Production of poly(3-hydroxybutyrate) by Halomonas boliviensis in an air-lift reactor

Bacterial strain and maintenance

Halomonas boliviensis LC1
^T
(= DSM 15516
^T
) was maintained at 4 °C on solid HM medium 15], containing (%, w/v): NaCl, 4.5; MgSO
₄
?·?7H
₂
O, 0.025; CaCl
₂
?·?2H
₂
O, 0.009; KCl, 0.05; NaBr, 0.006; peptone, 0.5; yeast extract, 1.0; glucose, 0.1;
and agar, 2.0. The pH of the medium was adjusted to 7.5 using 3 M NaOH.

Hydrolysis of starch obtained from cassava

Starch from cassava was suspended in 0.05 M citrate buffer solution (pH?=?6) in a
stirred tank, after which ?-amylase, Fungamyl, was added. Hydrolysis of starch was allowed to proceed at 60 °C and at a stirring
speed of 200 rpm for 24 hrs. The total sugar concentration in the resulting hydrolysate
solution was determined to be 200 g L
^?1
containing mainly glucose and maltose in a weight ratio of 1:2.

Culture medium and conditions for PHB production in shake flasks

Seed culture and PHB production media were formulated as described previously 16]. Seed culture medium contained (%, w/v): NaCl, 2.5; MgSO
₄
°7H
₂
O, 0.25; K
₂
HPO
₄
, 0.05; NH
₄
Cl, 0.23; FeSO
₄
°7H
₂
O, 0.005; glucose, 1.0; monosodium glutamate (MSG), 0.2 and TRIS, 1.5. The seed culture
was at 30 °C and 200 rpm for 15 hrs (OD
₆₀₀
?=?0.5-0.55; CDW?=?1.0?±?0.1 g L
^?1
). 5 % (v/v) was used to inoculate the PHB production medium, containing (%, w/v):
NaCl, 2.5; MgSO
₄
°7H
₂
O, 0.5; K
₂
HPO
₄
, 0.22; NH
₄
Cl, 0.4; FeSO
₄
°7H
₂
O, 0.005; MSG, 0.2; and the following carbohydrates (%, w/v): 1) 1.5 glucose and 0.5
xylose, 2) 1.4 glucose and 0.6 maltose, 3) 1.5 starch hydrolysate and 0.5 xylose,
4) 2.5 starch hydrolysate, respectively. A low amount, 0.2 % (w/v) MSG, was added
to the production medium to induce its depletion by H. boliviensis during the cultivation. All experiments were performed in 1000 ml shake flasks at
35 °C and 200 rpm of agitation.

Air-lift reactor construction and operation for PHB production

An air-lift glass with geometrical dimensions as depicted in Fig. 1 was used in this study. Air at a flow rate of 2 L min
^?1
(2.5 vvm) was pumped through a millipore glass net that acted as air sparger. Air
bubbles created by the sparger enter to an inner tube (known as the riser) which is
concentrically surrounded by an ungassed inner tube (known as the downcomer). The
different gas holdup in the raiser and downcomer zones result in dissimilar bulk densities
of the liquid culture medium in these regions which causes circulation of the liquid
in the reactor by gas-lift action. On top of the riser and downcomer a column section
of 180 mm height was placed in order to separate air bubbles from the culture medium.
The upper most section of the reactor had three openings: the middle one for inserting
a pH electrode and the ones on either side as an air outlet and for addition of acid,
base or antifoam to the reactor, respectively. Moreover on one side of the downcomer,
and at 105 mm from the sparger, a sampling device was coupled. A 245 mm height external
jacket was also included in the reactor to maintain the temperature constant by water
recirculation.

Halomonas boliviensis was cultured in a 1 L Erlenmeyer flask containing 50 ml of seed culture medium at
30 °C for 15 hrs. The seed culture was inoculated in 700 ml of PHB production medium
in the reactor and cultivation was allowed to proceed for 36 hrs. pH of the medium
was controlled at 7.5?±?0.3 by adding 3 M NaOH or 3 M HCl in the reactor whenever
a deviation from the set point value was observed. Water at 35 °C originating from
a thermostatic water bath was recirculated through the jacket of the reactor. Antifoam
was added when needed. Samples were withdrawn from the reactor at different time intervals.

Genome sequences studied and evolutionary analysis of proteins

Four genome sequences corresponding to strains of the family Halomonadaceae were selected for this study. The strains were Halomonas boliviensis LC1
^T
(=DSM 15516
^T
) 18], Halomonas elongata DSM 2581
^T24], Halomonas sp. TD01 25], and Chromohalobacter salexigens DSM 3043 26].

A total of 6901 alignments of clusters of orthologous proteins (COGs) of 88 microorganisms,
as classified in COGs 27] and EggNOG 28] databases, were obtained from that described by Puigbò et al. 29]. The protein sequences of these 88 microorganisms were used as reference for the
evolutionary analysis. Protein sequences of strains of the family Halomonadaceae were included in the alignments with the references for each corresponding COG using
the Muscle program 30] included in the MEGA 5 software package 31] with default parameters. Unrooted maximum likelihood phylogenetic trees for each
COG were constructed using MEGA 5 under a WAG with frequencies (+F) model, with uniform
mutation rates among amino acid sites and complete deletion of gaps and missing data.

Determination of biochemical pathways

Halomonas boliviensis was taken as the reference strain for the determination of biochemical pathways.
The functional annotation of enzymes involved in the pentose phosphate and Entner-Doudoroff
pathways was accomplished by analyses of protein sequences. Genes of H. boliviensis were aligned to others in databases to attain its corresponding functional annotation.
To ensure the biological meaning, only one high-quality information as annotation
to the genes from many results was chosen. BLAST was used to accomplish functional
annotation combined with different databases. BLAST version: blastall 2.2.21 software
(provided by the National Center for Biotechnology Information) was used for these
studies. Alignment results were obtained using the KEGG, COG, SwissProt, TrEMBL, and
NR databases. A BLAST search was also used to find enzymes in H. elongata, C. salexigens, and Halomonas sp. TD01 that share high identities with those annotated for H. boliviensis.

Quantitative analyses

Cell dry weight (CDW) and PHB content in H. boliviensis were determined as reported previously 15]. Residual cell mass (RCM) concentration was calculated as the difference between
the CDW and PHB concentration, while PHB content (wt.%) was obtained as the percentage
of the ratio of PHB concentration to the CDW as defined by Lee et al. 32]. All analyses were performed in triplicate. Glucose and maltose were determined using
the same HPLC system with a Polypore CA column (Perkin-Elmer), a RI detector at 80 °C
and water as mobile phase at a flow rate of 0.3 ml min
^?1
.