Microbiology (2005), 151, 135–143
DOI 10.1099/mic.0.27458-0
The effect of specific growth rate on protein
synthesis and secretion in the filamentous
fungus Trichoderma reesei
Tiina M. Pakula, Katri Salonen,3 Jaana Uusitalo and Merja Penttilä
VTT Biotechnology, PO Box 1500 (Tietotie 2, Espoo), FIN-02044 VTT, Finland
Correspondence
Tiina Pakula
[email protected]
Received 2 July 2004
Revised
23 September 2004
Accepted 24 September 2004
Trichoderma reesei was cultivated in chemostat cultures on lactose-containing medium. The
cultures were characterized for growth, consumption of the carbon source and protein production.
Secreted proteins were produced most efficiently at low specific growth rates, 0?022–0?033 h”1,
the highest specific rate of total protein production being 4?1 mg g”1 h”1 at the specific
growth rate 0?031 h”1. At low specific growth rates, up to 29 % of the proteins produced were
extracellular, in comparison to only 6–8 % at high specific growth rates, 0?045–0?066 h”1. To
analyse protein synthesis and secretion in more detail, metabolic labelling of proteins was
applied to analyse production of the major secreted protein, cellobiohydrolase I (CBHI, Cel7A).
Intracellular and extracellular labelled CBHI was quantified and analysed for pI isoforms in
two-dimensional gels, and the synthesis and secretion rates of the molecule were determined.
Both the specific rates of CBHI synthesis and secretion were highest at low specific growth
rates, the optimum being at 0?031 h”1. However, at low specific growth rates the secretion
rate/synthesis rate ratio was significantly lower than that at high specific growth rates, indicating
that at low growth rates the capacity of cells to transport the protein becomes limiting. In
accordance with the high level of protein production and limitation in the secretory capacity, the
transcript levels of the unfolded protein response (UPR) target genes pdi1 and bip1 as well as the
gene encoding the UPR transcription factor hac1 were induced.
INTRODUCTION
The filamentous fungus Trichoderma reesei produces a
variety of extracellular cellulases and hemicellulases that
hydrolyse plant-derived polysaccharides into monomeric
sugars which in turn are used as a source of carbon and
energy by the fungal cells (for reviews see Biely & Tenkanen,
1998; Penttilä, 1998; Penttilä et al., 2004). The cellulolytic
system of the fungus consists of synergistically acting
endoglucanases, cellobiohydrolases and b-glucosidases. The
enzymes active on hemicellulose include xylanases and a
mannanase as well as various enzymes cleaving hemicellulose side chains. The most abundant of these enzymes are
the cellobiohydrolases and endoglucanases, with cellobiohydrolase I (CBHI) comprising the major part of the total
extracellular protein produced by the fungus (Keränen &
Penttilä, 1995).
3Present address: University of Helsinki, Biocentrum Helsinki, Cell and
Protein Production Unit, PO Box 63, (Haartmaninkatu 8, Biomedicum),
00014 University of Helsinki, Finland.
Abbreviations: CBHI or Cel7A, cellobiohydrolase I; D, dilution rate; 2D,
two-dimensional; EGI or Cel7B, endoglucanase I; UPR, unfolded protein
response.
0002-7458 G 2005 SGM
Regulation of the cellulase genes on various carbon sources
has been studied in detail. The cellulase genes are to a large
extent induced in a coordinate manner in the presence
of cellulose, its hydrolytic products, or certain oligosaccharides, such as sophorose and lactose (Ilmén et al., 1997;
Foreman et al., 2003). For the individual hemicellulase
genes also, specific and partially overlapping induction
mechanisms have been anticipated to exist based on
differential expression of the genes on various carbon
sources (Margolles-Clark et al., 1997). Wide-domain carbon
catabolite repression has been shown to control the
expression of both cellulase and hemicellulase genes in the
presence of glucose (Ilmén et al., 1996, 1997; MargollesClark et al., 1997; Strauss et al., 1995; Takashima et al.,
1996). Many of the transcription factors involved in
regulation of the genes, CREI mediating carbon catabolite
repression, the repressor ACEI, the activator ACEII, and
the CCAAT binding complex Hap2/3/5, have been characterized in molecular detail (reviewed by Kubicek & Penttilä,
1998; Mach & Zeilinger, 2003; Schmoll & Kubicek, 2003).
Although the carbon-source-dependent transcriptional
regulation of the cellulase genes in T. reesei is fairly well
characterized, information on cellulase gene expression
and production of the proteins at different physiological
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 17:56:12
Printed in Great Britain
135
T. M. Pakula and others
states of the cells is scarce. Furthermore, very little information is available on the cellular responses to protein production. T. reesei has the potential to produce extracellular
proteins in very large quantities, which sets a demand for
the cells to adjust the capacity of protein synthesis and
transport to the level required, and may also provoke
stress responses within the cells. To obtain information on
protein production and factors affecting the processes at
different growth rates of the organism, we have analysed
carbon-limited chemostat cultures of the strain RUT-C30
in detail. Specifically, the capacity of the cells to synthesize
and secrete proteins has been studied using metabolic
labelling of the proteins, a methodology previously set up
for analysis of protein synthesis and secretion in defined
culture conditions (Pakula et al., 2000). In addition, the
expression levels of the major cellulase genes cbh1 and egl1,
as well as genes involved in protein folding and transport,
have been analysed under these conditions.
METHODS
Strain and cultivation conditions. Trichoderma reesei strain Rut-
C30 (ATCC 56765) (Montenecourt & Eveleigh, 1979) was used in
this study. Composition of the culture medium was as follows:
(NH4)2SO4 7?6 g, KH2PO4 5?0 g, MgSO4.7H2O 0?5 g, CaCl2.H2O
0?2 g, CoCl2 3?7 mg, FeSO4.7H2O 5 mg, ZnSO4.7H2O 1?4 mg,
MnSO4.7H2O 1?6 mg l21 and lactose 20 g l21 (in batch culture) or
8 g l21 (in continuous culture). The inoculum was prepared by
transferring 26107 spores into a 500 ml flask containing 200 ml
growth medium (lactose 20 g l21). The preculture was grown for
4 days in a conical flask at 30 uC with shaking at 200 r.p.m., and
finally transferred to the bioreactor (1?8 l laboratory fermenter,
Chemap) to the final volume of 1?5 l growth medium. The cultivations in the bioreactors were carried out at 30 uC, with aeration of
1?3 VVM (volumes of air per volume of liquid per minute) and stirring
with impeller tip speed 2 m s21. The pH was kept at 4?8±0?2 by
the addition of 10 % KOH. Occasionally Struktol SB 2023 (Schill &
Seilacher) was added to prevent foaming. When performing the
carbon-limited chemostat cultivations, the feed (growth medium
containing 8 g lactose l21) was started at the time when the lactose
in the batch phase was nearly consumed, about 40 h after transferring the preculture into the fermenter. The feed was supplied at a
constant rate, specific for each dilution rate used in the study. For
every dilution rate studied, a new culture was started.
Usually steady state in the cultures was achieved after cultivation of
four to five residence times. Steady state of the cultures was monitored
by measuring biomass dry weight and dissolved oxygen concentration
in the culture, carbon dioxide content of the exhaust air, as well as
lactose, phosphate and ammonium concentrations in the culture broth.
Analysis of the chemostat cultures. Dry weight was measured
by filtering and drying mycelium samples at 105 uC to a constant
weight (24 h). Residual lactose and glucose in the culture filtrate
were measured using either HPLC or an enzymic test kit for lactose
(Boehringer Mannheim 176 303) and the GOD-Perid method for
glucose (Boehringer Mannheim 124 036). The amount of phosphate
was measured as described by Basset et al. (1987). An enzymic test
kit (Boehringer Mannheim 1112 732) was used for measuring
ammonium concentration. Soluble protein was analysed using the
Bio-Rad Protein Assay. The carbon content of the biomass was
determined using a Carlo Erba C/N analyser, and the value for
carbon content determined by Nielsen & Villadsen (1994) was used
for extracellular proteins.
136
Analysis of intracellular methionine. Cells resuspended in
double-distilled water to 13 mg biomass dry weight ml21 were disrupted by sonication (Pakula et al., 2000) and boiled for 15 min.
The extracts were centrifuged at 14 000 g for 10 min. Sulphosalicylic
acid was added to the supernatant to a final concentration of 5 %
(w/v), and the samples were centrifuged at 14 000 g for 10 min. The
amino acids were analysed from the supernatant using HPLC.
Analysis of intracellular protein. Cell extracts were prepared and
the protein amount in the extracts was measured as described previously (Pakula et al., 2000). Samples of the cultures were filtered
through Millipore HVLP02500 filters to collect the mycelium, and
the mycelium was washed with double-distilled water and resuspended in 20 mM N-ethylmaleimide, 10 mM NaN3 (3–7 mg biomass dry weight ml21). The cells were disrupted by sonication
(868 s with an MSE 150 W sonicator, 18 mm amplitude, 30 s cooling on ice between the sonication cycles). The cell lysates were first
lyophilized, and then resuspended in two-dimensional (2D) lysis
buffer [9 M urea, 2 % (v/v) Triton X-100, 286 mM b-mercaptoethanol,
2 % (v/v) Pharmalyte 3–10; 560 mg biomass dry weight per 25 ml
buffer], 3 vols 2D sample buffer [the lysis buffer containing 0?5 %
(v/v) Triton X-100] was added, and the samples were incubated at
room temperature for 1 h. Insoluble material was removed by
centrifugation. Proteins were analysed using the Bio-Rad Protein
Assay Kit.
Metabolic labelling of the proteins. Metabolic labelling of the
proteins was carried out essentially as described previously (Pakula
et al., 2000). Aliquots (25 ml) of the chemostat culture were gently
transferred into shake flasks on a rotary shaker (210 r.p.m., 30 uC),
and the labelling was started immediately after the transfer. To avoid
changes in concentration of the nutrients, fresh medium was added
into the shake flasks. Medium was added at intervals of 15 s, the
amount of the added medium corresponding to the amount of
medium fed into the chemostat culture during the same time
period. [35S]Methionine (Amersham SJ 1015, in vivo cell labelling
grade, 1000 Ci mmol21, 10 mCi ml21) was added to the shake flasks,
either 5?5 or 12?5 mCi (mg biomass dry weight)21 in the labelling
experiment (1 Ci=3?761010 Bq), and samples of 1 ml were
collected at short intervals. The samples were rapidly filtered
through Millipore HVLP02500 filters to separate the culture
medium and mycelium. The mycelium on the filters was washed
with 10 ml double-distilled water and frozen immediately in liquid
nitrogen.
Analysis of the labelled proteins. The cells were disrupted by
sonication and the cell extracts were prepared as previously described (Pakula et al., 2000). Protein samples prepared from the
cell extracts and culture medium were subjected to 2D gel electrophoresis (Pakula et al., 2000) for analysis of specific proteins. Equal
amounts of total protein (measured using the Bio-Rad Protein Assay
Kit as described above) were loaded in the gels. The amount of
protein loaded was confirmed to be in a range giving a linear response between the signal analysed and the sample volume. Typically,
20 mg intracellular protein or 3 mg extracellular protein was loaded.
The 2D gels were analysed using a phosphorimager (Molecular
Dynamics). The intracellular labelled protein per culture volume
and per biomass amount was calculated by taking into account the
protein content of the biomass and the biomass amount under each
of the culture conditions. For analysis of total labelled protein, the
proteins in the cell extracts and in culture supernatant were precipitated using TCA, and the radioactivity in the TCA-insoluble material
was measured by scintillation counting (Pakula et al., 2000).
Parameters describing protein synthesis and secretion. The
parameters, protein synthesis and secretion rate, the mean synthesis
time of specific proteins, and the minimum secretion time of the
molecules were determined essentially as described by Pakula et al.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 17:56:12
Microbiology 151
Protein synthesis and secretion in Trichoderma
(2000). The amount of labelled protein per biomass and culture
volume in cell extract and in culture supernatant was plotted against
time (scintillation counting of TCA-insoluble material was used for
quantification of total labelled protein, and specific labelled proteins
were quantified in 2D gels using a phosphorimager). The specific
rate of protein synthesis was determined as the slope of the linear
part of the curve representing intracellular labelled protein at the
early time points where protein secretion was not yet detectable. The
specific rate of protein secretion was determined as the slope of
the linear part of the curve representing extracellular labelled protein.
The specific rates of protein synthesis and secretion per biomass and
time unit were normalized with the ratio of [35S]methionine taken
up by the cells to the total intracellular methionine. The mean
synthesis time for a specific protein was determined by extrapolating
the linear part of the curve representing the amount of the labelled
protein in cell extract to the abscissa. The intercept of the curve and
the abscissa corresponds to half the synthesis time of the protein
(see also Braakman et al., 1991; Horwitz et al., 1969; Loftfield &
Eigner, 1958). The minimum secretion time of the protein was
determined as the distance of the intercepts of the intracellular and
extracellular protein curves extrapolated to the abscissa.
Northern analysis. Mycelium samples of 50 ml were filtered,
washed with equal volume of 0?7 % NaCl, frozen immediately in
liquid nitrogen, and stored at 280 uC. Total RNA was isolated using
the Trizol Reagent (Gibco-BRL), essentially according to the manufacturer’s instructions. Northern blotting and hybridization on nitrocellulose filters were carried out according to standard procedures
(Sambrook et al., 1989). Five micrograms of total RNA isolated
from each sample was used in the Northern analysis. cDNAs of the
genes (cbh1, E00389; egl1, M15665; pdi1, AJ222773; sar1, Y08636;
ypt1, AJ277108) were used as probes. The signals were normalized
against the signals of act1 encoding actin.
RESULTS
Growth parameters and protein production in
chemostat cultures of T. reesei Rut-C30
Growth characteristics and protein production of T. reesei
strain Rut-C30 were studied in carbon-limited chemostat
cultures on lactose-containing minimal medium at dilution
rates of 0?021–0?076 h21. At steady state, reached typically
after four to five residence times, the levels of produced
biomass and extracellular protein and residual amount of
lactose were constant (Fig. 1a). Between dilution rates of
0?021 and 0?033 h21, the amount of biomass increased
Fig. 1. Growth, lactose consumption and production of extracellular proteins in chemostat cultures of T. reesei RUT-C30.
(a) Biomass dry weight (#), extracellular protein ($) and residual lactose (m) in steady-state cultures with different dilution
rates; (b) the inverse of specific growth rate plotted against the inverse of residual lactose concentration; (c) the specific
consumption rate of lactose (g lactose consumed per g biomass dry weight per hour, #) and the specific production rate of
extracellular total protein (mg protein produced per g biomass dry weight per hour, $); (d) growth yield Yx (g biomass formed
per g lactose consumed, #) and yield of extracellular proteins Yp (g protein produced per g lactose consumed, $). D is the
dilution rate which corresponds to the specific growth rate of the organism in the steady-state cultures, except for the culture
at the dilution rate 0?076 h”1, which exceeded the maximal specific growth rate so that a steady state could not be reached.
The error bars indicate SEM in 3–5 samples collected during the steady state in the cultures.
http://mic.sgmjournals.org
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 17:56:12
137
T. M. Pakula and others
significantly with increasing dilution rate (the dilution rate
in the cultures corresponds to the specific growth rate),
after which the biomass amount remained constant until
the maximal specific growth rate of the organism was
exceeded (between 0?066 h21 and 0?076 h21), leading to
washout of the biomass (Fig. 1a). The growth yield (grams
of biomass formed per gram of lactose consumed) at the
different dilution rates displayed a similar trend (Fig. 1d).
By applying the Monod equation [1/m=(1/S)(KS/mmax)+1/
mmax, where m is the specific growth rate, S the substrate
concentration, KS the substrate saturation constant and
mmax the maximal specific growth rate], and plotting 1/m vs 1/
S (Fig. 1b), the values for mmax and KS were deduced. The
obtained value, mmax 0?068 h21, was close to the maximal
specific growth rate measured in batch cultures on the same
culture medium, which was 0?073 h21 (data not shown).
The value obtained for KS was 0?025 g l21.
The amounts of residual lactose and its cleavage product,
glucose, in the cultures were low. The concentration of
lactose varied between 0?012 and 0?150 g l21 (Fig. 1a)
and the amount of glucose was below the detection limit.
The specific consumption rate of lactose increased with
increasing specific growth rate from 0?068 g g21 h21 to
0?137 g g21 h21 (Fig. 1c). The equation introduced by Pirt
(1965) (qS=m/Ymax+ms, where q is the specific rate of
substrate consumption, Y the molar growth yield and m
the maintenance coefficient) has often been used for
estimation of the amount of energy source required for
the maintenance of the cells (ms), assuming that the energy
requirement for the maintenance and the maximal growth
yield are constant over the different growth rates and that
no metabolites other than carbon dioxide and water are
produced. A fairly good linear regression was obtained
for the values of the specific lactose consumption rate
plotted against the specific growth rate, giving the values
0?027 g g21 h21 and 0?60 g g21 for the maintenance
coefficient and the maximal growth yield, respectively.
The highest specific production rate of extracellular proteins was obtained at low specific growth rates (0?022–
0?033 h21), the maximal value being 4?1 mg g21 h21 at
the specific growth rate 0?031 h21 (Fig. 1c). However, at
the lowest specific growth rate studied (0?021 h21), the
specific protein production rate was reduced close to the
level measured at high specific growth rates of 0?045–
0?066 h21 (1?4–1?6 mg g21 h21). Similarly, the yield of
extracellular protein produced per the amount of carbon
source consumed was the highest at the specific growth
rate 0?031 h21 (Fig. 1d). At low specific growth rates
(0?021–0?031 h21), both the biomass yield and the yield
of protein produced increased with increasing specific
growth rate, after which much less protein was produced
per amount of lactose consumed, whereas the production
of biomass related to lactose consumption remained at a
high level (Fig. 1d). Carbon balances accounting for biomass, secreted protein and CO2 produced in the cultures
showed a mean closure of 90±7 % total carbon consumed.
138
For the determination of the balances, the carbon content
of the mycelial samples was measured, and the elemental
composition CH1?72O0?31N0?27S0?004 described by Nielsen &
Villadsen (1994) was used for secreted proteins.
To elucidate how the capacity of protein production was
directed to production of intracellular and extracellular
proteins, the specific synthesis rates of intracellular and
extracellular protein were compared at the different specific
growth rates. The amount of intracellular protein in the cell
extracts was determined, and the specific synthesis rate of
intracellular proteins (grams of intracellular protein per
gram of biomass dry weight per hour) was deduced. The
specific synthesis rate of intracellular and extracellular
protein as well as the specific rate of total protein synthesis
(combined synthesis of extracellular and intracellular protein) plotted against the dilution rate in the culture are
shown in Fig. 2(a). The specific rate of total protein
synthesis increased with increasing specific growth rate
until a constant level was reached at the specific growth
rate 0?045 h21 (Fig. 2a). At low specific growth rates,
between 0?022 and 0?033 h21, a markedly higher percentage of the total protein synthesis was directed to
Fig. 2. The synthesis of intracellular and extracellular proteins
in the chemostat cultures. (a) The specific rate of total protein
synthesis ($), the specific rate of synthesis of intracellular
proteins (%) and the specific rate of synthesis of extracellular
proteins (&) (g protein per g biomass dry weight per hour) at
different growth rates in the chemostat cultures. (b) Synthesis
of intracellular proteins (%) and extracellular (&) proteins as a
percentage of the total proteins produced.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 17:56:12
Microbiology 151
Protein synthesis and secretion in Trichoderma
production of extracellular proteins compared to that of
cultures with the high specific growth rates of 0?045–
0?066 h21 (Fig. 2b). At most, at the specific growth rate
0?022 h21, 29 % of the synthesized proteins were extracellular, whereas at high specific growth rates (0?045–
0?066 h21), only 6–8 % of the synthesized proteins were
those transported into the culture medium.
Synthesis and secretion of CBHI at different
specific growth rates
To be able to study in more detail the synthesis and transport of the extracellular proteins at different specific growth
rates, a series of metabolic labelling experiments were set
up using the methodology described by Pakula et al. (2000).
For the labelling, aliquots of the chemostat cultures were
withdrawn into shake flasks on a rotary shaker. A feed of
fresh medium was supplied to maintain the cells under
conditions resembling those in the bioreactor. Newly
synthesized proteins were metabolically labelled by the
addition of [35S]methionine to the cultures, and the synthesis and secretion of the major cellulase Cel7A (CBHI)
was analysed as a model system. Frequent samples were
collected from the labelled cultures and intracellular and
extracellular protein extracts were subjected to 2D gel
analysis. The amount of labelled CBHI in the samples was
quantified and plotted against time to determine the parameters which indicated the efficiency of protein synthesis
and secretion. The specific synthesis rate of the protein was
measured at the early time points of the labelling experiment, before any secretion of the protein into the culture
medium was detectable. At these early time points, the most
prominent pI isoforms of the protein were those representing the early, nascent forms of the protein, whereas the
extracellular CBHI was represented by multiple pI isoforms
forming a pattern typical for secreted CBHI under these
conditions (Pakula et al., 2000).
The specific rates of CBHI synthesis and secretion (the
amount of protein synthesized or produced in the culture
medium per biomass and time unit) at different specific
growth rates are shown in Fig. 3. The highest specific
synthesis and secretion rates of labelled CBHI were both
obtained at the specific growth rate 0?031 h21, which is
in accordance with the result of total protein production
into the culture medium. However, at low specific growth
rates (0?022–0?045 h21), under conditions where the
specific CBHI synthesis rate was high, the ratio between
the secretion rate and the synthesis rate was much lower
than the ratio at high specific growth rates. At 0?031 h21, the
secretion rate was 37 % of the synthesis rate, whereas at
0?066 h21, the secretion rate was 62 % of the synthesis
rate. The result indicates that although CBHI was efficiently synthesized at the low specific growth rates, the
protein secretion capacity limits protein production under
these conditions. The mean time of synthesis of CBHI
molecules and the minimum time of secretion of the
protein did not differ significantly at the different specific
growth rates. The mean synthesis time of CBHI was
http://mic.sgmjournals.org
Fig. 3. The specific synthesis rate (%) and the specific secretion rate (&) of 35S-labelled CBHI in the aliquots of the
chemostat cultures labelled with [35S]methionine. The specific
rates are shown as PIU per mg biomass dry weight per min,
where PIU is the CBHI signal in the 2D gels analysed by a
phosphorimager.
3?9±0?3 min and the minimum secretion time of the
molecules was 12?6±0?2 min. The values are close to
the ones obtained previously for cultures carried out at the
specific growth rate 0?07 h21 (Pakula et al., 2000).
Northern analysis of genes encoding cellulases,
factors involved in protein transport and
folding, and ribosomal proteins in the
chemostat cultures
Northern analysis was carried out to elucidate whether or
not the cellulase genes were differentially expressed at different specific growth rates and whether or not the proposed limitation in the secretion capacity was manifested
at the expression level of genes involved in protein transport and folding. The steady-state transcript levels of cbh1
and egl1 (encoding CBHI/Cel7A and EGI/Cel7B, respectively) were the highest at the specific growth rate 0?031 h21
(Fig. 4a, b). Thus, the high synthesis rate of labelled CBHI
can, at least in part, be explained by the high transcript
levels of the genes, as well as the high specific production
rate of total extracellular proteins at this growth rate, since
CBHI and EGI form the major part of extracellular proteins produced by T. reesei. However, by comparing the
protein production rates and the transcript levels over
the full range of dilution rates studied, it is obvious that
the transcript levels do not fully explain the differences in
protein production levels at the different dilution rates,
but that other factors are involved as well. Especially at
dilution rates above 0?06 h21, the mRNA levels of cbh1
and egl1 were relatively high compared to the synthesis rate
of CBHI (Fig. 3) and the specific production rate of total
extracellular protein into the culture medium (Fig. 1).
The mRNA levels of the chaperon/foldase genes bip1 and
pdi1 were higher at low specific growth rates (0?022–
0?033 h21) than at high specific growth rates (0?045–
0?066 h21) (Fig. 4c). This is in accordance with the
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 17:56:12
139
T. M. Pakula and others
Fig. 4. Steady-state mRNA levels of selected genes encoding cellulases, proteins involved in protein folding and transport or
ribosomal proteins in chemostat cultures of T. reesei. Samples of total RNA (5 mg) collected from chemostat cultures were
subjected to Northern analysis. The specific mRNA signals were quantified using a phosphorimager and normalized against
the signal of actin gene act1. The mRNA signals of cbh1 (a), egl1 (b), bip1 ($) and pdi1 (#) (c), hac1 (the induced form of
the transcript $, the uninduced form #) (d), ypt1 ($) and sar1 (#) (e) and rpl4 ($) and rps11 (#) (f) are shown as relative
signals, the value for the specific growth rate 0?045 h”1 set as 1. The error bars shown indicate SEM in two or three parallel
Northerns. Examples of the mRNA signals on Northern filters as well the rRNA signals in the gels stained with acridine orange
are shown in panel (g).
observation that at these growth rates a high amount of
secreted protein was synthesized; however, the secretion
capacity was postulated to be a limiting factor under these
conditions, which would necessitate an increased capacity
of the folding machinery. The genes bip1 and pdi1 are
known to be induced by the unfolded protein response
(UPR) pathway, which is mediated by the hac1 gene. As
the UPR pathway is activated, a shorter form of the hac1
transcript is generated via splicing of an unconventional
intron and via truncation of the transcript at the 59 flanking region. The shortened form of the transcript then
produces the transcription factor HACI (Saloheimo et al.,
2003). The hac1 transcript levels were slightly induced at
low growth rates (Fig. 4d), the presence of the induced
form being more pronounced in the conditions corresponding to the highest rate of synthesis and secretion of
extracellular proteins, at the dilution rate 0?031 h21. In
contrast, the transcript levels of ypt1 and sar1, encoding
proteins involved in protein transport, were not induced
at low specific growth rates (Fig. 4e). For comparison, the
transcript levels of rps11 and rpl4, encoding ribosomal
140
components, were also analysed (Fig. 4f). The Northern
analysis showed that the signals of rps11 and rpl4 increased
with increasing growth rate.
DISCUSSION
A detailed analysis of carbon-limited chemostat cultures
of the strain RUT-C30 was carried out to elucidate the
capacity of T. reesei to synthesize and secrete proteins under
different physiological conditions and under the influence
of different factors affecting protein production by the
fungus. The cultures were characterized for growth, consumption of the carbon source, and production of extracellular proteins. Metabolic labelling of newly synthesized
proteins was used for analysis of the synthesis and secretion of the major secreted protein produced by the fungus
CBHI, and Northern analysis was carried out to analyse
cellulase gene expression and selected cellular stress responses to protein production under these conditions.
The dilution rates studied (0?021–0?076 h21) covered the
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 17:56:12
Microbiology 151
Protein synthesis and secretion in Trichoderma
range from a very low specific growth rate to a rate exceeding the maximal specific growth rate of the fungus. The
maximal specific growth rate estimated from the chemostat
data was 0?068 h21, which is close to the value 0?073 h21,
determined in batch bioreactor cultures for the strain on
the same medium, and to the value previously obtained
for T. reesei strain C5 on lactose-containing minimal
medium, 0?07 h21 (Chaudhuri & Sahai, 1994). At low
specific growth rates (0?021–0?033 h21), the fungal biomass
increased significantly with increasing growth rate, but at
higher specific growth rates (0?033–0?066 h21), the biomass amount remained approximately constant. Cultivation at the dilution rate 0?076 h21 resulted in washout of
the biomass, as predicted. The specific consumption rate
of the carbon source lactose increased until the critical
dilution rate was reached. Assuming that the amount of
energy source used for maintenance as well as the
maximal growth yield would be constant over the range
of growth rates studied, the maintenance coefficient
(0?027 g g21 h21) and maximal growth yield (0?60 g g21)
were determined (Pirt, 1965). A good linear regression
was obtained for the specific lactose consumption rate
over the dilution-rate range studied, indicating that the
assumptions of the Pirt equation were valid under these
conditions.
Protein production into the culture medium was most
efficient at low specific growth rates (below 0?033h21). The
specific production rate of extracellular proteins increased
first with increasing dilution rate, reaching the maximal
level, 4?1 mg g21 h21, at the specific growth rate
0?031 h21, after which the production rate decreased
significantly. Similar trends for specific protein production rates at different dilution rates have previously been
reported for lactase production by Rut-C30 from continuous cultures on lactose medium (Castillo et al., 1984)
and for production of cellulase activity by T. reesei strain
C5 (Chaudhuri & Sahai, 1994) and have been modelled for
cellulase productivity on medium containing xylose and
sorbose by T. reesei strain QM9414 (Schafner & Toledo,
1992). In accordance with the other data on production of
biomass and extracellular protein, at low specific growth
rates (0?022–0?033 h21), a higher proportion of the total
protein synthesis was also directed to production of extracellular proteins compared to the cultures at higher
specific growth rates (0?045–0?066 h21). At low specific
growth rates, up to 29 % of the proteins synthesized were
extracellular and at the high growth rates only 6–8 % were
extracellular.
Production of extracellular proteins by filamentous fungi,
such as a-amylase in Aspergillus oryzae (Carlsen et al., 1996;
Spohr et al., 1998) and glucoamylase in Aspergillus niger
(Pedersen et al., 2000; Schrickx et al., 1993; Withers et al.,
1998), has been shown to be growth-associated in many
cases. However, examples of growth-rate-disassociated
production are also known, such as recombinant protein
production by Fusarium venenatum (Wiebe et al., 2000). As
http://mic.sgmjournals.org
active growth of the cells and the fungal hyphae requires
efficient transportation of, for example, cell wall material,
protein transport is expected to take place efficiently in the
growing hyphae. However, a majority of the extracellular
enzymes produced by T. reesei are needed for degradation
of polymeric compounds derived from plant material to
provide the fungus with a source of carbon and energy.
Therefore, production of hydrolytic enzymes would be
beneficial for the fungus under low nutrient conditions in
which easily metabolized carbon sources, such as glucose,
are not available and growth might be slow.
Taking into account the data from the wide range of growth
rates studied, the results in our study as well as those
obtained by other groups (Castillo et al., 1984; Chaudhuri
& Sahai, 1994; Schafner & Toledo, 1992) indicate that
production of the hydrolytic enzymes that constitute the
major part of the extracellular proteins produced by T.
reesei under these conditions is not directly growth-rate
associated. In terms of specific protein production rate as
well as the yield of extracellular protein per amount of
carbon source consumed, the production of extracellular
proteins was maximal at the specific growth rate 0?031 h21,
but at higher growth rates a significant reduction in production was observed. However, in the range of the low
dilution rates 0?021–0?031 h21, both protein and biomass
production are positively correlated with the specific
growth rate. The reduction in biomass and protein production at very low specific growth rates is due to the
maintenance requirement of the cells. At low specific
growth rates, the specific consumption rate of the carbon
and energy source lactose was very low. At the dilution
rate 0?021 h21, 0?068 g lactose g21 h21 was consumed,
of which the estimated maintenance coefficient 0?027 g
g21 h21 forms 40 %. The high substrate consumption for
maintenance under these conditions would limit production of both biomass and extracellular protein.
As cellulase gene expression by T. reesei is known to be
regulated by the carbon source, one explanation for the
differences in the protein production levels at different
growth rates could be differences in the concentrations of
the residual carbon sources. However, the concentration
of the repressing carbon source glucose in the cultures in
this study was very low, and thus unlikely to cause repression of cellulase production. In addition, the strain RutC30 is defective in the cre1-mediated carbon catabolite
mechanism (Ilmén et al., 1996) and therefore cellulase
production by the strain is partially derepressed even in a
glucose medium. Further, the residual levels of lactose were
low in the cultures and no positive correlation between
cellulase expression and the residual lactose concentration
in the cultures was observed. Thus the residual lactose in
the cultures was unlikely to act as an inducer for cellulase
production. Interestingly, under high protein production
conditions at low specific growth rates, the specific consumption rate of lactose was low. Previous studies have
shown that cellulase gene expression is induced in batch
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 17:56:12
141
T. M. Pakula and others
cultures after the carbon source is exhausted (Ilmén et al.,
1997). It is possible that the low consumption rate of the
carbon source might trigger a similar type of induction
as does carbon and energy source starvation. In natural
habitats of the fungus, this type of mechanism might be
important for activation of synthesis of hydrolytic enzymes
to enable sequestering of easily metabolizable carbon
sources from complex plant polymers.
To address the question of transcriptional regulation of
cellulase genes at different specific growth rates, we carried
out a Northern analysis of the genes cbh1 and egl1. The
transcript levels of both genes were slightly increased at
low specific growth rates, thus being in accordance with
the results on production of total extracellular proteins. In
addition, metabolic labelling of the cultures was used for
analysis of the synthesis and secretion of CBHI specifically.
The results showed that both the specific synthesis and
secretion rate of labelled CBHI were the highest at the
low specific growth rate 0?031 h21. Thus, the differences
in transcript levels at the different growth rates could, at
least in part, explain the differences in the amounts of the
corresponding proteins synthesized per time unit. However,
further inspection of the full range of dilution rates studied
indicated that the transcript levels alone do not fully
explain the differences in protein production. Especially at
high specific growth rates, the specific protein production rate (Fig. 1) and the specific synthesis rate of CBHI
(Fig. 3) were lower than that which could have been
anticipated based on the transcript levels of the major
cellulase genes cbh1 and egl1.
Interestingly, the metabolic labelling studies also showed
that the ratio between the specific secretion rate and the
synthesis rate of labelled CBHI was much lower at the
low specific growth rates than at the high specific growth
rates. The result indicates that at the low growth rates
the capacity of the cells to transport the proteins would
become limiting as the amount of secreted protein synthesized increases. Accumulation of unfolded proteins in
the endoplasmic reticulum is known to activate the UPR,
in both lower and higher eukaryotes. In filamentous
fungi, UPR induction has been reported to take place
under conditions in which protein transport or folding
is impaired by treatment of the cultures with different
chemical agents or under conditions where a heterologous
protein is produced (Mulder et al., 2004; Ngiam et al.,
1997; Pakula et al., 2003; Punt et al., 1998; Saloheimo et al.,
1999, 2003; van Gemeren et al., 1997). In addition, we have
recently shown that the UPR pathway is activated along
with cellulase gene induction when the cultures are shifted
from a glucose repressed state to an induced state (Collén
et al., 2004). In the present study, the transcript levels of the
UPR transcription factor gene hac1 and the UPR target
genes pdi1 and bip1 were more abundant at low specific
growth rates than at high growth rates. The result indicates that the UPR pathway was activated in response to
increased production of secreted proteins, such as CBHI,
142
and also in response to the postulated limitation in the
transport process under those conditions. For comparison,
no such increase at low specific growth rates was observed
in the transcript levels of ypt1 and sar1, which encode
proteins involved in other functions in protein transport.
It has been previously shown that ypt1 and sar1 transcript
levels are not induced by endoplasmic reticulum stress
caused by treatment of the cultures with DTT, an agent
known to induce UPR via inhibition of protein folding and
transport (Saloheimo et al., 2004).
In conclusion, our study offers new insight into the protein
production characteristics of the industrially important
host organism T. reesei. Production of extracellular proteins was favoured at low specific growth rates, which is
beneficial for fed-batch processes where efficient protein production without extensive biomass formation is
required. However, our study indicated that the capacity
of the cells to transport the newly synthesized proteins is a
limiting factor in the process under these conditions, and,
for optimal protein production, strategies to improve the
transport step should be designed. In addition, information on cellulase gene regulation at different growth rates
and on consumption of the carbon and energy source was
obtained to form a basis for further studies, especially in
the field of starvation signalling for cellulase expression.
ACKNOWLEDGEMENTS
Dr Aristos Aristidou and Dr Marilyn Wiebe are thanked for useful
consultation and comments on the work, and Riitta Nurmi and Aili
Grundström are thanked for skilful technical assistance. The work was
funded by the EU Biotechnology programme BIO-CT94-2045 and
Roal Oy, and the work was a part of the research programme ‘VTT
Industrial Biotechnology’ (Academy of Finland; Finnish Centre of
Excellence programme, 2000–2005, Project no. 64330).
REFERENCES
Basset, J., Deney, R. C., Jeffery, G. F. & Mendham, J. (1987). Vogel’s
Textbook of Quantitative Inorganic Analysis, 4th edn. New York:
Wiley.
Biely, P. & Tenkanen, M. (1998). Enzymology of hemicellulose
degradation. In Trichoderma and Gliocladium, pp. 25–47. Edited by
G. E. Harman & C. P. Kubicek. London. Bristol: Taylor & Francis
Ltd.
Braakman, I., Hoover-Litty, H., Wagner, K. R. & Helenius, A. (1991).
Folding of influenza hemagglutinin in the endoplasmic reticulum.
J Cell Biol 114, 401–411.
Carlsen, M., Nielsen, J. & Villadsen, J. (1996). Growth and a-amylase
production by Aspergillus oryzae during continuous cultivations.
J Biotechnol 45, 81–93.
Castillo, F. J., Blanch, H. W. & Wilke, C. R. (1984). Lactase production
in continuous culture by Trichoderma reesei Rut-C30. Biotechnol Lett
6, 593–596.
Chaudhuri, B. K. & Sahai, V. (1994). Comparison of growth and
maintenance parameters for cellulase biosynthesis by Trichoderma
reesei-C5 with some published data. Enzyme Microb Technol 16,
1079–1083.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 17:56:12
Microbiology 151
Protein synthesis and secretion in Trichoderma
Collén, A., Saloheimo, M., Bailey, M., Penttilä, M. & Pakula, T. M.
(2004). Protein production and induction of the unfolded protein
response in Trichoderma reesei strain Rut-C30 and its transformant
expressing endoglucanase I with a hydrophobic tag. Biotechnol
Bioeng, (in press).
Foreman, P. K., Brown, D., Dankmeyer, L. & 14 other authors
(2003). Transcriptional regulation of biomass-degrading enzymes
in the filamentous fungus Trichoderma reesei. J Biol Chem 278,
31988–31997.
Horwitz, M. S., Scharff, M. D. & Maizel, J. V. (1969). Synthesis and
Penttilä, M., Limón, C. & Nevalainen, H. (2004). Molecular biology of
Trichoderma and biotechnological applications. In Mycology, Vol. 20,
Handbook of Fungal Biotechnology, pp. 413–427. Edited by D. K.
Arora. 2nd edn. New York, Basel: Marcel Dekker.
Pirt, S. J. (1965). The maintenance energy of bacteria in growing
cultures. Proc R Soc B 163, 224–231.
Punt, P. J., van Gemeren, I. A., Drint-Kuijvenhoven, J., Hessing,
J. G., van Muijlwijk-Harteveld, G. M., Beijersbergen, A., Verrips,
C. T. & van den Hondel, C. A. (1998). Analysis of the role of the gene
assembly of adenovirus 2. I. Polypeptide synthesis, assembly of
capsomeres, and morphogenesis of the virion. Virology 39, 682–694.
bipA, encoding the major endoplasmic reticulum chaperone protein
in the secretion of homologous and heterologous proteins in black
Aspergilli. Appl Microbiol Biotechnol 50, 447–454.
Ilmén, M., Thrane, C. & Penttilä, M. (1996). The glucose repressor
Saloheimo, M., Lund, M. & Penttilä, M. E. (1999). The protein
gene cre1 of Trichoderma: isolation and expression of a full-length
and a truncated mutant form. Mol Gen Genet 251, 451–460.
disulphide isomerase gene of the fungus Trichoderma reesei is
induced by endoplasmic reticulum stress and regulated by the
carbon source. Mol Gen Genet 262, 35–45.
Ilmén, M., Saloheimo, A., Onnela, M.-L. & Penttilä, M. E. (1997).
Regulation of cellulase gene expression in the filamentous fungus
Trichoderma reesei. Appl Environ Microbiol 63, 1298–1306.
Saloheimo, M., Valkonen, M. & Penttilä, M. (2003). Activation
Keränen, S. & Penttilä, M. (1995). Production of recombinant
mechanisms of the HACI-mediated unfolded protein response in
filamentous fungi. Mol Microbiol 47, 1149–1161.
proteins in the filamentous fungus Trichoderma reesei. Curr Opin
Biotechnol 6, 534–537.
Saloheimo, M., Wang, H., Valkonen, M., Vasara, T., Huuskonen, A.,
Riikonen, M., Pakula, T., Ward, M. & Penttilä, M. (2004). Charac-
Kubicek, C. & Penttilä, M. (1998). Regulation of production of plant
polysaccharide degrading enzymes by Trichoderma. In Trichoderma
and Gliocladium, pp. 49–72. Edited by G. E. Harman & C. P.
Kubicek. London: Taylor & Francis Ltd.
terization of secretory genes ypt1/yptA and nsf1/nsfA from two
filamentous fungi: induction of secretory pathway genes of
Trichoderma reesei under secretion stress conditions. Appl Environ
Microbiol 70, 459–467.
Loftfield, R. B. & Eigner, E. A. (1958). The time required for the
Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). Molecular Cloning:
synthesis of a ferritin molecule in rat liver. J Biol Chem 231, 925–943.
Mach, R. L. & Zeilinger, S. (2003). Regulation of gene expression in
a Laboratory Manual, 2nd edn. Cold Spring Harbor, NY: Cold
Spring Harbor Laboratory.
industrial fungi: Trichoderma. Appl Microbiol Biotechnol 60, 515–522.
Schafner, D. W. & Toledo, R. T. (1992). Cellulase production in
Margolles-Clark, E., Ilmén, M. & Penttilä, M. (1997). Expression
continuous culture by Trichoderma reesei on xylose-based media.
Biotechnol Bioeng 39, 865–869.
patterns of ten hemicellulase genes of the filamentous fungus
Trichoderma reesei on various carbon sources. J Biotechnol 57,
167–179.
Schmoll, M. & Kubicek, C. P. (2003). Regulation of Trichoderma
Montenecourt, B. S. & Eveleigh, D. E. (1979). Selective screening
cellulase formation: lessons in molecular biology from an industrial
fungus. A review. Acta Microbiol Immunol Hung 50, 125–145.
methods for the isolation of high yielding cellulase mutants of
Trichoderma reesei. Adv Chem Ser 181, 289–301.
Schrickx, J. M., Krave, A. S., Verdoes, J. C., van den Hondel, C. A.,
Stouthamer, A. H. & van Verseveld, H. W. (1993). Growth and
Mulder, H. J., Saloheimo, M., Penttila, M. & Madrid, S. M. (2004).
The transcription factor HACA mediates the unfolded protein
response in Aspergillus niger, and up-regulates its own transcription.
Mol Genet Genomics 271, 130–140.
product formation in chemostat and recycling cultures by Aspergillus
niger N402 and a glucoamylase overproducing transformant,
provided with multiple copies of the glaA gene. J Gen Microbiol
139, 2801–2810.
Ngiam, C., Jeenes, D. J. & Archer, D. B. (1997). Isolation and
Spohr, A., Carlsen, M., Nielsen, J. & Villadsen, J. (1998). a-amylase
characterisation of a gene encoding protein disulphide isomerase,
pdiA, from Aspergillus niger. Curr Genet 31, 133–138.
production in recombinant Aspergillus oryzae during fed-batch and
continuous cultivations. J Ferment Bioeng 86, 49–56.
Nielsen, J. & Villadsen, J. (1994). Bioreaction Engineering Principles,
Strauss, J., Mach, R. L., Zeilinger, S., Hartler, G., Stoffler, G.,
Wolschek, M. & Kubicek, C. P. (1995). Cre1, the carbon catabolite
pp. 480. New York: Plenum.
Pakula, T. M., Uusitalo, J., Saloheimo, M., Salonen, K., Aarts, R. J.
& Penttilä, M. (2000). Monitoring the kinetics of glycoprotein
synthesis and secretion in the filamentous fungus Trichoderma reesei:
cellobiohydrolase I (CBHI) as a model protein. Microbiology 146,
223–232.
Pakula, T. M., Laxell, M., Huuskonen, A., Uusitalo, J., Saloheimo, M.
& Penttilä, M. (2003). The effects of drugs inhibiting protein
secretion in the filamentous fungus Trichoderma reesei. Evidence
for down-regulation of genes that encode secreted proteins in the
stressed cells. J Biol Chem 278, 45011–45020.
repressor protein from Trichoderma reesei. FEBS Lett 376, 103–107.
Takashima, S., Nakamura, A., Iikura, H., Masaki, H. & Uozumi, T.
(1996). Cloning of a gene encoding a putative carbon catabolite
repressor from Trichoderma reesei. Biosci Biotechnol Biochem 60,
173–176.
van Gemeren, I. A., Punt, P. J., Drint-Kuyvenhoven, A.,
Broekhuijsen, M. P., van’t Hoog, A., Beijersbergen, A., Verrips,
C. T. & van den Hondel, C. A. (1997). The ER chaperone encoding
bipA gene of black Aspergilli is induced by heat shock and unfolded
proteins. Gene 198, 43–52.
Pedersen, H., Beyer, M. & Nielsen, J. (2000). Glucoamylase
Wiebe, M. G., Robson, G. D., Shuster, J. & Trinci, A. P. (2000).
production in batch, chemostat, and fed-batch cultivations by an
industrial strain of Aspergillus niger. Appl Microbiol Biotechnol 53,
272–277.
Growth-rate-independent production of recombinant glucoamylase
by Fusarium venenatum JeRS 325. Biotechnol Bioeng 68, 245–251.
Penttilä, M. (1998). Heterologous protein production in Tricho-
Withers, J. M., Swift, R. J., Wiebe, M. G., Robson, G. D., Punt, P. J.,
van den Hondel, C. A. M. J. J. & Trinci, A. P. J. (1998). Optimisation
derma. In Trichoderma and Gliocladium, pp. 365–382. Edited by
G. E. Harman & C. P. Kubicek. London: Taylor & Francis Ltd.
and stability of glucoamylase production by recombinant strains of
Aspergillus niger in chemostat culture. Biotechnol Bioeng 59, 407–418.
http://mic.sgmjournals.org
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 17:56:12
143