environ-

ment during the lag phase, it is necessary to limit

the incidence of oxidative stress following inocu-

lation of a shake-flask into the bioreactor. This

could potentially be achieved by, for example,

lowering aeration during inoculation and lag

phase, and increasing it as cultures enter exponen-

tial growth. By doing so, the organism will be

afforded a much reduced likelihood of becoming

oxidatively stressed following bioreactor inocula-

tion, and as such the period of adaptation will be

shortened. Alternatively, an initial short burst of

highly O2-enriched air may serve to bolster the

cellular defence mechanisms and make them

better able to cope with the highly aerated biore-

actor environment. Whichever path is taken, the

aim must be to minimize the occurrence of oxida-

tive stress during the latter stages of the lag phase,

as it seems that it is this point at which culture

degeneration is most likely to occur, and thus aim

to maximise the culture productivity by maintain-

ing this accurate control of the process aeration.

Fig. 3 Total residual carbohydrate and ammonia concen-trations in lag phase A. niger cultures sparged with either1 vvm air or 1 vvm air/O2 mix (50% O2 enrichment).Symbols: d 1 vvm air, � 50% O2 enrichment

Biotechnol Lett (2007) 29:895–900 899

123

References

Aebi H (1984) Catalase in vitro. Meth Enzymol 105:121–126

Archer DB, Jeenes DJ, MacKenzie DA et al (1990a) Henegg white lysozyme expressed in, and secreted from,Aspergillus niger is correctly processed and folded.Bio Technology 8:741–745

Bai Z, Harvey LM, McNeil B (2001) Use of the chemilu-minescent probe lucigenin to monitor the productionof the superoxide anion radical in a recombinantAspergillus niger (B1-D). Biotechnol Bioeng 75:204–211

Bai Z, Harvey LM, McNeil B (2003) Physiological re-sponses of chemostat cultures of Aspergillus niger(B1-D) to simulated and actual oxidative stress. Bio-technol Bioeng 82:691–701

Bai Z, Harvey LM, White S, McNeil B (2004) Effects ofoxidative stress on production of heterologous andnative protein, and culture morphology in batch andchemostat cultures of Aspergillus niger (B1-D). En-zyme Microb Technol 34:10–21

Dalton TP, Shertzer HG, Puga A (1999) Regulation ofgene expression by reactive oxygen. Annu RevPharmacol Toxicol 39:67–101

Fridovich I (1978) The biology of oxygen radicals. Science201:875–880

Gottlieb SF (1971) Effect of hyperbaric oxygen onmicroorganisms. Annu Rev Microbiol 25:111–152

Grapo JD, McCord JM, Fridovich I (1978) Preparationand assay of superoxide dismutase. Meth Enzymol53:382–393

Gupta A, Rao G (2003) A study of oxygen transfer inshake-flasks using a non-invasive oxygen sensor.Biotechnol Bioeng 84:351–358

Halliwell B, Gutteridge JMC (1999) Free radicals inbiology and medicine. Oxford University Press,Oxford

Kreiner M, Harvey LM, McNeil B (2002) Oxidative stressresponse of a recombinant Aspergillus niger to exog-enous menadione and H2O2 addition. Enzyme MicrobTechnol 30:346–353

Kreiner M, Harvey LM, McNeil B (2003) Morphologicaland enzymatic responses of a recombinant Aspergillusniger to oxidative stressors in chemostat cultures.J Biotechnol 100:251–260

Manjula-Rao Y, Sureshkumar GK (2001) Improvement inbioreactor productivities using free radicals: HOCl-induced overproduction of xanthan gum from Xan-thamonas campestris and its mechanism. BiotechnolBioeng 72:62–68

Nystrom T (1998) To be or not to be: the ultimate decisionof the growth-arrested bacterial cell. FEMS MicrobiolRev 21:283–290

Sies H (1993) Strategies of antioxidant defense. Eur JBiochem 215:213–219

Storz G, Imlay JA (1999) Oxidative stress. Curr OpinMicrobiol 2:188–194

Turrens JF, Alexandre A, Lehninger AL (1985) Ubisem-iquinone is the electron donor for superoxide forma-tion by complex III of heart mitochondria. ArchBiochem Biophys 237:408–414

Wongwicharn A, Harvey LM, McNeil B (1999a) Secretionof heterologous and native proteins, growth andmorphology in batch cultures of Aspergillus nigerB1-D at varying agitation rates. J Chem TechnolBiotechnol 74:821–828

900 Biotechnol Lett (2007) 29:895–900

123