Content uploaded by Yissue Woo
Author content
All content in this area was uploaded by Yissue Woo on Feb 10, 2025
Content may be subject to copyright.
Towards sustainable aquafeeds: Safe and consistent microbial protein
grown on food-processing wastewater
Ramanujam Srinivasan Vethathirri
a,b
, Ezequiel Santillan
a,*
, Yissue Woo
a
, Sara Swa Thi
a
,
Hui Yi Hoon
a
, Stefan Wuertz
a,b,*
a
Singapore Centre for Environmental Life Sciences Engineering, Nanyang Technological University, Singapore, 637551, Singapore
b
School of Civil and Environmental Engineering, Nanyang Technological University, Singapore, 639798, Singapore
ARTICLE INFO
Keywords:
Circular bioeconomy
Resource recovery
Single cell protein
Sustainable animal feed
Essential amino acid
Feed safety
ABSTRACT
Microbial community-based single cell protein (SCP) holds promise as a sustainable source of protein in livestock
feed; yet its feed-safety and consistency in composition and production when using variable real-world waste-
water has not been investigated. Here, the effect of heterogeneity in soybean-processing wastewater on SCP
quality was tested using four replicate sequencing batch reactors over 92 days. The microbial community-based
SCP grown on soybean wastewater demonstrated high consistency, with replicates showing similar patterns of
biomass growth and protein accumulation. The dry microbial biomass exhibited a protein content of 39.8 ±5.8
%, and the yield was 17.7 ±1.7 g dry weight/g soluble total Kjeldahl nitrogen (sTKN). Azospirillum, a nitrogen-
xing bacterium, was the prevalent SCP-producing genus in all replicates at a relative abundance of 40.6 ±5.1
%. The organism was not detected in wastewater, where Lactococcus and Weissella dominated. SCP contained
essential amino acids to supplement conventional animal diets and was deemed safe for sh due to the very low
abundance of sh-pathogen-like sequences (<0.009 %) via metabarcoding. This study demonstrates the con-
sistency of microbial community-based SCP derived from food-processing wastewater and addresses feed safety
through pathogen screening, highlighting its potential to substitute protein in traditional animal feed and
contribute to sustainable aquaculture practices.
1. Introduction
The present warning that ten out of every 100 people in the world are
suffering from severe malnutrition (FAO, 2021) raises serious concerns
for both current and future generations (Godfray et al., 2010). A sus-
tainable answer to protein deciency could be the microbial protein or
single cell protein (SCP) derived from microbial biomass grown on
nutrient rich wastewaters (Asiri and Chu, 2022; Joris et al., 2024; Hül-
sen et al., 2022; Verstraete et al., 2022).
Such a closed-loop approach of recovering valuable protein from
wastewater can help meet the projected protein demand in 2050
(Durkin et al., 2022). SCP is a promising alternative protein-rich diet for
aquaculture animals like sh and shrimp (Asiri et al., 2022; Asiri, 2024;
El Abbadi and Criddle, 2019; Jones et al., 2020; Santillan et al., 2024).
Unlike single-organism-based (axenic) approaches (Ritala et al., 2017;
Suman et al., 2015), a microbial community-based (non-axenic) method
of SCP production reduces overall protein production costs by excluding
steps such as wastewater source sterilization and pH control (Vethathirri
et al., 2021). The approach has great potential for animal feed appli-
cations (Vethathirri et al., 2023), reducing the land and water footprint
and greenhouse gas emissions of traditional farming systems by at least
5 % (Pikaar et al., 2018).
Microbial community-based SCP production from food-processing
wastewaters must account for variable C:N in the inuent wastewater
batches due to uctuations during production in the source industry
(Nayyar et al., 2021). Such differences can affect biomass growth, amino
acid content in the biomass, and the SCP-producing bacterial taxa
(Vethathirri et al., 2023). As a result, distinct microbial communities are
enriched, which could further alter the performance of SCP production
(Vethathirri et al., 2023). Replicate reactors are therefore required to
evaluate the heterogeneity introduced by the variable chemical char-
acteristics of inuent wastewater along with variability due to sampling
and analysis (Prosser, 2010). Further, replication can help assess the
variable behavior of the microbial communities, especially in a
* Corresponding authors.
E-mail addresses: esantillan@ntu.edu.sg (E. Santillan), swuertz@ntu.edu.sg (S. Wuertz).
Contents lists available at ScienceDirect
Cleaner and Circular Bioeconomy
journal homepage: www.elsevier.com/locate/clcb
https://doi.org/10.1016/j.clcb.2025.100139
Received 5 September 2024; Received in revised form 19 December 2024; Accepted 20 January 2025
Cleaner and Circular Bioeconomy 10 (2025) 100139
Available online 27 January 2025
2772-8013/© 2025 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license ( http://creativecommons.org/licenses/by/4.0/ ).
microbial community-based production approach (Santillan and
Wuertz, 2022). Microbial communities may perform differently in terms
of SCP yield or production rate when subject to inuent wastewaters of
variable chemical composition despite replicated experimental condi-
tions (Prosser, 2010). To date, no study has investigated the variability
in microbial community-based SCP production from real
food-processing wastewater using replicate reactors.
SCP can be produced without a microbial inoculum (Hülsen et al.,
2022; Vethathirri et al., 2023) or with an inoculum obtained from a prior
experiment (Hülsen et al., 2018a, 2018b, 2020; Lee et al., 2015). The
former case was explored using purple phototrophic bacteria, which use
light as energy source for the assimilation of organics and nutrients
(Hülsen et al., 2022), and aerobic heterotrophic bacteria, which utilize
organic compounds as carbon source and energy donor, and oxygen as
electron acceptor (Vethathirri et al., 2023). However, replication was
not used to investigate the variability in the product SCP in the absence
of a microbial inoculum. Further, the duration of a reactor run may
inuence the microbial composition of the community-based SCP grown
using wastewaters of variable chemical composition (Hülsen et al.,
2022). Here, a longer reactor operation of 92 days was chosen to select
for a wider range of taxa grown from food-processing wastewater of
variable chemical composition. An additional benet is that a longer
growth period may yield a more suitable microbial community-based
inoculum for applications focusing on enhancing SCP yield and pro-
duction rate.
While performance metrics are important for bioprocess optimiza-
tion, the SCP’s suitability for animal consumption, ensuring health
compatibility without adverse effects, remains to be investigated (Carter
and Codabaccus, 2022; Sharif et al., 2021). A feed-safety assessment of
microbial community-based SCP is needed for the targeted animal spe-
cies, with the aim of substituting a portion or the entirety of conven-
tional feed protein (Bratosin et al., 2021; Hadi and Brightwell, 2021).
The objectives of this study were to (1) characterize the microbial
community in inuent wastewater and the reactor biomass; (2) deter-
mine the growth rate and protein content in biomass across replicate
reactors; (3) characterize the essential amino acid content in the mi-
crobial protein produced; (4) assess the overall consistency of
community-based SCP by evaluating production rate and yield; and (5)
evaluate the potential health risks for sh associated with the
community-based SCP product when used as a substitute for aquafeeds.
This study demonstrates the consistent production of microbial
community-based SCP from food-processing wastewater, initiated
without a microbial inoculum and validated through a replicated reactor
design. It also evaluates feed safety through pathogen screening, high-
lighting its potential to replace traditional protein sources in animal feed
and support sustainable aquaculture practices.
2. Materials and methods
2.1. Experimental design
Four 4-L bioreactors were operated in sequencing batch mode on
continuous 12-h cycles with intermittent aeration, receiving wastewater
from a soybean processing company in Singapore. Over the course of
three months, twelve 20-L carboys of wastewater from soybean soaking
were collected on ten different occasions. The soybean wastewater
collection batches were numbered from 1 to 10. Batch 2 included three
carboys collected on the same day (2.1, 2.2, 2.3), while all other batches
consisted of a single carboy, each collected on a different day. All four
reactors were operated under identical conditions for 92 days to test
whether growing biomass directly from soybean-processing wastewater
without inoculum was consistent in the long term. Samples were
collected thrice a week on alternate days from each reactor to evaluate
the weekly variation of the parameters measured.
2.2. Operational parameters and bioreactor arrangement
The reactor temperature was maintained at 30 ◦C and sludge was
continuously mixed at 375 rpm. The feeding phase occurred during the
initial 5–10 min of a cycle, followed by alternating 180-min anoxic/
anaerobic and 540-min aerobic phases. Cycles nished once soluble
chemical oxygen demand (sCOD) of the mixed liquor was less than 400
mg/L (arbitrarily set value), after which the biomass was left to settle for
60 min at a minimum settling velocity of 0.09 m/h and then 1.35 L of
supernatant was discarded. Thereafter, the reactor was lled with the
same volume of soybean wastewater to start a new cycle. This feeding
regime resulted in the following average hydraulic residence times
(HRTs) for the four bioreactors operated: 7.2 d (F1), 7.2 d (F2), 7.2
d (F3) and 7.2 d (F4) (details in supplementary information). The pH
was between 6.0 and 8.5 and the DO level was controlled between 0.2
and 0.5 mg/L during the aerobic phase. Each of the sequencing batch
reactors (SBRs) employed in this study was equipped with a magnetic
stir plate to ensure mixed liquor homogeneity, a pair of EasySense pH
and DO probes with their corresponding transmitters (Mettler Toledo), a
dedicated air pump, a dedicated feed pump, a solenoid valve for su-
pernatant discharge, and a surrounding water jacket connected to a re-
circulating water heater. The different portions of the cycle were
controlled by a computer software specically designed for these re-
actors (VentureMerger, Singapore).
2.3. Analytical methods
Water quality parameters were measured in accordance with Stan-
dard Methods (APHA-AWWA-WEF, 2005) and targeted chemical oxygen
demand (COD) (Standard Methods 5220 D) and nitrogen species
(ammonium, nitrite, and nitrate ions) using spectrophotometric tests
(Hach Company, Loveland, Colorado, USA) and Ion Chromatography
(Standard Methods 4500-NH
3
for ammonium; 4110 B for nitrate and
nitrite). Total organic carbon (TOC) and total Kjeldahl nitrogen (TKN)
were also measured in inuent samples using a TOC-L analyzer (Shi-
madzu Corp., Kyoto, Japan). Total suspended solids (TSS) and volatile
suspended solids measurements were performed according to Standard
Methods (APHA-AWWA-WEF, 2005). Prior to chemical analysis,
efuent samples were ltered through a 0.2-
μ
m pore size lter and the
ltrate was stored at 4⁰C for less than one week.
2.4. Biomass protein analysis
The protein content of the biomass was determined by quantifying
amino acids using high-performance liquid chromatography (HPLC),
using a methodology adapted from Chen et al. (2019). Protein deter-
mination via HPLC yields higher accuracy than traditional protein
methods while also revealing which amino acids are present in the SCP
(Vethathirri et al., 2021) and is the preferred methodology to measure
protein content in food (FAO, 2003). First, 0.05 g of freeze-dried
biomass was mixed with 5 mL of 6 M HCl and ushed with nitrogen
gas for about 50 s. Then the samples were digested for 22 h using a
heating block at 110 ◦C followed by ltration with a 0.22-µm membrane
lter after cooling down. The solution was vacuum evaporated to dry-
ness at 38 ◦C and 25 Pa in a rotary evaporator and re-dissolved in a
volumetric ask with 5 mL of 0.1 M HCl. It was centrifuged for 40 min at
10,000 g and 4 ◦C. The supernatant was ltered using a 0.22-µm
membrane lter and the ltrate maintained at 4 ◦C until derivatization
(Chen et al., 2019). Pre-column derivatization was used with o-phtha-
laldehyde and 9-uorenylmethoxycarbonyl (FMOC
–
Cl) via an auto-
sampler (SIL −30 AC Autosampler). Seven and a half microliter of
sample or standard was mixed with 45 µL of mercaptopropionic acid and
22 µL of OPA for 1 min in online derivatization. After mixing 8 µL of
FMOC for 2 min, 5 µL of 0.1 M HCl was added. The reaction mixture was
injected into an HPLC (Prominence UFLC, Shimadzu, Japan) equipped
with a UV diode array detector (DAD, SPD-M20A), and detected at a
R.S. Vethathirri et al.
Cleaner and Circular Bioeconomy 10 (2025) 100139
2
wavelength of 338 nm for primary and 266 nm for secondary amino
acids after passage through a Shimadzu Shim-pack Scepter C18 column
(3 µm, 3.0 ×150 mm). The ow rate was adjusted to 0.8 mL/min and the
column temperature was set to 40 ◦C. Gradient programs were applied
for HPLC analysis. The average protein content in biomass (as a per-
centage of dry weight) of the various batches of soybean processing
wastewaters used on day 1 across the four reactors was estimated at 37.7
%, the average value measured on two samples of soybean processing
wastewaters (Figure S1). Hence, the initial microbial protein (d1) for
each reactor starting without inoculum was estimated as 37.7 % of its
initial biomass (TSS) value.
2.5. Microbial analysis
Each soybean wastewater feed batch was subsampled once for mi-
crobial analysis and sludge samples were collected thrice a week from
each reactor. A 50-mL feed sample was centrifuged at 10,000 rpm for 3
min and 45 mL of supernatant discarded resulting in a tenfold increase in
biomass concentration. Aliquots of 2 mL of concentrated wastewater and
2 mL of sludge samples were stored in cryogenic vials at −80 ◦C for DNA
extraction as previously described (Santillan et al., 2019b). Bacterial 16S
rRNA gene amplicon sequencing was done in two steps as described in
Santillan et al. (2020b). Primer set 341f/785r targeted the V3-V4 vari-
able regions of the 16S rRNA gene (Thijs et al., 2017). The libraries were
sequenced on an Illumina MiSeq platform (v.3) with 20 % PhiX spike-in
and at a read-length of 300 bp paired-end. Sequenced sample libraries
were processed following the DADA2 bioinformatics pipeline (Callahan
et al., 2016). DADA2 allows inference of exact amplicon sequence var-
iants (ASVs) providing several benets over traditional OTU clustering
methods (Callahan et al., 2017). Illumina sequencing adaptors and PCR
primers were trimmed prior to quality ltering. Sequences were trun-
cated after 280 and 255 nucleotides for forward and reverse reads,
respectively, a length at which the average quality dropped below a
Phred score of 20. After truncation, reads with expected error rates
higher than 3 and 5 for forward and reverse reads were removed. After
ltering, error rate learning, ASV inference and denoising, reads were
merged with a minimum overlap of 20 bp. Chimeric sequences were
identied and removed, with an average per sample of 0.77 %. For a
total of 92 samples, an average of 34,074 reads were retained per sample
after processing, constituting 68.8 % of the average input reads. Tax-
onomy was assigned using the SILVA database (v.138) (Gl¨
ockner et al.,
2017). The adequacy of sequencing depth after reads processing was
corroborated with rarefaction curves at the ASV level (Figure S2). Heat
maps for bacterial relative abundances of the SCP were constructed
using the DivComAnalyses package (v.0.9) in R (Constancias and
Sneha-Sundar, 2022).
2.6. Fluorescence in situ hybridization
Microbial characterization was further supported by uorescence in
situ hybridization (FISH) using probes for the domain bacteria and
selected core SCP taxa. Sludge samples were amended with 4 % para-
formaldehyde and placed on ice for 2 – 3 h The xed samples were
washed with 1 ×phosphate-buffered saline (PBS) solution and stored in
a mixture of 1 ×PBS and ethanol (1:1) at −20 ◦C until use. The cells
were allowed to dry on microscopic slides and dehydrated in an ethanol
series of 50, 80 and 96 % for 3 min each. Hybridization buffer (0.9 M
NaCl, 20 Mm Tris–HCl, 35 % formamide, 0.01 % sodium dodecyl sul-
fate) and probes were added to detect microorganisms of interest.
Eubmix (Eub338, Eub 338ll, Eub 338lll) targets most bacteria (Daims
et al., 1999) and AZOI 655 probe targets species belonging to Azospir-
illum (Stoffels et al., 2001). After hybridization, the slides were washed
with warm buffer for 10 min (0.9 M NaCl, 20 Mm Tris–HCl, 5 mM EDTA,
0.01 % sodium dodecyl sulfate) and rinsed thoroughly with cold water.
FISH images were acquired using a LSM780 confocal laser scanning
microscope. The Zen software was used for image processing and
graphical analysis (Carl Zeiss, Germany).
2.7. Phylogenetic feed-safety analysis
The ASV table (see supplementary le) from all SCP samples was
screened for suspected sh pathogens, by comparing the listed genera in
the table to known sh pathogens. The nucleotide sequences of the
suspected ASVs were aligned with Basic Local Alignment Search Tool
(Johnson et al., 2008) using the Nucleotide collection database and the
Megablast algorithm to obtain the reference taxa to construct the
phylogenetic trees. The same query sequences were aligned against
alternative families from the same order, using the Nucleotide collection
database and the BLAST algorithm to obtain the outgroups for the trees.
The evolutionary history was inferred by using the Maximum Likelihood
method and the predicted most likely model. The tree with the highest
log likelihood is shown. The percentage of trees in which the associated
taxa clustered together is shown next to the branches. Initial tree(s) for
the heuristic search were obtained automatically by applying
Neighbor-Join and BioNJ algorithms to a matrix of pairwise distances
estimated using the Maximum Composite Likelihood (MCL) approach,
and then selecting the topology with a superior log likelihood value. The
predicted most likely model distribution, and most likely rate variation
model, was used to model evolutionary rate differences among sites. All
positions with less than 90 % site coverage were eliminated, i.e., fewer
than 10 % alignment gaps, missing data, and ambiguous bases were
allowed at any one position (partial deletion option). Evolutionary an-
alyses were conducted in MEGA X (Kumar et al., 2018; Stecher et al.,
2020).
3. Results and discussion
3.1. Microbial composition of inuent wastewaters
Microbial characterization of the inuent food-processing waste-
waters is crucial, especially in microbial community-based SCP studies
that do not use a microbial inoculum. The different batches of soybean-
processing wastewaters used in this study (Table 1) were dominated by
the same two genera, Weissella and Lactococcus, at relative abundances
of 42 ±25.9 % and 34 ±16.6 %, respectively (Fig. 1). Both genera are
lactic acid bacteria of probiotic potential, and strains have been isolated
from the intestinal environment of sh species exhibiting antimicrobial
activity against pathogens (Mortezaei et al., 2020). Another genus of
lactic acid bacteria identied in all inuent soybean-processing waste-
waters was Lactobacillus (at a relative abundance of 16 ±13.6 %). The
most abundant ASV at the species level was Lactobacillus fermentum (up
to 15.5 % relative abundance). Lactobacillus fermentum strains are of
economic value due to their widespread application in biomedical and
food preservation elds (Naghmouchi et al., 2020).
Besides these three genera, Streptococcus was also found in all
wastewater batches at a relative abundance of up to 8 % (Fig. 1).
Although the genus Streptococcus includes several pathogens, species
such as Streptococcus thermophilus have “generally recognized as safe”
status and several health benets (Iyer et al., 2010). Other genera
including Leuconostoc, Floricoccus, Selenomonas, Veillonella and Mega-
sphaera occurred less frequently at a relative abundance greater than 2
% in at least one of the inuent wastewater batches (Fig. 1). Strains of
Leuconostoc are lactic acid producing bacteria of great value in dairy
technology and functional foods (Hemme and Foucaud-Scheunemann,
2004). Floricoccus penangensis has shown great probiotic potential in
terms of lactic acid production, antibacterial activities, auto-aggregation
and bacterial adhesion (Rungsirivanich et al., 2019). In a past study, it
was mentioned that Selenomonas ruminantium might have potential for
industrial lactic acid production (Shimizu et al., 1989). Veillonella
atypica has shown its ability in maintaining the health of the human gut
through the metabolic conversion of lactate into propionate (Scheiman
et al., 2019). Megasphaera elsdenii was reported to support lactate-driven
R.S. Vethathirri et al.
Cleaner and Circular Bioeconomy 10 (2025) 100139
3
dark fermentation in achieving stable biohydrogen production (Ohnishi
et al., 2022). In summary, the microbial characterization of inuent
soybean-processing wastewaters identied mostly lactic acid bacteria
that may be suitable for both SCP production and other applications.
3.2. Biomass growth and biomass protein content in replicate reactors
Growth in the four community-based SCP reactors was monitored as
total suspended solids (TSS) over 92 days (Fig. 2). The initial biomass
concentration was the same as the TSS of the rst batch of soybean-
processing wastewater (180 mg/L). Due to the low sTKN level of 38
mg/L in the inuent wastewater batch, there was not much biomass
growth in any of the replicate reactors from d1 to d5. Afterwards, due to
the shift in sTKN content to a higher range (69 – 161 mg/L), all repli-
cates showed an increase in biomass from d6 to d65. Following that, a
decline in cell growth from d66 to d76 was observed in all replicates due
to the lower sTKN level in the wastewater batch fed (41 mg/L). All re-
actors had either increased in biomass concentration or remained at the
same level whenever the sTKN level in the inuent wastewater fell
within 58 and 161 mg/L. Considering the entire bioconversion period,
an increase in biomass concentration of 5065 ±381 mg/L was achieved.
Overall, the replicate reactors showed similar patterns in biomass
growth on soybean-processing wastewaters.
The amount of protein in the biomass followed the pattern of TSS
production with either increasing or decreasing trends depending on the
wastewater batch used in each bioreactor (Fig. 2). A trend of increasing
Table 1
Chemical characteristics of soybean processing wastewaters used for microbial community-based SCP production.
Soybean wastewater
collection batch
a
sCOD
[mg/L]
sTKN
b
[mg/L]
sPO
4
3-
-P
[mg/L]
NH
4
+
-N
[mg/L]
NO
2
-
-N
[mg/L]
NO
3
-
-N
[mg/L]
TA
c
[mg/L]
TSS
[mg/L]
VSS
[mg/L]
pH sCOD:
sTKN (C:
N)
d
sTKN:
sPO
4
3-
-P (N:
P)
e
1 6872 38 56 4.89 0.00 0.06 0 180 180 4.02 181 0.7
2.1 15,301 161 120 9.29 0.00 0.13 253 450 420 4.79 95 1.3
2.2 9376 69 51 5.28 0.00 0.01 253 320 320 4.79 137 1.4
2.3 11,428 101 71 6.70 0.00 0.02 253 380 360 4.79 114 1.4
3 7382 72 48 5.67 0.00 0.02 165 420 300 4.64 103 1.5
4 7482 72 53 6.74 0.00 0.16 120 240 240 4.52 104 1.4
5 9676 82 75 4.83 0.00 0.20 33 300 300 4.10 118 1.1
6 8946 74 63 4.94 0.00 0.14 28 240 240 4.07 120 1.2
7 12,051 123 91 4.78 0.04 0.18 156 380 380 4.68 98 1.4
8 4397 41 15 3.78 0.31 0.02 197 160 160 4.90 107 2.7
9 6615 58 28 4.28 0.23 0.00 52 180 180 4.17 114 2.1
10 13,434 108 80 8.45 0.10 0.06 17 340 300 3.80 124 1.4
Average
f
(s.d.m.) 9413
(3020)
83 (34) 63 (27) 5.80
(1.61)
0.06
(0.10)
0.08
(0.07)
127
(95)
299
(95)
282
(80)
4.4
(0.4)
118 (22) 1.5 (0.5)
a
A total of twelve 20-L carboys was collected at ten different time points over three months. The soybean wastewater collection batches were numbered 1 to 10, with
Batch 2 consisting of three carboys (2.1, 2.2, 2.3), while all others consisted of a single carboy.
b
sTKN was calculated using the value of sTN, nitrate, and nitrite: sTKN =sTN – nitrate - nitrite.
c
Total alkalinity expressed as calcium carbonate.
d
Ratio of carbon and nitrogen in wastewater as calculated from chemical data.
e
Ratio of nitrogen and phosphorus in wastewater as calculated from chemical data.
f
Average values of inuent wastewater batches used, including standard deviation of the mean (s.d.m.) in parentheses.
Fig. 1. Microbial characterization of inuent soybean-processing wastewaters used for microbial community-based SCP production as assessed through 16S rRNA
gene amplicon sequencing. Shown are the top 20 genera of twelve batches of wastewater (20-L carboys), collected at ten different time points over three months. The
soybean wastewater collection batches were numbered from 1 to 10, with Batch 2 consisting of three carboys (2.1, 2.2, 2.3), while all others consisted of a sin-
gle carboy.
R.S. Vethathirri et al.
Cleaner and Circular Bioeconomy 10 (2025) 100139
4
biomass protein content was observed in reactors F1, F3 and F4 from
d44 to d68, and F2 from d44 to d68, due to the usage of wastewater
batches with lower sCOD:sTKN and higher sTKN levels (82 – 123 mg/L)
(Fig. 2). On the other hand, a decrease or no change in biomass protein
content was observed in reactors F1, F3 and F4 from d68 to d82, and in
F2 from d61 to d82 due to a higher sCOD:sTKN and lower sTKN levels in
wastewaters (41 – 58 mg/L) (Fig. 2). Higher protein percentages of 54.1
%, 58.9 %, 58.2 % and 47.9 % (expressed as TSS) were observed in
reactors F1 (on d44), F2 (on d54), F3 (on d47) and F4 (on d54),
respectively. These protein fractions are considered suitable to satisfy
the protein requirements of sh species (Azim et al., 2008; Webster and
Lim, 2002). The variable C:N (sCOD:sTKN) in the inuent wastewater
batches was the likely reason for the changeable protein fraction in all
reactors during the bioconversion period. Overall, microbial
community-based SCP production is inuenced by the C:N ratio in
wastewater, with a higher nitrogen content supporting greater protein
accumulation (Vethathirri et al., 2023). Wastewaters with a C:N ratio
between 10 and 20 are ideal for SCP production, as demonstrated by
various studies showing successful microbial growth and nutrient
removal in different wastewater types (Vethathirri et al., 2021).
3.3. Microbial community-based protein as a supplementary ingredient
for animal diet
The SCP produced from soybean-processing wastewaters in four
replicate reactors contained all essential amino acids (EAA) for several
animals, except for tryptophan, which was not measured. For instance,
EAA quantied in the SCP (Fig. 3) could partially replace the protein
source in the traditional diet of aquaculture animals (Millamena, 2002).
Further, the SCP produced could help meet the EAA requirements of
poultry animals such as chicken and turkey (Almquist, 1952). Microbial
biomass produced in all reactors had an EAA content of 41 ±0.2 %
protein on the nal day of the study. This was within the EAA range
obtained in another study, where the aerobic heterotrophic
bacteria-based SCP was produced using synthetic brewery-processing
wastewater (Papini et al., 2023). Among EAAs, leucine was the most
enriched in all reactors (8 ±0.2 % protein on d92). The leucine content
obtained was nearly double the estimated level required for maximum
growth of ngerling rainbow trout (Ahmad et al., 2021b). In a recent
study, a leucine supplementation in the diet of largemouth bass led to a
signicant increase in specic growth rate, weight gain, body protein
and total plasma protein (Shao et al., 2022). Apart from sh species,
leucine can also be employed in laying hen feed formulations, ac-
counting for half of the required quantity (Almquist, 1952). Next to
leucine, arginine was found to be abundant in all replicates (7 ±0.2 %
protein on d92), and this proportion of arginine was 1 % higher than the
desired requirement of chicken (Almquist, 1952). Poultry health in
terms of musculoskeletal development, fat accretion and antioxidant
system can be promoted to a greater extent using arginine supplemen-
tation (Castro and Kim, 2020). Arginine also plays an important role in
the metabolism of nutrients, stimulation of insulin release, and
enhancement of disease resistance for sh (Wang et al., 2021). Further,
valine, which was the third most prevalent EAA (6 ±0.3 % protein on
d92), is a branched-chain amino acid with widespread applications in
cosmetic, pharmaceutical, and agricultural sectors besides the animal
feed industry (Gao et al., 2021). The benets of valine in enhancing
growth and intestinal enzymatic capabilities in sh species such as
Fig. 2. Biomass growth and biomass protein content in four SCP-enriched
bioreactors. Biomass was measured as total suspended solids (TSS, open cir-
cles), and protein content in the biomass (closed triangles) via amino acid
quantication using HPLC, in four replicate reactors (F1–4) run for 92
d without inoculum.
Fig. 3. Essential amino acids needed for aquaculture animals in microbial community-based single cell protein (SCP) produced from soybean processing waste-
waters. Four replicate reactors were operated without inoculum: F1 (pattern-lled dark blue bar), F2 (pattern-lled light blue bar), F3 (pattern-lled dark green bar),
and F4 (pattern-lled light green bar). Data used correspond to samples collected on the nal day of the study, d92.
R.S. Vethathirri et al.
Cleaner and Circular Bioeconomy 10 (2025) 100139
5
rainbow trout have been established (Ahmad et al., 2021a). Comparing
all amino acids, the dry biomass produced in replicate reactors using
soybean-processing wastewaters was rich in aspartic acid, glutamic acid,
and alanine (Figure S3 and Figure S4). In conclusion, the SCP produced
by microbial communities in replicate reactors run without inoculum
could potentially be used as a protein ingredient in feed for a variety of
animals.
3.4. Enriched SCP-producing taxa in replicate reactors run without
inoculum
The most enriched SCP genus in all replicate reactors was Azospir-
illum at a relative abundance of 40.6 ±5.6 % (Fig. 4). FISH imaging
further conrmed the presence of Azospirillum in the SCP community of
all reactors (Fig. 5). Although it was below the limit of detection in the
different batches of inuent wastewaters characterized (Fig. 1), this
genus was enriched and became the most abundant SCP bacterial taxon
given the bioconversion conditions provided and available chemical
constituents. It is a plant-growth promoting nitrogen-xing bacterium
whose prevalence can be explained by the low oxygen and nitrogen
levels in the reactors (Lee et al., 2015). Certain species of Azospirillum
were reported to produce riboavin (vitamin B2), which can stimulate
plant growth (Lopez et al., 2019; Rodelas et al., 1993). For instance,
Azospirillum brasilense enhanced the physiological and growth parame-
ters of radish under stress conditions such as waterlogging
(Salazar-Garcia et al., 2022). The production of amino acids by this
species was also discussed as one of the reasons in the promotion of plant
growth (Gonz´
alez-L´
opez et al., 2005). Another species, Azospirillum
lipoferum (present in reactor F3 on days 5 and 10), was found to support
growth parameters in salt-stressed maize plants (Abdel Latef et al.,
2020). Hence, the enrichment of Azospirillum directly from
soybean-processing wastewater could be a way to produce effective
microbe-based fertilizers as proposed by Matassa et al. (2022). Such a
benecial bacterium can be cultivated in different geographical areas,
and further customization is possible based on soil and climatic condi-
tions (Ramos et al., 2023). Growing nitrogen-xing bacteria on real
food-processing wastewater could be a strategy in formulating
high-quality plant growth promoting biofertilizers to replace synthetic
nitrogen fertilizers. Similar temporal dynamics in microbial composition
were observed during long-term operation under uctuating nutrient
levels. These dynamics, shown through genus-level relative abundances
from samples collected every 4 – 5 days over the 92-day study period for
each reactor (Figure S5), highlight the stability and adaptability of the
microbial consortia over time despite nutrient uctuations in the feed
waste streams.
Like Azospirillum, the genera Propioniciclava and
Fig. 4. Microbial characterization based on 16S rRNA gene amplicon sequencing of community-based SCP biomass produced from soybean-processing wastewaters.
The 30 most abundant bacterial genera across the four replicate reactors (F1-F4) after 92 days of operation without inoculum are shown. Heat maps were generated
using 16S rRNA gene amplicon v3-v4 data; ASVs were grouped at the genus taxonomic level. Temporal dynamics of bacterial genera abundance throughout the study
are available as supplementary material (Figure S5).
R.S. Vethathirri et al.
Cleaner and Circular Bioeconomy 10 (2025) 100139
6
Acidipropionibacterium were below the limit of detection in the inuent
wastewaters but were enriched in all replicate reactors at a relative
abundance of 4.3 ±3.7 % and 3.0 ±2.1 %, respectively. Propioniciclava
was identied as one of the key contributors to the removal of organic
pollutants in paper-making wastewater efuents (Zhuang et al., 2020).
Acidipropionibacterium acidipropionici, which was identied at species
level in all replicate reactors, is the highest producer of propionic acid
among all known Propionibacteria and has the potential for efcient
feedstock utilization through multiple pathways (Ammar and Philip-
pidis, 2021; Parizzi et al., 2012). Weissella, one of the most abundant
bacterial genera in the inuent soybean-processing wastewaters (Fig. 1),
was not enriched in any of the reactors. On the other hand, Lactococcus
and Lactobacillus ourished in the SCP biomass of all replicates with a
relative abundance of 6.1 ±1.8 % and 3.8 ±2.2 %, respectively.
Although these genera were identied in the produced biomass, their
average relative abundances were reduced by about a factor of 5
compared to the inuent soybean-processing wastewaters. The available
nutrients in the inuent wastewater batches and the chosen process
conditions in reactors can explain an increase or decrease in relative
abundance for a specic genus (Santillan et al., 2020a, 2019b; Seshan
et al., 2023). Variations in nutrient levels, such as organic carbon and
nitrogen, inuence the growth of specic microbes, while reactor con-
ditions, including oxygen levels, temperature, and pH, also signicantly
impact microbial activity. For example, Azospirillum, a nitrogen-xing
bacterium, thrived under low oxygen and nitrogen conditions. Like-
wise, Propioniciclava and Acidipropionibacterium were enriched despite
not having been detected in the inuent wastewater, likely due to their
capacity to efciently utilize available substrates. These ndings un-
derscore how nutrient availability and reactor conditions can play a
more critical role in shaping microbial community composition during
SCP production than bacterial immigration from the inuent waste-
water (Vethathirri et al., 2023).
3.5. Need for replication in microbial community-based SCP production
The enrichment of core SCP-producing genera was similar (Fig. 4),
and comparable proles of EAAs were obtained in replicate reactors
through microbial community-based protein production from soybean-
processing wastewaters (Fig. 3). Towards the end of the study, the
biomass concentration was higher in reactor F1 compared to the other
three reactors (Fig. 2). The biomass production rate and biomass yield
were also higher (Table 2). This could be because of the presence of
other microbial taxa at low abundances in F1 that may have supported
the overall biomass growth. Since F1 did not have the highest SCP
production rate and SCP yield among its replicates (Table 2), higher
biomass yields and production rates do not necessarily mean higher
protein contents and production rates. This could have been due to a
higher C:N (sCOD:sTKN) in the inuent wastewater, which favored the
production of internal carbon storage such as glycogen or extracellular
polysaccharide rather than intracellular protein accumulation but still
resulted in an increase in TSS values (Durmaz and Sanin, 2001). Pre-
votella, the second most abundant genus in reactor F1 at a relative
abundance of 9.7 %, was below the detection limit in reactors F3 and F4
on the last day of the study (Fig. 4). Although all the replicate reactors
Fig. 5. FISH images from four replicate reactors (F1–4) operated without inoculum on d73 showing cells belonging to Azospirillum (orange) and the domain bacteria
(green). Cell abundances are consistent with 16S rRNA gene amplicon sequencing data for Azospirillum in F1, F2, F3 and F4 indicating relative abundances of 46.9 %,
49.5 %, 55.2 %, and 63.1 %, respectively. FISH images showing Azospirillum were also obtained from all reactors on d40 and d46 (Figure S6).
R.S. Vethathirri et al.
Cleaner and Circular Bioeconomy 10 (2025) 100139
7
received the same batch of soybean-processing wastewater, the micro-
bial community in the SCP in each of the reactors differed somewhat
(Fig. 4). This could be due to the non-constant C:N in the inuent
wastewater, which inhibited certain genera and favored others to grow
based on the resistance (the degree to which a community is insensitive
to a disturbance) and resilience (the capacity of such community to re-
turn to a pre-disturbance condition) of microbial communities to such
disturbances (Santillan et al., 2019a, 2020b, 2021; Santillan and
Wuertz, 2022). Clearly, the use of replicate reactors is essential to un-
derstanding the degree of variability in community-based SCP
production.
3.6. Potentially hygienically relevant taxa in community-based SCP
Putative pathogen-like sequences were present at very low relative
abundances up to a total of 0.009 % in the microbial community-based
SCP produced using soybean-processing wastewater in four replicate
reactors for sh diet applications (Table 3). At the genus level, six po-
tential pathogens were detected in all the replicate SCP reactors, namely
Flavobacterium, Lactococcus, Mycolicibacter (formerly Mycobacterium),
Mycobacterium, Pseudomonas, and Streptococcus. These genera accounted
for 0.64 % of all the taxa detected, with the Flavobacterium genus being
the most abundant at about 0.2 % (Table 3). As these taxa are below the
1 % relative abundance threshold, they are classied as rare organisms.
The ASVs of the 16S rRNA gene sequences underwent phylogenetic
analysis to obtain the most probable species related to the queried
sequences.
Members of the Flavobacterium genus are associated with three major
bacterial sh diseases, which are columnaris disease caused by Fla-
vobacterium columnare, bacterial gill disease cause by Flavobacterium
branchiophilum, and bacterial cold-water disease caused by Fla-
vobacterium psychrophilum (Loch and Faisal, 2015). Other less commonly
reported Flavobacterium sh pathogens include Flavobacterium johnso-
niae, Flavobacterium hydatis, Flavobacterium succinicans, Flavobacterium
chilense, Flavobacterium araucananum, Flavobacterium oncorhynchi, Fla-
vobacterium plurextorum, Flavobacterium tructae, Flavobacterium piscis,
Flavobacterium collinsii, Flavobacterium branchiarum, and Flavobacterium
branchiicola, many of which are co-infecting with Flavobacterium psy-
chrophilum (Loch and Faisal, 2015). None of the above mentioned Fla-
vobacterium species were detected; however, Flavobacterium
lindanitolerans (ASV 90, Figure S8), detected at 0.2 % relative abundance
(Table 3), was found to be associated with a case of fatal pulmonary
edema and hemorrhage in a patient (Tian et al., 2011), while Fla-
vobacterium ummariense (ASV 690, Figure S8) detected at 0.004 %
relative abundance (Table 3) had no reported case of pathogenicity.
Since no known Flavobacterium sh pathogens were detected, and the
remaining detected ASVs are of very low relative abundance, it can be
concluded that the consumption of SCP by aquaculture sh species
would present a minimal risk for Flavobacterium infection.
The species of Lactococcus involved in lactococcosis in sh is Lacto-
coccus garvieae (Meyburgh et al., 2017), with Lactococcus petaurid
recently discovered in lactococcosis in Nile tilapia (Egger et al., 2023).
In all the SCP reactors, Lactococcus constituted about 5.9 % of all the
detected taxa with Lactococcus lactis being the most abundant at roughly
5.8 %, and the remaining species being Lactococcus hircilactis and Lac-
tococcus protaetiae (Table 3) (Figure S9). Lactococcus lactis is a prevalent
prebiotic organism used in cultivated shes (Cano-Lozano et al., 2022;
Moroni et al., 2021; Tan et al., 2022; Zhu et al., 2021), and Lactococcus
hircilactis and Lactococcus protaetiae are not associated with any known
sh diseases. Since the pathogenic species of Lactococcus are absent in
the reactors, the consumption of SCP by sh would not pose a risk of
lactococcosis.
Many members of the genus Mycobacterium are responsible for
mycobacteriosis in shes, mostly known to be associated with Myco-
bacterium marinum, Mycobacterium fortui- tum, and Mycobacterium che-
lonae (Gauthier and Rhodes, 2009). Two Mycobacterium species were
detected at very low relative abundances; they are Mycobacterium mar-
inum (ASV 1188, Figure S10) at 0.001 % and Mycolicibacter sp. (formerly
Mycobacterium) (ASV 985 & 1010, Figure S10) at 0.002 % (Table 3). The
SCP products are processed through spray-drying after harvesting from
the reactors, before being added into the aquaculture feed preparation.
Some species of Mycobacterium are able to be preserved in a
cryo-desiccated state, which include Mycobacterium avium, Mycobacte-
rium phlei, Mycobacterium aquae, Mycobacterium microti, Mycobacterium
fortuitum, and Mycobacterium smegmatis, but Mycobacterium marinum
was not reported in the published literature (ˇ
Slos´
arek et al., 1976).
Therefore, there is a potential for viable Mycobacterium marinum to be
Table 2
Microbial community-based SCP production from soybean-processing wastewaters without inoculum in replicate reactors.
Reactor
a
sTKN
b
[mg/L]
C: N
b
[sCOD:
sTKN]
Biomass
production rate
[g TSS/L
R
/d]
c
Biomass
yield_sCOD [g
TSS/g sCOD]
Biomass
yield_sTKN [g
TSS/g sTKN]
Protein
production rate
[g protein/L
R
/d]
Protein
yield_sCOD [g
protein/g sCOD]
Protein
yield_sTKN [g
protein/g sTKN]
R
sCODd
[%]
R
sTNd
[%]
F1 83 (34) 118 (22) 0.77 0.153 20.55 0.25 0.051 6.72 93 78
F2 83 (34) 118 (22) 0.68 0.140 17.49 0.32 0.062 7.88 91 82
F3 83 (34) 118 (22) 0.65 0.131 16.31 0.30 0.048 6.10 93 83
F4 83 (34) 118 (22) 0.63 0.130 16.59 0.25 0.059 7.44 91 80
a
Reactors F1 – 4 run without a microbial inoculum for 92 d under identical conditions.
b
Average values of inuent soybean-processing wastewater batches used, including standard deviation of the mean (s.d.m.) in parentheses.
c
L
R
=reactor volume in liters. All reactors had a working volume of 4 L.
d
Average nutrient removal efciency of carbon (R
sCOD
) and nitrogen (R
sTN
) calculated for each reactor based on feed and efuent characteristics (Figure S7).
Table 3
Potential pathogens at the species level in combined microbial community-
based SCP product from four replicate reactors and their sh pathogenicity.
Most probable
potential pathogen
a
Amplicon
sequence variant
number
Percent of total
16S rRNA read
[%]
Fish pathogen
[Yes/No]
b
Flavobacterium
lindanitolerans
90 0.2014 No
Flavobacterium
ummariense
690 0.0037 No
Lactococcus hircilactis 1408 0.0898 No
Lactococcus protaetiae 1398 0.0003 No
Mycolicibacter sp. 985 and 1010 0.0021 No
Mycobacterium
marinum
1188 0.0011 Yes
Pseudomonas
nicosulfuronedens
1308 0.1991 No
Pseudomonas
sichuanensis
325 0.0316 No
Pseudomonas
alcaligenes
625 0.0075 Yes
Pseudomonas juntendi 1173 0.0008 No
Pseudomonas sediminis 1178 0.0006 No
Streptococcus urinalis 52 and 131 0.0989 No
a
Species identied using phylogenetic analysis of corresponding amplicon
sequence variants from 16 s rRNA metabarcoding of the SCP reactors
(Figure S8–12).
b
Analysis provided under Section 3.6 to identify potential pathogenic and
non-pathogenic SCP-producing species for sh feed applications.
R.S. Vethathirri et al.
Cleaner and Circular Bioeconomy 10 (2025) 100139
8
present in the nished SCP. When pelleting sh feed, the conditioning
step in the process involved the introduction of steam, which raises the
temperature of the feed mix to 80 – 90 ◦C with the temperature rising to
92 ◦C during extrusion (Fisheries, 2008). While Mycobacterium marinum
was identied from ASV 1188 (Figure S10) through phylogenetic anal-
ysis, the sequence queried was from the 16S rRNA gene sequence, which
is inadequate to effectively identify the species of Mycobacterium.
A variety of Pseudomonas species are involved in pseudomonadiasis
in shes, which include Pseudomonas aeruginosa, Pseudomonas alcali-
genes, Pseudomonas baetica, Pseudomonas chlororaphis, Pseudomonas u-
orescens, Pseudomonas koreensis, Pseudomonas luteola, Pseudomonas
mosselii, Pseudomonas plecoglossicida, Pseudomonas pseudoalcaligenes,
Pseudomonas putida, and most prominently Pseudomonas anguilliseptica
(Austin and Austin, 2016). All reported species were absent in the re-
actors, except for Pseudomonas alcaligenes (ASV 625, Figure S11) that
was detected at 0.008 % relative abundance (Table 3). It has been re-
ported that an infectious dose of 2 ×10
6
CFU/sh of Pseudomonas
alcaligenes in healthy hybrid sturgeons resulted in 33 % mortality 15
days post infection (Xu et al., 2015). Therefore, the ultra-low detection
of the putative pathogen-like sequences, coupled with the combination
of SCP drying and food pelleting, would make the threat of contracting a
sh disease linked to Pseudomonas negligible.
“Warm water” streptococcosis is known to involve Lactococcus gar-
vieae, Streptococcus iniae, Streptococcus agalactiae and Streptococcus par-
auberis, which cause sh mortalities in water temperatures above 15 ◦C.
Streptococcus iniae is a highly infectious streptococcosis agent, which had
a reported LD
50
(lethal dose at 50 % mortality) of 1 ×10
3
in the oral
cavity of barramundi Lates calcarifer, and 2.0 – 3.2 ×10
4
CFU/ml in
freshwater or salt water (Bromage and Owens, 2002). Additionally,
Streptococcus iniae has been shown to infect humans handling infected
shes (Goh et al., 1998), with reported cases of bacterial cellulitis,
endocarditis, meningitis, and arthritis (Lau et al., 2003). All reported
pathogens were absent in all reactors, with the only Streptococcus
detected being most related to Streptococcus urinalis (ASV 52 & 131,
Figure S12) at 0.1 % relative abundance (Table 3). Streptococcus urinalis
has not been reported to be pathogenic to sh, and the only known case
was a urinary tract infection in a 60-year-old man
(Peltroche-Llacsahuanga et al., 2012). As such, the consumption of SCP
should not present a threat of streptococcosis.
3.7. Future work and challenges in SCP production from food-processing
wastewaters
The current research involved four replicate reactors, which were
run without a microbial inoculum to produce SCP from soybean-
processing wastewaters. An average of 92 % of sCOD and 81 % of sTN
removal was observed (Table 2). After 92 d of reactor operation using
several batches of inuent wastewaters with variable C:N, sufcient
biomass was produced. The average biomass yield was 0.1 – 0.2 g TSS/g
sCOD, which was lower than theoretical considerations (0.4–0.5 g TSS/
g sCOD) for produced microbial biomass at shorter sludge ages (Pikaar
et al., 2022); this may be attributed to the absence of a microbial
community-based inoculum. Taxa in the inoculum would have an
advantage over the initial wastewater microbial community in terms of
adaptation to the bioreactor environment (Vethathirri et al., 2023). This
observation is similar to the previous study by Vethathirri et al. (2023),
which reported a maximum biomass yield of 0.089 g TSS/g sCOD
feed
.
These yield values of aerobic heterotrophic bacteria were signicantly
lower than the yield of photosynthetic nonsulfur bacteria (ten times
higher) produced in a volatile fatty acid-based medium with a controlled
COD:TN ratio of 17 (Peng et al., 2022). The highest biomass production
rate occurred in reactor F1 (0.77 g TSS/LR/d) and was about one order
of magnitude lower than the rate reported for aerobic heterotrophic
bacteria using synthetic brewery-processing wastewater (Papini et al.,
2023). Vethathirri et al. (2023) attributed their lower yield to the use of
real food-processing wastewaters with varying C:N ratios, which were
higher than the recommended C:N ratios (10 to 20) for microbial
community-based single-cell protein (SCP) production, particularly due
to the high sCOD:sTKN values in the soybean wastewaters. Optimizing
parameters such as solids retention time, HRT, cycle time,
aerobic-to-anoxic processing ratio, mixing rate, temperature, pH, and C:
N ratios, as discussed in the review by Vethathirri et al. (2021), could be
critical to improving SCP productivity in both studies by promoting the
enrichment of core microbes and enhancing the microbial protein con-
tent in community-based SCP production systems. Due to the lack of
research on the production of microbial community-based SCP from real
food-processing wastewater without inoculum using a replicated system
and a long-term approach, a true comparison of the yield and production
rate with previously published studies is challenging.
Although in the current study the same genus was dominant in
replicate reactors, there could be different community dynamics due to
stochastic effects in identically operated parallel reactors in the presence
of a microbial inoculum (Santillan et al., 2019a, 2020a; Wittebolle et al.,
2009). Hence, additional studies are needed to assess the performance
metrics with a well-designed inoculum in combination with a revised
feeding regime using replicate reactors. While 16S rRNA sequencing,
ASV identications, and phylogenetic analysis provide valuable insights
into microbial composition and potential pathogens, we acknowledge
that other sequencing methods, such as metagenomic sequencing fol-
lowed by metagenome-assembled genome (MAG) construction, could
offer a more comprehensive assessment of pathogen presence and
strain-level identication (Neshat et al., 2024).
Further, it was observed that during a 12 h cycle, most of the sCOD
and sTN were removed in the aerobic portion compared to the anoxic
phase (Figure S13). Hence, a reactor cycle with no anoxic phase could be
explored for faster conversion of carbon and nitrogen into SCP. In that
case, Azospirillum, the most abundant genus in all replicate reactors
would likely be outcompeted. Additionally, lactic acid bacteria would
not be enriched at their maximum specic growth rates in the absence of
an anaerobic phase (Rombouts et al., 2020). The current study employed
aerobic and anoxic stages at a ratio of 3:1 to favor the growth of both
aerobic and anaerobic microorganisms using soybean-processing
wastewaters in SBRs. Identifying the appropriate aerobic:anoxic phase
ratio to enrich and maintain the core SCP-producing bacterial taxa is a
key objective for future work.
The importance of ribonucleic acids (RNA) in feed has been reported
in the literature, and while bacterial SCP are known to contain high
levels of nucleic acids (8–12 %), particularly RNA (Ritala et al., 2017),
this may not pose a signicant issue for sh feeds. Studies by Li & Gatlin.
(2006) and Rairat et al. (2022) suggest that nucleic acids in sh diets can
provide benets, such as enhancing immune responses and offering
protection against infections. Although RNA levels were not measured in
this study, incorporating RNA content analysis in future research could
be useful for understanding its potential impacts on feed quality and
animal health.
While the present study showed microbial community-based SCP to
be safe for use in sh feed, further investigation is warranted to assess its
suitability for other animal species. The aim of this study was to
demonstrate the consistent production of microbial community-based
SCP from food-processing wastewater, starting without a microbial
inoculum, and to assess its safety for animal feed through pathogen
screening. Future research efforts could incorporate comparative ana-
lyses with conventional protein sources to evaluate the economic
feasibility and sustainability of SCP in aquaculture systems, offering a
clearer understanding of its potential to enhance aquaculture
sustainability.
4. Conclusions
This study establishes the long-term consistency and feed safety of
microbial community-based SCP production from soybean-processing
wastewater, achieved without a microbial inoculum and validated
R.S. Vethathirri et al.
Cleaner and Circular Bioeconomy 10 (2025) 100139
9
through a replicated reactor setup. Low aeration levels (0.2–0.5 mg/L)
and nitrogen content (38–161 mg/L) in wastewater batches led to a high
relative abundance of Azospirillum (40.6 ±5.1 %) in the SCP, a genus
known for its ability to perform nitrogen xation and promote plant-
growth, despite the dominance of Lactococcus (34 ±16.6 %) and Weis-
sella (42 ±25.9 %) in the wastewater feed. While high biomass yields
were achieved, they did not always correlate with high protein yields,
likely due to the accumulation of other intracellular components.
Pathogen screening indicated minimal sh pathogen-like sequences,
supporting the safety of SCP for animal feed. Further investigation is
needed to evaluate performance parameters using a microbial
community-based inoculum, optimized feeding strategies, and replicate
reactor systems to enhance robustness and reliability. These ndings
highlight the potential of microbial community-based SCP as a sus-
tainable alternative to conventional protein sources, offering a scalable
and environmentally friendly solution for aquaculture feed production.
CRediT authorship contribution statement
Ramanujam Srinivasan Vethathirri: Writing – review & editing,
Writing – original draft, Visualization, Methodology, Investigation,
Formal analysis, Data curation, Conceptualization. Ezequiel Santillan:
Writing – review & editing, Writing – original draft, Visualization, Su-
pervision, Project administration, Methodology, Formal analysis, Data
curation, Conceptualization. Yissue Woo: Writing – review & editing,
Visualization, Methodology, Formal analysis. Sara Swa Thi: Writing –
review & editing, Methodology. Hui Yi Hoon: Writing – review &
editing, Methodology. Stefan Wuertz: Writing – review & editing,
Writing – original draft, Supervision, Funding acquisition,
Conceptualization.
Declaration of competing interest
The authors declare that they have no known competing nancial
interests or personal relationships that could have appeared to inuence
the work reported in this paper.
Acknowledgements
This research was supported by the Singapore National Research
Foundation (NRF) and Ministry of Education under the Research Centre
of Excellence Program (EDUN C33-62-036-V4), and the NRF Competi-
tive Research Programme (NRF-CRP21-2018-0006) "Recovery and mi-
crobial synthesis of high-value aquaculture feed additives from food-
processing wastewater". We thank LCW Liew for support with waste-
water collection.
Supplementary materials
Supplementary material associated with this article can be found, in
the online version, at doi:10.1016/j.clcb.2025.100139.
Data availability
DNA sequencing data are available at NCBI BioProjects
PRJNA1074121. See supplementary information for details about mi-
crobial protein yield and production, nutrient removal efciency and
HRT estimation, chemical characteristics of bioreactor efuent and
inuent, rarefaction plots for 16S rRNA gene sequencing data, biomass
protein content (as % dry weight), temporal amino acid proles in re-
actors, amino acid prole in dry biomass of wastewater, and temporal
dynamics of the 20 most abundant genera in each reactor. Supplemen-
tary ASV-level data are also provided to identify particular species
present in the reactors and feed.
References
Abdel Latef, A.A.H., Abu Alhmad, M.F., Kordrostami, M., Abo–Baker, A.-B.A.-E., Zakir, A,
2020. Inoculation with azospirillum lipoferum or azotobacter chroococcum reinforces
maize growth by improving physiological activities under saline conditions. J. Plant
Growth Regul. 39 (3), 1293–1306.
Ahmad, I., Ahmed, I., Dar, N.A., 2021a. Dietary valine improved growth, immunity,
enzymatic activities and expression of TOR signaling cascade genes in rainbow trout,
oncorhynchus mykiss ngerlings. Sci. Rep. 11 (1), 22089.
Ahmad, I., Ahmed, I., Dar, N.A., 2021b. Effects of dietary leucine levels on growth
performance, hematobiochemical parameters, liver prole, intestinal enzyme
activities and target of rapamycin signalling pathway related gene expression in
rainbow trout, Oncorhynchus mykiss ngerlings. Aquac. Nutr. 27 (6), 1837–1852.
Almquist, H., 1952. Amino acid requirements of chickens and turkeys—a review. Poult.
Sci. 31 (6), 966–981.
Ammar, E.M., Philippidis, G.P., 2021. Fermentative production of propionic acid:
prospects and limitations of microorganisms and substrates. Appl. Microbiol.
Biotechnol. 105 (16–17), 6199–6213.
APHA-AWWA-WEF. 2005. Standard methods for the examination of water and
wastewater, 21, 258–259.
Asiri, F., Chu, K.-H., 2022. Valorization of agro-industrial wastes into
polyhydroxyalkanoates-rich single-cell proteins to enable a circular waste-to-feed
economy. Chemosphere 309, 136660.
Asiri, F., Kim, J., Chu, K.-H., 2022. Chitosan-harvested polyhydroxybutyrate-rich
Zobellella denitricans ZD1 as a multifunction feed for aquaculture. Aquaculture 561,
738655.
Asiri, F., 2024. Polyhydroxyalkanoates for sustainable aquaculture: a review of recent
advancements, challenges, and future directions. J. Agric. Food Chem. 72 (4),
2034–2058.
Austin, B., Austin, D.A., 2016. Bacterial Fish Pathogens. Springer International
Publishing.
Azim, M.E., Little, D.C., Bron, J.E., 2008. Microbial protein production in activated
suspension tanks manipulating C:n ratio in feed and the implications for sh culture.
Bioresour. Technol. 99 (9), 3590–3599.
Bratosin, B.C., Darjan, S., Vodnar, D.C., 2021. Single cell protein: a potential substitute in
Human and animal nutrition. Sustainability 13 (16), 9284.
Bromage, E., Owens, L., 2002. Infection of barramundi lates calcarifer with Streptococcus
iniae: effects of different routes of exposure. Dis. Aquat. Org. 52, 199–205.
Callahan, B.J., McMurdie, P.J., Holmes, S.P., 2017. Exact sequence variants should
replace operational taxonomic units in marker-gene data analysis. ISME J 11 (12),
2639–2643.
Callahan, B.J., McMurdie, P.J., Rosen, M.J., Han, A.W., Johnson, A.J.A., Holmes, S.P.,
2016. DADA2: high-resolution sample inference from Illumina amplicon data. Nat.
Methods 13 (7), 581–583.
Cano-Lozano, J.A., Villamil Diaz, L.M., Melo Bolivar, J.F., Hume, M.E., Ruiz Pardo, R.Y.,
2022. Probiotics in tilapia (Oreochromis niloticus) culture: potential probiotic
lactococcus lactis culture conditions. J. Biosci. Bioeng. 133 (3), 187–194.
Castro, F.L.D.S., Kim, W.K., 2020. Secondary functions of arginine and sulfur amino acids
in poultry health: review. Animals 10 (11), 2106.
Carter, C.G., Codabaccus, M.B., 2022. Assessing the value of single-cell ingredients in
aquafeeds. Curr. Opin. Biotechnol. 76, 102734.
Chen, S., Dong, B., Dai, X., Wang, H., Li, N., Yang, D., 2019. Effects of thermal hydrolysis
on the metabolism of amino acids in sewage sludge in anaerobic digestion. Waste
Manage. (Oxford) 88, 309–318.
Constancias, F., Sneha-Sundar, 2022. fconstancias/DivComAnalyses (v0.9). Zenodo,
6473394.
Daims, H., Brühl, A., Amann, R., Schleifer, K.H., Wagner, M., 1999. The domain-specic
probe EUB338 is insufcient for the detection of all bacteria: development and
evaluation of a more comprehensive probe set. Syst. Appl. Microbiol. 22 (3),
434–444.
Durkin, A., Finnigan, T., Johnson, R., Kazer, J., Yu, J., Stuckey, D., Guo, M., 2022. Can
closed-loop microbial protein provide sustainable protein security against the
hunger pandemic? Curr. Res. Biotechnol. 4, 365–376.
Durmaz, B., Sanin, F.D., 2001. Effect of carbon to nitrogen ratio on the composition of
microbial extracellular polymers in activated sludge. Water Sci. Technol. 44 (10),
221–229.
Egger, R.C., Rosa, J.C.C., Resende, L.F.L., de P´
adua, S.B., de Oliveira Barbosa, F.,
Zerbini, M.T., Tavares, G.C., Figueiredo, H.C.P., 2023. Emerging sh pathogens
Lactococcus petauri and L. garvieae in Nile tilapia (Oreochromis niloticus) farmed in
Brazil. Aquaculture 565, 739093.
El Abbadi, S.H., Criddle, C.S., 2019. Engineering the dark food chain. Environ. Sci.
Technol. 53 (5), 2273–2287.
FAO, 2003. Food Energy - Methods of Analysis and Conversion Factors.
FAO, 2021. The state of food security and nutrition in the world 2021, ISBN 978-92-5-
134325-8.
Fisheries, F.A.O, 2008. ADCP/REP/80/11-Fish Feed Technology. Feed Milling Processes,
Rome, Italy. Chapter 18.
Gao, H., Tuyishime, P., Zhang, X., Yang, T., Xu, M., Rao, Z., 2021. Engineering of
microbial cells for L-valine production: challenges and opportunities. Microb. Cell
Fact. 20 (1).
Gauthier, D.T., Rhodes, M.W., 2009. Mycobacteriosis in shes: a review. Vet. J. 180 (1),
33–47.
Gl¨
ockner, F.O., Yilmaz, P., Quast, C., Gerken, J., Beccati, A., Ciuprina, A., Ludwig, W.,
2017. 25 years of serving the community with ribosomal RNA gene reference
databases and tools. J. Biotechnol. 261, 169–176.
R.S. Vethathirri et al.
Cleaner and Circular Bioeconomy 10 (2025) 100139
10
Godfray, H.C.J., Beddington, J.R., Crute, I.R., Haddad, L., Lawrence, D., Muir, J.F.,
Toulmin, C., 2010. Food security: the challenge of feeding 9 billion people. Science
327 (5967), 812–818.
Goh, S.H., Driedger, D., Gillett, S., Low, D.E., Hemmingsen, S.M., Amos, M., Chan, D.,
Lovgren, M., Willey, B.M., Shaw, C., Smith, J.A., 1998. Streptococcus iniae, a human
and animal pathogen: specic identication by the chaperonin 60 gene
identication method. J. Clin. Microbiol. 36 (7), 2164–2166.
Gonz´
alez-L´
opez, J., Rodelas, B., Pozo, C., Salmer´
on-L´
opez, V., Martínez-Toledo, M.,
Salmer´
on, V., 2005. Liberation of amino acids by heterotrophic nitrogen xing
bacteria. Amino Acids 28 (4), 363–367.
Hadi, J., Brightwell, G., 2021. Safety of alternative proteins: technological,
environmental and regulatory aspects of cultured meat, plant-based meat, insect
protein and single-cell protein. Foods 10 (6), 1226.
Hemme, D., Foucaud-Scheunemann, C., 2004. Leuconostoc, characteristics, use in dairy
technology and prospects in functional foods. Int. Dairy J. 14 (6), 467–494.
Hülsen, T., Hsieh, K., Lu, Y., Tait, S., Batstone, D.J., 2018a. Simultaneous treatment and
single cell protein production from agri-industrial wastewaters using purple
phototrophic bacteria or microalgae - a comparison. Bioresour. Technol. 254,
214–223.
Hülsen, T., Hsieh, K., Tait, S., Barry, E.M., Puyol, D., Batstone, D.J., 2018b. White and
infrared light continuous photobioreactors for resource recovery from poultry
processing wastewater - A comparison. Water Res. 144, 665–676.
Hülsen, T., Sander, E.M., Jensen, P.D., Batstone, D.J., 2020. Application of purple
phototrophic bacteria in a biolm photobioreactor for single cell protein production:
biolm vs suspended growth. Water Res. 181, 115909.
Hülsen, T., Stegman, S., Batstone, D.J., Capson-Tojo, G., 2022. Naturally illuminated
photobioreactors for resource recovery from piggery and chicken-processing
wastewaters utilising purple phototrophic bacteria. Water Res., 118194
Iyer, R., Tomar, S.K., Maheswari, T.U., Singh, R., 2010. Streptococcus thermophilus strains:
Multifunctional lactic acid bacteria. Int. Dairy J. 20 (3), 133–141.
Johnson, M., Zaretskaya, I., Raytselis, Y., Merezhuk, Y., McGinnis, S., Madden, T.L.,
2008. NCBI BLAST: a better web interface. Nucleic Acids Res. 36, W5–W9 (Web
Server).
Jones, S.W., Karpol, A., Friedman, S., Maru, B.T., Tracy, B.P., 2020. Recent advances in
single cell protein use as a feed ingredient in aquaculture. Curr. Opin. Biotechnol. 61,
189–197.
Joris, P., Lombard, E., Paillet, A., Navarro, G., Guillouet, S.E., Gorret, N., 2024. Recycling
potential of Cupriavidus necator for life support in space: production of SCPs from
volatile fatty acid and urea mixture. J. Biotechnol. 396, 18–27.
Kumar, S., Stecher, G., Li, M., Knyaz, C., Tamura, K., 2018. MEGA X: molecular
evolutionary genetics analysis across computing platforms. Mol. Biol. Evol. 35 (6),
1547–1549.
Lau, S.K.P., Woo, P.C.Y., Tse, H., Leung, K.-W., Wong, S.S.Y., Yuen, K.-Y., 2003. Invasive
Streptococcus iniae infections outside north america. J. Clin. Microbiol. 41 (3),
1004–1009.
Lee, J.Z., Logan, A., Terry, S., Spear, J.R., 2015. Microbial response to single-cell protein
production and brewery wastewater treatment. Microb Biotechnol 8 (1), 65–76.
Li, P., Gatlin, D.M., 2006. Nucleotide nutrition in sh: current knowledge and future
applications. Aquaculture 251 (2–4), 141–152.
Loch, T.P., Faisal, M., 2015. Emerging avobacterial infections in sh: a review. J. Adv.
Res. 6 (3), 283–300.
Lopez, B.R., Palacios, O.A., Bashan, Y., Hern´
andez-Sandoval, F.E., de-Bashan, L.E., 2019.
Riboavin and lumichrome exuded by the bacterium Azospirillum brasilense promote
growth and changes in metabolites in Chlorella sorokiniana under autotrophic
conditions. Algal Res. 44.
Matassa, S., Boeckx, P., Boere, J., Erisman, J.W., Guo, M., Manzo, R., Meerburg, F.,
Papirio, S., Pikaar, I., Rabaey, K., Rousseau, D., Schnoor, J., Smith, P., Smolders, E.,
Wuertz, S., Verstraete, W., 2022. How can we possibly resolve the planet’s nitrogen
dilemma? Microb. Biotechnol. 1–13.
Meyburgh, C.M., Bragg, R.R., Boucher, C.E., 2017. Lactococcus garvieae: an emerging
bacterial pathogen of sh. Dis. Aquat. Org. 123 (1), 67–79.
Millamena, O.M., 2002. The essential nutrients: proteins and amino acids. In: nutrition in
Tropical Aquaculture: essentials of sh nutrition, feeds, and feeding of tropical aquatic
species. Aquaculture Department, Southeast Asian Fisheries Development Center
7–20.
Moroni, F., Naya-Catal`
a, F., Piazzon, M.C., Rimoldi, S., Calduch-Giner, J., Giardini, A.,
Martínez, I., Brambilla, F., P´
erez-S´
anchez, J., Terova, G., 2021. The effects of nisin-
producing lactococcus lactis strain used as probiotic on gilthead sea bream (Sparus
aurata) rrowth, gut microbiota, and transcriptional response. Front. Mar. Sci. 8,
659519.
Mortezaei, F., Royan, M., Allaf Noveirian, H., Babakhani, A., Alaie Kordghashlaghi, H.,
Balcazar, J.L., 2020. In vitro assessment of potential probiotic characteristics of
indigenous lactococcus lactis and weissella oryzae isolates from rainbow trout
(Oncorhynchus mykiss Walbaum). J. Appl. Microbiol. 129 (4), 1004–1019.
Naghmouchi, K., Belguesmia, Y., Bendali, F., Spano, G., Seal, B.S., Drider, D., 2020.
Lactobacillus fermentum: a bacterial species with potential for food preservation and
biomedical applications. Crit. Rev. Food Sci. Nutr. 60 (20), 3387–3399.
Nayyar, D., Nawaz, T., Noore, S., Singh, A.P., 2021. Food processing wastewater
treatment: current practices and future challenges. In: Singh, S.P., Rathinam, K.,
Gupta, T., Agarwal, A.K. (Eds.), Pollution Control Technologies: Current Status and
Future Prospects. Springer Singapore, Singapore, pp. 177–208.
Neshat, S.A., Santillan, E., Seshan, H., Wuertz, S., 2024. Non-redundant metagenome-
assembled genomes of activated sludge reactors at different disturbances and scales.
Sci. Data 11 (1), 855.
Ohnishi, A., Hasegawa, Y., Fujimoto, N., Suzuki, M., 2022. Biohydrogen production by
mixed culture of megasphaera elsdenii with lactic acid bacteria as lactate-driven dark
fermentation. Bioresour. Technol. 343, 126076.
Papini, G., Muys, M., Van Winckel, T., Meerburg, F.A., Van Beeck, W., Vermeir, P.,
Vlaeminck, S.E., 2023. Boosting aerobic microbial protein productivity and quality
on brewery wastewater: impact of anaerobic acidication, high-rate process and
biomass age. Bioresour. Technol. 368, 128285.
Parizzi, L.P., Grassi, M.C.B., Llerena, L.A., Carazzolle, M.F., Queiroz, V.L., Lunardi, I.,
Zeidler, A.F., Teixeira, P.J., Mieczkowski, P., Rincones, J., 2012. The genome
sequence of Propionibacterium acidipropionici provides insights into its
biotechnological and industrial potential. BMC Genomics 13 (1), 1–20.
Peltroche-Llacsahuanga, H., Frye, B., Haase, G., 2012. Isolation of Streptococcus urinalis
from a human blood culture. J. Med. Microbiol. 61 (5), 740–742.
Peng, L., Lou, W., Xu, Y., Yu, S., Liang, C., Alloul, A., Song, K., Vlaeminck, S.E., 2022.
Regulating light, oxygen and volatile fatty acids to boost the productivity of purple
bacteria biomass, protein and co-enzyme Q10. Sci. Total Environ. 822, 153489.
Pikaar, I., Matassa, S., Bodirsky, B.L., Weindl, I., Humpenoder, F., Rabaey, K., Boon, N.,
Bruschi, M., Yuan, Z., van Zanten, H., Herrero, M., Verstraete, W., Popp, A., 2018.
Decoupling livestock from land use through industrial feed production pathways.
Environ. Sci. Technol. 52 (13), 7351–7359.
Pikaar, I., Cruz, H., Hülsen, T., Batstone, D., Rabaey, K., Jensen, P., Avnimelech, Y.,
Verstraete, W., 2022. Producing microbial-based protein from reactive nitrogen
recovered from wastewater. Resource Recovery from Water. IWA Publishing,
pp. 223–244.
Prosser, J.I., 2010. Replicate or lie. Environ. Microbiol. 12 (7), 1806–1810.
Rairat, T., Chuchird, N., Keetanon, A., Carcano, P., Comi, M., Koppe, W., 2022. Effects of
dietary yeast-derived nucleotide and RNA on growth performance, survival, immune
responses, and resistance to Vibrio parahaemolyticus infection in Pacic white shrimp
(Litopenaeus vannamei). Aquac. Rep. 27, 101352.
Ramos, J.L., Bernal, P., Salvachúa, D., 2023. Microbial biotechnology in the effort to end
hunger. Microb. Biotechnol. 16 (8), 1581–1583.
Ritala, A., Hakkinen, S.T., Toivari, M., Wiebe, M.G., 2017. Single cell protein-state-of-
the-art, industrial landscape and patents 2001-2016. Front. Microbiol. 8, 2009.
Rodelas, B., Salmeron, V., Martinez-Toledo, M., Gonz´
alez-L´
opez, J., 1993. Production of
vitamins by Azospirillum brasilense in chemically-dened media. Plant Soil. 153 (1),
97–101.
Rombouts, J.L., Kranendonk, E.M.M., Regueira, A., Weissbrodt, D.G., Kleerebezem, R.,
van Loosdrecht, M.C.M., 2020. Selecting for lactic acid producing and utilising
bacteria in anaerobic enrichment cultures. Biotechnol. Bioeng. 117 (5), 1281–1293.
Rungsirivanich, P., Inta, A., Tragoolpua, Y., Thongwai, N., 2019. Partial rpoB gene
sequencing identication and probiotic potential of Floricoccus penangensis ML061-4
isolated from Assam tea (Camellia sinensis var. assamica). Sci. Rep. 9 (1), 16561.
Salazar-Garcia, G., Balaguera-Lopez, H.E., Hernandez, J.P., 2022. Effect of plant growth-
promoting bacteria azospirillum brasilense on the physiology of radish (Raphanus
sativus L.) under waterlogging stress. Agronomy 12 (3), 726.
Santillan, E., Constancias, F., Wuertz, S., 2020a. Press disturbance alters community
structure and assembly mechanisms of bacterial taxa and functional genes in
mesocosm-scale bioreactors. Msystems 5 (4) e00471-20.
Santillan, E., Phua, W.X., Constancias, F., Wuertz, S., 2020b. Sustained organic loading
disturbance favors nitrite accumulation in bioreactors with variable resistance,
recovery and resilience of nitrication and nitriers. Sci. Rep. 10 (1), 1–11.
Santillan, E., Seshan, H., Constancias, F., Drautz-Moses, D.I., Wuertz, S., 2019a.
Frequency of disturbance alters diversity, function, and underlying assembly
mechanisms of complex bacterial communities. NPJ Biolms Microbiom. 5 (1), 1–9.
Santillan, E., Seshan, H., Constancias, F., Wuertz, S., 2019b. Trait-based life-history
strategies explain succession scenario for complex bacterial communities under
varying disturbance. Environ. Microbiol. 21 (10), 3751–3764.
Santillan, E., Seshan, H., Wuertz, S., 2021. Press xenobiotic 3-chloroaniline disturbance
favors deterministic assembly with a shift in function and structure of bacterial
communities in sludge bioreactors. ACS ES&T Water 1 (6), 1429–1437.
Santillan, E., Wuertz, S., 2022. Microbiome assembly predictably shapes diversity across
a range of disturbance frequencies in experimental microcosms. NPJ Biolms
Microbiom. 8 (1), 41.
Santillan, E., Yasumaru, F., Vethathirri, R.S., Thi, S.S., Hoon, H.Y., Sian, D.C.P.,
Wuertz, S., 2024. Microbial community-based protein from soybean-processing
wastewater as a sustainable alternative sh feed ingredient. Sci. Rep. 14 (1), 2620.
Scheiman, J., Luber, J.M., Chavkin, T.A., MacDonald, T., Tung, A., Pham, L.-D.,
Wibowo, M.C., Wurth, R.C., Punthambaker, S., Tierney, B.T., 2019. Meta-omics
analysis of elite athletes identies a performance-enhancing microbe that functions
via lactate metabolism. Nat. Med. 25 (7), 1104–1109.
Seshan, H., Santillan, E., Constancias, F., Chandra Segaran, U.S., Williams, R.B.H.,
Wuertz, S., 2023. Metagenomics and metatranscriptomics suggest pathways of 3-
chloroaniline degradation in wastewater reactors. Sci. Total Environ. 903, 166066.
Shao, M., Liang, H., Xu, G., Zhu, J., Li, S., Ren, M., 2022. Dietary leucine
supplementation improves growth performance, metabolic responses of liver via
GCN2/ATF4, and insulin signaling pathways in largemouth bass (Micropterus
salmoides). Fish Physiol. Biochem. 1–17.
Sharif, M., Zafar, M.H., Aqib, A.I., Saeed, M., Farag, M.R., Alagawany, M., 2021. Single
cell protein: sources, mechanism of production, nutritional value and its uses in
aquaculture nutrition. Aquaculture 531, 735885.
Shimizu, G., Cotta, M., Bothast, R., 1989. The kinetics of glucose fermentation by
Selenomonas ruminantium HD4 grown in continuous culture. Biotechnol. Lett. 11 (1),
67–72.
ˇ
Slos´
arek, M., ˇ
Sourek, J., Mikov´
a, Z., 1976. Results of long-term preservation of
mycobacteria by means of freeze-drying. Cryobiology 13 (2), 218–224.
R.S. Vethathirri et al.
Cleaner and Circular Bioeconomy 10 (2025) 100139
11
Stecher, G., Tamura, K., Kumar, S., 2020. Molecular evolutionary genetics analysis
(MEGA) for macOS. Mol. Biol. Evol. 37 (4), 1237–1239.
Stoffels, M., Castellanos, T., Hartmann, A., 2001. Design and application of new 16S
rRNA-targeted oligonucleotide probes for the Azospirillum-Skermanella-Rhodocista-
cluster. Syst. Appl. Microbiol. 24 (1), 83–97.
Suman, G., Nupur, M., Anuradha, S., Pradeep, B., 2015. Single cell protein production: a
review. Int J Curr Microbiol Appl Sci 4 (9), 251–262.
Tan, C., Li, Q., Yang, X., Chen, J., Zhang, Q., Deng, X., 2022. Lactococcus lactis’ effect on
the intestinal microbiota of Streptococcus agalactiae-infected zebrash (Danio rerio).
Microbiol. Spectr. 10 (5) e01128-22.
Thijs, S., Op De Beeck, M., Beckers, B., Truyens, S., Stevens, V., Van Hamme, J.D.,
Vangronsveld, J., 2017. Comparative evaluation of four bacteria-specic primer
pairs for 16S rRNA gene surveys. Front. Microbiol. 8, 251189.
Tian, G.Z., Piao, D.R., Zhao, H.Y., Jiang, H., Cui, B.Y., Li, J.Y., 2011. A Flavobacterium
lindanitolerans strain isolated from the ascites sample of a chinese patient with EV71
virus infection. Biomed. Environ. Sci. 24 (6), 694–696.
Verstraete, W., Yanuka-Golub, K., Driesen, N., De Vrieze, J., 2022. Engineering microbial
technologies for environmental sustainability: choices to make. Microb. Biotechnol.
15 (1), 215–227.
Vethathirri, R.S., Santillan, E., Thi, S.S., Hoon, H.H., Wuertz, S., 2023. Microbial
community-based production of single cell protein from soybean-processing
wastewater of variable chemical composition. Sci. Total Environ. 873, 162241.
Vethathirri, R.S., Santillan, E., Wuertz, S., 2021. Microbial community-based protein
production from wastewater for animal feed applications. Bioresour. Technol. 341,
125723.
Wang, Q., Xu, Z., Ai, Q., 2021. Arginine metabolism and its functions in growth, nutrient
utilization, and immunonutrition of sh. Anim. Nutr. 7 (3), 716–727.
Webster, C.D., Lim, C., 2002. Nutrient Requirements and Feeding of Finsh For
Aquaculture. Cabi.
Wittebolle, L., Verstraete, W., Boon, N., 2009. The inoculum effect on the ammonia-
oxidizing bacterial communities in parallel sequential batch reactors. Water Res. 43
(17), 4149–4158.
Xu, J., Zeng, X., Jiang, N., Zhou, Y., Zeng, L., 2015. Pseudomonas alcaligenes infection and
mortality in cultured Chinese sturgeon, Acipenser sinensis. Aquaculture 446, 37–41.
Zhu, C.-Z., Li, D., Chen, W.-J., Ban, S.-N., Liu, T., Wen, H., Jiang, M., 2021. Effects of
dietary host-associated lactococcus lactis on growth performance, disease resistance,
intestinal morphology and intestinal microbiota of mandarin sh (Siniperca chuatsi).
Aquaculture 540, 736702.
Zhuang, H., Cheng, Z., Shan, S., Shen, H., Zhao, B., 2020. Demonstration on the
treatment of paper-making wastewater by a full-scale IC-A/O-membrane reactor
system for reclamation. J. Chem. Technol. Biotechnol. 95 (12), 3161–3168.
R.S. Vethathirri et al.
Cleaner and Circular Bioeconomy 10 (2025) 100139
12
