S. Oda et al.
Process Biochemistry xxx (xxxx) xxx–xxx
As for the L–L IBRtac, 800 μl of the 1-day broth was mixed into 80 ml
tacky properties (Figs. 1D and 4 ). Indeed, the BM micro-pieces were
present after vigorous mixing and standing for 4 days. First of all,
Sunrose SLD-F1 and FM were selected as the favorable insoluble binder
material for sticking MS micro-particles to each other.
of the CA-medium containing 2.5 g of the MS (HB-2051) and 100 mg of
the BM (SLD-FM). After vigorous agitation, the mixture was poured 4
glass vials (50 ml; 30 mm i.d.) by 15 ml. The precultivation, cultivation,
and determination of the products were done by the manner same as
those of the S–L IBR.
3.2. Floating test of MS and collection test of BM in an MS layer
2
.7. Analytical procedures
It was assumed that the collection efficiency of the BM micro-pieces
in the MS layer depends on the floating rate of the MS because the
specific gravity of the BM was larger than 1.0. The floating rate of the
MS and BM to a liquid surface was estimated by measuring the turbidity
at OD610. As shown in Fig. 5, OD610 values of HB-2051, HB-2051 plus
SLD-F1, and HB-2051 plus SLD-FM were 0.813, 1.435, and 1.614 at
60 min, respectively. The floating rates (OD610 decrease for first 15 min)
of HB-2051, HB-2051 plus SLD-F1, and HB-2051 plus SLD-FM were
0.895, 0.191, and 0.127, respectively (Table 2). Thus, it was worried
that the collection rate of the BM in the MS layer was not so high.
The collection rate of the BM micro-pieces in the MS layer was es-
timated by measuring sediment dry weight on a bottom of the vial. As
shown in Table 2, while 22.5 ± 7.9% of SLD-F1 settled on the bottom
of the vial, the sedimentation rate of SLD-FM reached 29.9 ± 2.5%.
Thus, the collection rates of SLD-F1 and SLD-FM were 77.5 ± 7.9 and
70.1 ± 2.5%, respectively. In conclusion, SLD-F1 whose collection rate
was higher than that of SLD-FM and it was selected as the best BM.
Although the collection rate of SLD-F1 in an MS layer was limited to
77.5 ± 7.9%, the MS–BM layer was stably held for 3 days with re-
volution (60 rpm). Thus, SLD-F1 is efficiently effective for maintenance
of the MS–BM layer.
All of the conversion products, citronellal, 2-methylcyclohexanone,
-octanone, citronellic acid, and citronellyl acetate were determined by
2
the GLC. Stereochemistry of all products was not analyzed. Citronellal
and 2-octanol were determined by below conditions: the column
(
0.25 mm i.d. × 60 m) contained Equity-5 (Supelco Co., Ltd.,
Bellefonte, PA), the column and detector temperatures were 150 and
60 °C, respectively, the carrier gas was He (20 cm/sec), and the split
ratio was 100:1.
1
2-Methylcyclohexanone was determined by below condition: the
column (0.25 mm i.d. × 60 m) contained SUPELCOWAX (Supelco Co.,
Ltd.), the column and detector temperatures were 90 and 160 °C, re-
spectively, the carrier gas was He (20 m/sec), and the split ratio was
1
00:1.
Citronellic acid was determined by below condition: the column
0.25 mm i.d. × 60 m) contained Equity-5, the column and detector
temperatures were 150 and 255 °C, respectively, the carrier gas was He
20 cm/sec), and the split ratio was 100:1. Citronellyl acetate was de-
(
(
termined by below condition: the column (0.25 mm i.d. × 60 m) con-
tained Equity-5, the column and detector temperatures were 200 and
210 °C, respectively, the carrier gas was He (20 cm/sec), and the split
ratio was 100:1.
3.3. Application of L–L IBRtac to bioconversion with an actinomycete
3
. Results and discussion
As mentioned above, Sunrose SLD-F1 can sustain its tackiness and
effectively stuck the MS and microbial cells (Figs. 1,3 and 4). The
MS–BM layer formed on a liquid-surface did not collapse by overlaying
a hydrophobic organic solvent such as dimethylsilicone oil. Next, the
novel interface cultivation system with the MS and BM, L–L IBRtac, was
applied to some microbial transformations in order to verify its wide
availability.
The L–L IBR which is an interfacial bioconversion system with a
fungal cells–MS layer generally enables to produce high accumulations
of hydrophobic products [16,25–29]. However, the application of the
system has been limited to well-growing fungi. The applications to
slow-growing fungi and actinomycetes, and unicellular bacteria and
yeasts are very difficult because of the collapse of a cells–MS layer by
overlaying of hydrophobic organic solvent. To overcome the dis-
advantage of the former L–L IBR, the BM micro-pieces was added into
the cells–MS layer in order to attach of MS to each other (L–L IBRtac).
First, the MS–BM layer was characterized in terms of the stability,
floating ability and collection rate. Next, the availability of the L–L
IBRtac was estimated by applying the system to some microbial trans-
formations.
Rhodococcus hoagii NBRC 3730 catalyzes many kinds of oxidation of
alcohols [33] and sulfides [34]. The authors have also reported that this
actinomycete catalyzes the oxidation of citronellol [18,32], 2-methyl-
cyclohexanol [22], and 2-octanol [22]. First, the L–L IBRtac was applied
to these microbial oxidations by the actinomycete, and the efficacy of
the L–L IBRtac was compared with that of the SmC and the TLP.
As shown in Fig. 6, citronellol, 2-methylcyclohexanol, and 2-octanol
were not oxidized in SmC because these substrates exhibited strong
biotoxicity at the level of only 1% in the SmC. The coenzyme-re-
generating cycle might be shut by killing the actinomycete cells by the
toxic substrates. In the TLP system, the biotoxicity of substrates were
partially alleviated except 1% citronellol because n-decane played as a
reservoir of toxic substrate [35,36]. Five % of 2-methylcyclohexanol
and 2-octanone could be efficiently oxidized in the TLP, whereas, 2-
methylcyclohexanone and 2-octanone were decreased after 7th day. It
was assumed that the both products might be decomposed via Bayer-
Villiger oxidation [37,38]. Moreover, an organic phase in the TLP
playing as an extractant of product [38] and an oxygen vector [36,39]
reduced by vigorous agitation. Thus, TLP system may be also un-
favorable to practical bioconversions.
3.1. Stability test of an MS–BM layer formed on a water-solvent interface
In the case that a fungal mat is not enough formed on the MS layer,
the fungal cells–MS layer easily collapses by overlaying with hydro-
phobic organic solvent as shown in Fig. 1A and B. It is expected that the
collapse of the cells–MS layer could be prevented by addition of tacky
BM micro-pieces (Fig. 1C and D). First, the BM having sustained
stickiness was screened for the protection of the fungal cells–MS layer
in the L–L IBRtac. In a preliminary experiment, it was confirmed that
starch, sodium alginate, and polyvinyl alcohol could not be used as the
BM because of their water-soluble properties. The BM must be water-
insoluble and maintain tacky property for a long time (Fig. 1C).
As the water-insoluble and tacky micro-pieces, two kinds of car-
boxymethyl cellulose, Sunrose SLD-F1 and SLD-FM, were selected. The
physicochemical properties of both BMs are shown in Table 1. The
supplementary effects of the BM are shown in Fig. 3. Both BMs sus-
tained an MS–BM layer under the condition with rotation (60 rpm) for 3
On the other hand, the L–L IBRtac effectively catalyzed all the oxi-
dations. The accumulation of citronellal, 2-methylcyclohexanone, and
2-octanone reached 3.1 (16 days), 2.3 (12 days), and 32.9 g/l (12 days)
in spite of strong biotoxicity of the substrates and/or products without
collapse of the actinomycete cells–MS–BM layer. The Bayer-Villiger
oxidation of the ketones produced was partially repressed compared
with the TLP system (Fig. 6). The repression of Bayer-Villiger oxidation
in the L–L IBR was also observed in the oxidation of n-decane to 4-
2
days by addition more than 5 mg/vial (0.7 mg/cm -liquid surface). The
suitable effect of the BM was largely due to the water-insoluble and
4