See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/236139190

Algae biofilm growth and the potential to stimulate lipid accumulation

through nutrient starvation

Article  in  Bioresource Technology · March 2013

DOI: 10.1016/j.biortech.2013.03.036 · Source: PubMed

CITATIONS READS

71 621

3 authors, including:

Peter Schnurr D. Grant Allen

Minnow Environmental University of Toronto
9 PUBLICATIONS   182 CITATIONS    148 PUBLICATIONS   3,218 CITATIONS   

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Peter Schnurr on 17 May 2019.

The user has requested enhancement of the downloaded file.

 Bioresource Technology 136 (2013) 337–344

Contents lists available at SciVerse ScienceDirect

Bioresource Technology
journal homepage: www.elsevier.com/locate/biortech

Algae biofilm growth and the potential to stimulate lipid accumulation

through nutrient starvation
Peter J. Schnurr a, George S. Espie b, D. Grant Allen a,⇑
a
Department of Chemical Engineering & Applied Chemistry, University of Toronto, 200 College Street, Toronto, Ontario, Canada M5S 3E5
b
Department of Cell & Systems Biology, University of Toronto, 3359 Mississauga Road North, Mississauga, Ontario, Canada L5L 1C6

h i g h l i g h t s

 A system was developed to study the growth kinetics of algal biofilms.

 Algal biofilm growth kinetics are linear.
 Nutrient starvation caused sloughing and/or growth cessation of algae biofilms.
 Nutrient starvation did not increase in algal biofilm lipid concentrations.
 Algal biofilm lipid productivities were comparable to terrestrial biofuel sources.

a r t i c l e i n f o a b s t r a c t

Article history: An algae biofilm growth system was developed to study the growth kinetics and neutral lipid productiv-
Received 17 December 2012 ities of Scenedesmus obliquus and Nitzschia palea, and to determine if algal biofilms can be starved of key
Received in revised form 5 March 2013 nutrients to increase their neutral lipid concentrations. Linear growth curves were determined for each
Accepted 7 March 2013
species until nutrient starvation commenced, at which point growth ceased and/or biofilms sloughed
Available online 15 March 2013
from their substratum. Nutrient starvation did not increase neutral lipid concentrations in any of the bio-
films; however, it approximately doubled their lipid concentrations when grown in suspension. Biomass
Keywords:
productivities of 2.8 and 2.1 g/m2/d and lipid productivities of 0.45 and 0.18 g/m2/d were determined for
Biofuels
Algae
N. palea and S. obliquus, respectively. The results suggest that nutrient starvation of biofilms is not a desir-
Biofilms able method of lipid production for algae biofilm biofuel production systems, but that lipid production
Lipids rates compare favorably with conventional terrestrial biofuel sources.
Nutrients Ó 2013 Elsevier Ltd. All rights reserved.

1. Introduction
In recent years researchers around the world have been focus-
ing their efforts on producing biofuels and biochemicals from the
growth of algae cultures. This effort has been inspired by the need
to develop renewable biofuels and other biochemicals in order to
alleviate the environmental effects associated with the use of
petroleum fuels and chemicals, and because conventional fuel re-
sources are dwindling and are less secure (Li et al., 2008). In addi-
tion, algae produce some potentially high value chemicals such as
b-carotene and astaxanthin (Barowitzka, 1992). Biofuel and bio-
chemical development from culturing microalgae simultaneously
addresses the above challenges if the technology can be economi-
Abbreviations: EPS, extracellular polymeric substance; PBR, photobioreactor;
LED, light emitting diode; BSE, backscatter electron; SE, secondary electron; FAME,
fatty acid methyl ester; GC, gas chromotography.
⇑ Corresponding author. Tel.: +1 (416) 9788517; fax: +1 (416) 9788605.
E-mail addresses: peter.schnurr@utoronto.ca (P.J. Schnurr), george.espie@
utoronto.ca (G.S. Espie), dgrant.allen@utoronto.ca (D.G. Allen).
cal and energy efficient, allowing nations to ‘‘grow’’ fuels and
chemicals from inorganic nutrients through photosynthesis.
One of the most compelling advantages for algae as a source of
renewable energy is the potentially high rates of productivity of
many algal species compared to conventional biofuel stocks such
as corn, soybean, rapeseed, and other plants (Chisti, 2007). It has
been reported that some algae species can produce up to 30 times
more bioenergy per area of land than the most productive terres-
trial oil seed plants (Johnson and Wen, 2010; Mata et al., 2010).
This is due to rapid growth with population doubling times as
low as 3.5 h and to algae biomass lipid/oil concentrations in the or-
der of 15–75% of the dry weight (Chisti, 2007). Algal systems also
offer the potential to simultaneously treat wastewater and flue
gas streams by utilizing the nitrogen, phosphorus and carbon diox-
ide within these streams.
Algal lipid yields may be enhanced by exposing cultures to a
range of environmental stresses prior to harvest (Hu et al., 2008;
Opute, 1974). Most of this increase in lipid concentration is in
the form of neutral lipids, particularly triacylglycerol – the type
generally used for biofuel production (Hu et al., 2008), aquaculture

0960-8524/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.biortech.2013.03.036
338 P.J. Schnurr et al. / Bioresource Technology 136 (2013) 337–344
feeds, and nutraceuticals (Sicko-Goad and Anderson, 1991). These
neutral lipids form in the cytoplasm as dense lipid bodies (Hu
et al., 2008; Sriharan et al., 1991), and arise from the assimilation
of new carbon (from CO2), and from the conversion of already
assimilated carbon in carbohydrates, proteins and other internal
biochemical molecules (Rodolfi et al., 2009; Roessler, 1988).
Although many types of environmental stress can trigger a lipid
accumulation response, the most commonly researched are nutri-
ent starvation stresses. Of the many potential types of nutrient
starvation stresses, the majority of research has been done on
nitrogen and silicon starvation. Silicon starvation is only relevant
for diatom algae species because it is a macronutrient essential
for cell division as it composes a significant fraction of the frustule
exoskeleton unique to diatoms (Spilling et al., 2010). Nitrogen is a
necessary macronutrient for all algae species because it is essential
for cell division as it composes functional groups of nucleic acids,
proteins and other biomolecules (Madigan et al., 2009).
Although the underlying mechanism is not well understood,
researchers have clearly demonstrated that starvation of key
macronutrients can significantly increase lipid concentrations
within algae biomass of many different species. Opute (1974)
determined that the lipid concentration of Nitzschia palea could
be increased from 22% to 35% after 5 days of silicon starvation.
Roessler (1988) reported an increase from 20% to 28% lipid concen-
tration when starving the marine diatom Cyclotella cryptica of sili-
con for just 12 h. Sriharan et al. (1991) starved C. cryptica until lipid
concentrations peaked (unreported amount of time) increasing the
lipid concentrations from 21% to 42%, and from 17% to 45%, for sil-
ica deficiencies and nitrogen deficiencies, respectively. Rodolfi
et al. (2009) starved a marine algae Nannochloropsis sp. of nitrogen
and increased the lipid concentration 4-fold (from 15% to 60%)
after 3 days of starvation. Lastly, Shifrin and Chisholm (1981)
starved 30 different algal species of nitrogen and silicon for 4 to
9 days and observed a 2 to 3-fold increase in lipids for green algae,
and both increases and decreases for diatoms. Thus, nutrient star-
vation can be an important industrial strategy to control and in-
crease the yield of neutral lipids prior to biomass harvest.
Although there are many attractive attributes to the production
of biofuels and biochemicals from algae cultures, there are signifi-
cant economical challenges to overcome before commercialization
of these technologies can occur. One of the most significant limita-
tions to the economical use of algae is the high cost of harvesting
and concentrating the biomass (Christenson and Sims, 2011; Gudin
and Therpenier, 1986; Uduman et al., 2010). In their review article,
Uduman et al. (2010) state that algal suspensions are often be-
Fig. 1. Schematic of the algae biofilm culturing system used to grow algae biofilms with controls on key growth parameters.
tween 0.02% and 0.06% total suspended solids (TSS), and that sig-
nificant energy is spent to harvest and concentrate the cells to 5–
25% TSS. Gudin and Therpenier (1986) report that energy for har-
vesting and de-watering accounts for 20–30% of the total biomass
production costs.
To circumvent these biomass-processing costs, we have been
investigating the utility of growing algae as a biofilm, rather than
in a conventional suspension culture. The idea is that by growing
algae as a biofilm the biomass can effectively be concentrated
and immobilized on a solid subsurface. Typically, biofilms range
from 6–16% total suspended solids (TSS) (Christenson and Sims,
2011; Johnson and Wen, 2010; Ozkan et al., 2012) thereby mini-
mizing the effort required to concentrate and de-water the bio-
mass during harvest. Additionally, the immobilized and
concentrated biomass can potentially reduce difficulty in down-
stream processing i.e. removing substratum-biomass structure
and directly immersing in solvent, or removing growth medium
from the reactor and adding solvent.
To date, there has been very little research on biofilm systems
for commercial production of algae and lipids (Christenson and
Sims, 2012; Irving and Allen, 2011; Johnson and Wen, 2010; Ozkan
et al., 2012). Most of the algal biofilm literature focuses on using
them as a means of wastewater treatment. In this paper the
dynamics of a laboratory-scale algal biofilm growth system were
explored to determine algal biofilm growth kinetics and lipid pro-
duction. Additionally, the hypothesis that nutrient starvation
(nitrogen and silicon) would elicit enhanced lipid production in
biofilm cultures, as is the case in algal suspensions, was tested.
2. Methods
An algal biofilm growth system was designed to characterize
and control many algal biofilm growth parameters while maintain-
ing the ability to manipulate the bulk medium nitrogen and silicon
concentrations part way through the experiment to elicit nutrient
starvation. Additionally, the system allows for the collection of
growth kinetics data while doing destructive lipid analysis on the
biofilms.
2.1. Biofilm culturing system
The semi-continuous flat plate parallel horizontal photobioreac-
tor (PBR) system (Fig. 1) had significant re-circulation (95% by
volume) to re-use the large amounts of medium going through
the reactors. The fresh feed medium (5% by volume of total to
 P.J. Schnurr et al. / Bioresource Technology 136 (2013) 337–344
each reactor) allowed for control of key nutrient (N and Si) concen-
trations once biofilms were grown and the starvation period com-
menced. Flow rates for the re-circulation and fresh feed streams
within the system were controlled with Cole Parmer Masterflex,
Model #7523-90 peristaltic pumps. Carbon dioxide was sparged
into the growth medium in the mixing vessel at 2% CO2 (v/v),
which corresponds to 30 mg/L at 25 °C and a total pressure of
1 atm according to Henry’s Law. The concentration of CO2 was con-
trolled through the use of rotameters to control the relative flow
rates of compressed air and CO2. Growth medium exited the mix-
ing vessel into a manifold, which divided the stream into 18 sepa-
rate tubes. A Cole Parmer Masterflex, Model 7520-50 peristaltic
pump and 18 dampeners controlled the flow rates and hence res-
idence time and shear rates into 18 replicate PBR’s. The shear rates
used in these experiments were calculated to be 4 and 7 s 1 corre-
sponding to flow rates of between 13.3 and 23.3 mL/min, respec-
tively, through each bioreactor. These shear rates were calculated
by assuming fully developed flow, by calculating the hydraulic
diameter of the reactors which had rectangular cross section, the
Reynolds Number (19 and 33 for 4 and 7 s 1, respectively), the
friction factor based on laminar flow, the wall shear stress, and
then finally, the shear rate. There were 18 (20  2  0.4 cm,
length  width  height) replicate PBR’s in order to facilitate
destructive analysis with time, particularly dry weight per area of
substrate, and lipid concentrations in the dry biomass. The PBR’s
were illuminated with custom-made strips of 8 Watt, red-light emit-
ting diodes (LED) positioned at the top and the bottom of the PBR.
An artificial (vs natural) light system was selected to ensure that
we had well controlled, reproducible conditions to compare the var-
ious biofilm growth conditions. Red LED lights were chosen because
chlorophyll a absorbs in the red part of the visible light spectrum
and studies have shown that there are no differences in growth rates
when systems are irradiated with red light compared to all photo-
synthetically active radiation (PAR) (Wang et al., 2007). Additionally,
Gordon and Polle (2007) state that on a per watt basis PAR is approx-
imately half as effective as red light in getting a photosynthetic spec-
tral response. Pond Biofuels, Toronto, Canada, supplied the lighting
system. Light was maintained at 80 lmol/m2/s at both surfaces
(top and bottom each) of the PBRs on a 16/8 h light/dark cycle. Light
intensities were measured as photosynthetically active radiation
with a FieldScout light meter and quantum sensor from Spectrum
Technologies. Bulk medium temperatures were controlled at
26 ± 2 °C by controlling the ambient temperature of the laboratory.
The pH was controlled at 7 ± 0.2 with an Omega PHP-700 Series
pH meter and ORP Controller and Dosing Pump. Because the med-
ium drift was towards acidic conditions, a 0.5 M NaOH solution
was used to control the pH. Manifolds collected the PBR outlet med-
ium into a single tube, which emptied into a collection flask, where-
by the medium either left the system or was re-circulated.
2.2. Experimental approach
2.2.1. Biofilm growth
Irving and Allen (2011) determined that the presence of waste-
water is an important factor for enhancing the formation of an al-
gae biofilm. The presence of bacteria and extracellular polymeric
substances (EPS) in the wastewater apparently created a positive
environment on which to seed an algal biofilm. Therefore, the PBRs
were conditioned with flowing unsterile secondary wastewater
effluent from Ashbridge’s Bay Wastewater Treatment Facility, Tor-
onto, ON, at a rate of 13.3 or 23.3 ml/min, for 48 h. It was assumed
that this step formed an ‘initial’ biofilm and EPS matrix for which
the inoculated cells could adhere. After 48 h the flow was stopped
and approximately 90% of the wastewater was removed from the
culturing system to immediately be replaced with a biofilm growth
medium. The 10% of residual wastewater that remained in the sys-
339
tem was intended to ‘seed’ the biofilm growth medium with bacte-
ria, dissolved organic compounds, and EPS.
A synthetic biofilm growth medium was developed because,
compared to wastewater, it is consistent and easily characterized
and controllable for when nutrient starvation is commenced part
way through the experiments. This synthetic medium was created
using the macro and micro nutrients of the CHU-10 diatom med-
ium. The phosphate concentrations were amended to 75 and
175 mg/L for K2HPO4 and KH2PO4, respectively, while the NH4Cl
and Ca(NO3)24H2O concentrations were amended to 120 and
20.2 mg/L, respectively.
Immediately following the switch to biofilm growth medium,
the pumps were stopped in order to inoculate the PBR’s. This
was done by injecting 15 mL of an algal inoculum directly into each
of the 18 reactors. The inoculum was 200–300 mg/L (dry weight) of
a pure culture of N. palea (CPCC 160) or Scenedesmus obliquus (CPCC
5) supplied by the Canadian Phycological Culture Centre, Univer-
sity of Waterloo, ON. After inoculation, the pumps were left idle
for 24 h in order to facilitate initial adhesion of algae cells to the
glass substrate in the PBR’s.
The pumps then were started and the first samples were taken
for the experiment. During each experiment biofilms were grown
for 20–21 days; however, at day 13–14 the nutrient starvation per-
iod began by exchanging nutrient replete medium with nutrient
depleted medium. This was accomplished in 2 ways: (1) 95% of
nutrient replete medium was immediately taken from system
and replaced with nutrient deficient medium; (2) the remaining
nitrogen and silicon in the lines, dampeners, reactors, etc., was di-
luted out with fresh feed of nutrient deficient medium. Given that
the residence time in the system was about 0.11 days, virtually
complete flushing of the system was expected within 1 day (i.e. 9
residence times).
The first set of experiments was run in the biofilm culturing sys-
tem at a shear rate of 4 s 1. Under these conditions triplicate
experiments of 20 days each were run with S. obliquus as the inoc-
ulum. After 13 days of growth the biofilms were starved of nitro-
gen and silicon. A negative control non-starved S. obliquus culture
was grown for 26 days to compare to the starved cultures. The sec-
ond set of experiments was run at 7 s 1. These two separate exper-
iments utilized two different inoculating species – N. palea and S.
obliquus as the inocula. Nutrient starvation commenced after
14 days of biofilm growth.
2.2.2. Suspended cultures
Suspension cultures of N. palea and S. obliquus were also ex-
posed to nutrient deficient conditions in order to determine the
enhancement of lipid accumulation, as is suggested in the litera-
ture. This was done by growing two pure cultures of each of the
two species in flasks sparged with 2% CO2 at a volumetric flow rate
of 1 L/min. Additionally, each flask was maintained at 25 °C, buf-
fered at a pH of 6.8, and illuminated at 100 lmol/m2/s with Lumina
34 Watt cool white fluorescent lights. In order to accumulate en-
ough biomass to measure lipids each culture was grown for 3–
4 weeks with the same nutrient replete growth medium used in
the biofilm experiments. Then, one flask was harvested – the
non-starved culture – for lipid analysis. The second culture was
centrifuged at 4000g for 10 min to remove the nutrient replete
medium, and the cell pellet was re-suspended in the flask with
nutrient deficient medium – void of nitrogen and silicon chemical
species. These cultures were then grown for another 3 days under
nutrient deficient conditions and then harvested for lipid analysis.
2.3. Sampling and analysis
Each of the 18 replicate reactors represented a sample for anal-
ysis. The reactors were harvested in duplicate or triplicate, 2–3
340 P.J. Schnurr et al. / Bioresource Technology 136 (2013) 337–344

times per week. Additionally, the sampling was designed such that used, and absorbances were measured to calculate and directly
there were 2 sets of duplicate or triplicate samples after nutrient read out the concentration of the analyte.
starvation commenced. One set was harvested and analyzed after
3–4 days of nutrient starvation while the other set was harvested
after 6–7 days. To harvest, the glass substrata were removed from 3. Results and discussion
the PBRs and scraped clean of their algal biomass. The algal bio-
mass was then lyophilized with a Labconco Lyph Lock 6 Freeze 3.1. Biofilm growth kinetics and the affects of nutrient starvation
Dryer to determine dry weight. The algae biomass was then ex-
tracted to determine lipid content. In all experiments the algae biofilm growth system was effec-
Biofilm and suspended algae biomass neutral lipid concentra- tive in reproducibly growing thick and robust algal biofilms.
tions were determined by first extracting total lipids with the Folch Fig. 2a shows that prior to starvation the algae biofilm growth
extraction method using 2:1 (v/v) chloroform to methanol as the kinetics of the three S. obliquus experiments were reproducible
solvent. Upon extraction the methanol phase of the Folch and polar and linear. Although grown in the presence of bacteria and other
lipids solution was discarded because only neutral lipids were tar- microbes, qualitative observation of the biofilms shows a very
geted. Neutral lipids, predominately composed of triacylglycerides, green film indicating a high proportion of algae (Fig. 3a). From
were targeted because they are the ones which accumulate during the date of inoculation onward the biofilms gradually progressed
nutrient starvation, and because they are the preferred feedstock to being more stable and robust, having the cells firmly bonded to-
for biodiesel production (Hu et al., 2008). The chloroform was then gether as one film. Although the biofilms were primarily comprised
evaporated under a nitrogen blanket to isolate the extracted of algae, qualitative observations with SEM show significant pres-
neutral lipids from solution. The neutral lipids were then methyl- ence of bacteria and EPS as well.
ated using the Fatty Acid Methyl Ester (FAME) technique based The algae biofilm growth kinetics and biofilm sloughing were
on the Microbial Identification System, Microbial ID Inc. (MIDI significantly affected by the levels of nitrogen and silicon. As
Method) (Smid and Salfinger, 1994). The samples were then ana- shown in Figs. 2a and 2b, the algal biofilms grew in an approxi-
lyzed by gas chromatography (GC) using a Perkin Elmer Clarus mately linear fashion until nitrogen and silicon were flushed from
680 GC. A Hewlett Packard HP-5 MS 30 m  0.25 mm  0.25 lm the media to below their detection limit (<1 mg/L, TN; <0.01 mg/L
column was used to separate the methylated fatty acids in the
gas chromatography instrument. In order to determine the mass
of fatty acids/lipids present in the algae biomass samples, the ratio
of the area under the calibration curve to that of the internal stan-
dard was compared to those ratios in the unknowns. The calibra-
tion curve was made from 100% olive oil, while the internal
standard was hexadecane. Lastly, the lipid mass was compared to
the dry weight of algae mass to determine the percent lipid
concentration.
Scanning electron microscopy was used to observe the presence
of microbes and EPS within the biofilm. Small samples of live bio-
films were pelletized and treated for 10 min with osmium tetrox-
ide. The osmium tetroxide bonded with cellular lipids increasing
density of the high lipid biofilm components, mainly algae cells.
Backscatter electron (BSE) mode was then used to observe these
osmium tetroxide treated high-density cells. Secondary electron
(SE) mode was used to image everything else present within the
field of view i.e. bacteria, EPS, inert solids. BSE and SE images were Fig. 2a. Algae biofilm growth kinetics grown at a shear rate of 4 s 1. Triplicate
experiments were run to determine the affects of nutrient starvation on growth
overlayed upon each other in order to show the high resolution al-
kinetics. Nutrient starvation of N and Si began at the 13th day of growth. A fourth
gae cells within the biofilm matrix. The scanning electron micro- experiment was run for 26 days without any nutrient deficiencies.
scope used was a Hitachi S-3400N.
Total nitrogen and total silicon analysis was conducted on the
bulk medium to measure the level of these nutrients prior to star-
vation, and to follow the disappearance of these nutrients after the
starvation period commenced. These analyses were conducted
using HACH chemical reagent kits, which utilizes principles of col-
orimetric analysis. The chemical reagent kits used for total nitro-
gen were TNTplus (827) HR for high range analysis of 5–40 mg/L,
and TNTplus (826) LR for low range analysis of 1–16 mg/L. For
determination of total silicon the reagent kit Silica Reagent Set
High Range (1–100 mg/L), and the Silica Reagent Set Low Range
(0.01–1.60 mg/L) were used.
For preparation and analysis of the samples a HACH digestion
instrument and a HACH spectrophotometer were used. The HACH
digestion (DRB200) unit was used to digest all nitrogen compounds
into a single form that would react with the HACH reagent sup-
plied in the kit. Upon digestion and treatment of samples with
the HACH reagent, a HACH DR3900 spectrophotometer was used
to analyze all samples for total nitrogen and total silicon concen- Fig. 2b. Algae biofilm growth kinetics grown at a shear rate of 7 s 1 for both
trations. The wavelengths for analysis were programed directly Nitzschia palea and Scenedesmus obliquus. Nutrient starvation at the 14th day of
into the spectrophotometer according to the reagent kit being growth.
 P.J. Schnurr et al. / Bioresource Technology 136 (2013) 337–344
Fig. 3a. Algae biofilms inoculated with S. obliquus and grown for 14 days under nutrient replete conditions.
Fig. 3b. Algae biofilms inoculated with S. obliquus and grown for 14 days under nutrient replete conditions followed by 3 days of exposure to nutrient deficient conditions.
Si), after which point the biofilms significantly sloughed from the
substratum it was fixed to (Fig. 3b), and/or ceased to grow. Lag
phases and sloughing events were variable between the triplicate
S. obliquus experiments ran at 4 s 1; these types of variability are
not uncommon in biological systems. A non-starved culture of S.
obliquus grew for 26 days without any cessation of growth or sig-
nificant sloughing. The N. palea biofilm grown at 7 s 1 had the least
amount of observable and quantifiable sloughing.
The cessation of growth with nitrogen limitations is expected
because without this nutrient these organisms cannot produce
the amino acids, nucleic acids, and proteins necessary for growth
and reproduction (Madigan et al., 2009). Similarly, diatoms cannot
produce the frustule exoskeleton without silicon (Spilling et al.,
2010). Opute (1974) suggested another potential reason for growth
inhibition. He suggested that, when starved of key nutrients, algae
release an autotoxin that inhibits cell division while photosynthe-
sis continues to assimilate carbon and store it in the form of lipids.
Since growth is inhibited during nutrient starvation the overall va-
lue of using this method to increase lipid productivities is debat-
able because the growth rate-lipid concentration trade-off is
delicate (Opute, 1974; Rodolfi et al., 2009).
Sloughing of the biofilms may be a result of EPS matrix degrada-
tion by enzyme release induced by nutrient starvation. (Davies,
1999; Wingender et al., 1999). Davies (1999) suggests that chang-
ing environments will cause bacterial cells to down regulate algi-
nate promoters causing much less alginate to be produced,
eventually leading to cell dettachment and sloughing. Additionally,
and perhaps more importantly, Davies (1999) states that when ex-
posed to vital nutrient deficiencies, bacteria can release alginate
lyase, leading to the degradation of the EPS matrix bonding the al-
gae cells to each other and to the substratum. Recall that the liter-
ature suggests, and that the research in this paper anticipated,
biofilms bound together by EPS produced by bacteria; however, al-
gae cells may also behave in a similar fashion as bacteria when ex-
posed to nutrient deficient conditions i.e. down regulate alginate
promoters and release enzymes.
Hydrolysis and degradation of the EPS may also occur during
periods of nutrient starvation in an attempt to either utilize the
nutrients within the biomolecules in the matrix (Wingender
et al., 1999), or to dis-attach and re-populate where more nutrient
favorable conditions exist. Wingender et al. (1999) and Wolfaardt
et al. (1999) both suggest that bacteria will use enzyme-substrate
interactions in order to utilize nutrients in EPS during periods of
nutrient stress. Although these authors mostly write about carbon
nutrient limitations on bacterial biofilms, and not about nitrogen
341
or silicon limitations within algal-bacterial biofilm consortia, a
similar phenomenon of microbe nutrient acquisition may be occur-
ring in these consortiums.
In addition, quorum-sensing mechanisms within biofilms could
lead to other reasons for dis-attachment within the film (Decho,
1999). Wingender et al. (1999) suggest that cell-to-cell communi-
cations are fundamental for biofilms to adapt and respond to cer-
tain environmental conditions, including nutrient limitations. It
may be that the cells exposed to adverse conditions such as nutri-
ent limitations will utilize an adaptive strategy whereby they deat-
tach from the substratum in order to reinhabit a new environment
where nutrients are more abundant.
Linear, rather than exponential growth, suggests some chemical
or physical limitations to growth. This could be due to limitations
of light and nutrient transport, and/or the result of cell deattach-
ment from the biofilm. The linear growth could also be a result
of allocation of electrons and other resources i.e. carbon, nitrogen,
into building the EPS matrix rather than reproduction (Laspidou
and Rittmann, 2002). Wolfaardt et al. (1999) suggest that in bacte-
rial systems up to 70% of the carbon and energy can go into the
production of EPS, rather than cell reproduction. Additionally, Da-
vies (1999) determined that when bacterial cells form biofilms
they make up to 19 times more EPS than planktonic cells. Algae
cells may behave similarly in the sessile state.
The differences in sloughing observed between N. palea and S.
obliquus (Fig. 2b) might be due to the biological properties of the
two species – N. palea is a diatom and diatoms are generally more
of a sessile group of algae species whereas green algae (S. obliquus)
are more of a planktonic-type group of species (Congestri et al.,
2006). Additionally, through observations in the lab, and corre-
spondence with the Canadian Phycological Culture Centre, N. palea
was determined to be a very adherent species, readily forming bio-
films and flocs under sterile environments. Thus, N. palea is likely
to form a more robust and resilient biofilm, even while exposed
to nutrient deficient conditions for relatively short periods of time.
3.2. Lipid concentrations of algae grown in nutrient replete and
deficient conditions
In contrast to planktonic algae, algal biofilms did not accumu-
late more neutral lipids when exposed to nutrient deficient condi-
tions for 3–4 or 6–7 days (Fig. 4). There was no significant
difference between the neutral lipid concentrations of either
S. obliquus or N. palea after 3 or 6 days of starvation. On the
other hand, consistent with the literature, planktonic N. palea and
342 P.J. Schnurr et al. / Bioresource Technology 136 (2013) 337–344
Fig. 4. Algal biofilms show no increase in biofilm lipid concentrations after 3–4 and
6–7 days of nutrient starvation. Conversely, lipid concentrations in planktonic
cultures of both Nitzschia palea and Scenedsmus obliquus were nearly doubled when
exposed to 3 days of nutrient deficient conditions.
S. obliquus exposed to nutrient deficient conditions significantly in-
creased their neutral lipid concentrations when starved for 3 days
(Fig. 4). Thus, there appears to be a fundamental difference in the
way that suspension cultures and biofilm cultures respond to
nutrient deprivation.
The biofilm lipid concentrations of S. obliquus and N. palea ob-
served under nutrient replete and deficient conditions were simi-
lar, though slightly lower than the lipid concentrations of the
same species grown planktonically (Fig. 4). Additionally, the lipid
concentrations of the planktonic cultures grown in these experi-
ments are considerably lower than the values reported in the liter-
ature (Mata et al., 2010; Shifrin and Chisholm, 1981).
Though the inability for the algal biofilms to store lipids under
nutrient deficient conditions in the bulk medium was unexpected,
we offer a number of possible reasons for this result: (1) the EPS
matrix adsorbs various nutrients acting as a reservoir during nutri-
ent starved conditions (Laspidou and Rittmann, 2002; Neu and
Lawrence, 1999; Wolfaardt et al., 1999); (2) biofilms are notori-
ously resilient to environmental stresses – it is one of the reasons
microbes form biofilms (Wingender et al., 1999); (3) as discussed
earlier, the microbes may be releasing enzymes which break up
the EPS (Davies, 1999; Wolfaardt et al., 1999) and, with the aid
of mineralizing bacteria, can utilize the released nutrients; (4)
there are other forms of nutrient recycling and regeneration within
the biofilm, including use of dead biomass, inert solids, and exog-
enous biochemicals (Laspidou and Rittmann, 2002; Wetzel, 1993).
The lower lipid concentrations of the biofilms compared to the
planktonic systems may be a result of the composition of the
microbial community in the biomass. Mixed community biofilms
can have a high proportion of bacteria and EPS (Wingender et al.,
1999), which tend to have a significantly lower neutral lipid con-
centration than algae (Ames, 1968; Flemming et al., 2007). Ames
(1968) showed that the neutral lipid fraction of Salmonella typ-
hilmurium and of Escherichia coli was 0.7% (w/w) of bacteria bio-
mass. Biofilms also accumulate other forms of inert biomass
(Laspidou and Rittmann, 2002), which have little or no lipid con-
tent. Additionally, the lowering of the lipid concentrations with
time might be a result of succession to more EPS and (cyano) bac-
teria within the biofilm. Indeed, observation with SEM showed that
a considerable fraction of the biomass present in the biofilms
grown was non-algal biomass.
Consistent with previous research, both N. palea and S. obliquus
grown planktonically, significantly increased lipid concentrations
when exposed to 3 days of nutrient deficient conditions. Perhaps,
as is the case with bacterial cells, algae cells make significantly less
EPS matrix when grown planktonically (Davies, 1999), which
makes them more susceptible to nutrient deficiencies, causing
them to respond with lipid accumulation (discussed already
above). Additionally, these planktonic cultures were grown with-
out the seeding from untreated secondary wastewater effluent,
thus do not have bacteria, and likely have far less EPS. Having less
bacteria and EPS to compose the biomass would theoretically re-
sult in higher overall biomass lipid concentrations, as discussed
above. It is likely one, or a combination of these things, which
causes planktonic algae to accumulate neutral lipids when exposed
to nutrient deficient conditions, and to have higher lipid concentra-
tions than algae biofilms.
Hu et al. (2008) state that one or a combination of chemical and
physical stresses can cause significant neutral lipid accumulation
in planktonic algal systems. These stresses include extreme tem-
perature and light regimes for physical stresses, and nutrient star-
vation, extreme salinity and pH for chemical stress (Hu et al.,
2008). Although nitrogen or silicon starvation does not seem to
cause lipid accumulation in algae biofilms, perhaps these other
stresses may lead to significant accumulation of neutral lipid in al-
gae biofilm systems. Moreover, although biofilm starvation did not
achieve lipid accumulation, but rather, caused significant slough-
ing, perhaps different carbon to nitrogen ratios can be explored
as a means to accumulate neutral lipids, as was demonstrated by
Piorreck et al. (1984) using planktonic systems. These non-nutrient
starved carbon to nitrogen ratios may not cause growth cessation
and sloughing observed in completely starved cultures.
The lower quantities of lipids for the biofilms reported here may
be a result of several things. First, the biomass could be signifi-
cantly composed of low-lipid containing biomass, resulting in low-
er overall lipid percentages. Second, neutral and non-neutral lipids
are often not differentiated in the literature, thus total lipids are of-
ten reported. Finally, researchers sometimes use gravimetric meth-
ods of lipids determination, which is controversial because other
extracted compounds wrongfully get weighed as lipids.
3.3. Algae biofilm biomass and lipid productivities and implications for
algal biofilm growth system for biofuels
Comparisons of the biomass and lipid productivities indicate
significant differences between the species studied (Fig. 5). The
biofilm biomass productivities were determined using linear
regression on the data before nutrient starvation began on day
14. For the first 14 days of growth, prior to nutrient starvation,
the bioreactors inoculated with N. palea had significantly greater
biomass productivities than the bioreactors inoculated with S. obli-
quus – 2.8 g/m2/d compared to 2.1 g/m2/d. Due to the higher
Fig. 5. Comparing biomass and lipid productivities of N. palea and S. obliquus at
shear rates of 7 s 1.
 P.J. Schnurr et al. / Bioresource Technology 136 (2013) 337–344
biomass productivities, and because N. palea was found to have
twice the lipid concentrations than that of S. obliquus – 15% and
8%, respectively – N. palea had significantly higher lipid productiv-
ities – 0.45 g/m2/d compared to 0.18 g/m2/d.
N. palea biomass productivity rates may be higher than S. obli-
quus because it has a high propensity to grow in a sessile state
(Congestri et al., 2006). Besemer et al. (2007) showed that diatom
species compose approximately 60–80% of the biomass of biofilms,
supporting the claim that N. palea, being a diatom, might outper-
form S. obliquus. Additionally, N. palea may form biofilms that are
more robust, due to its propensity to form biofilms and flocs. This
robustness is also consistent with the N. palea biofilms being less
prone to gradual cell detachment and nutrient induced sloughing
compared to S. obliquus.
In the planktonic state N. palea is known for having higher lipid
concentrations than S. obliquus, thus the higher lipid concentra-
tions in biofilms are anticipated (Opute, 1974; Shifrin and Chis-
holm, 1981). At the higher biofilm biomass productivities, and
higher lipid concentrations within these biofilms, significantly
greater lipid productivities resulted for N. palea than for S. obliquus
(Fig. 5).
The results have implications for photobioreactor scale-up for
the production of biodiesel – starvation of biofilms may not be
suitable for optimizing lipid yields. Instead, optimizing lipid con-
centrations through species control – using species’ with high lipid
concentrations and biofilm biomass productivities – and optimiz-
ing growth kinetics through physical and chemical growth param-
eters, may be the preferred strategy. As mentioned, other forms of
biofilm stress may also be explored to accumulate lipids in algae
biofilms.
Another important implication of the results of these studies is
that knowledge on planktonic algae systems may not be directly
transferable to algae biofilm systems. Although it was believed
that knowledge on planktonic systems would be transferable to
biofilm systems i.e. starvation would cause lipid accumulation,
the results prove this is not the case. Because biofilms are unique
with distinct (micro) environments compared to planktonic
systems, it may be important to broaden algae biofilm studies to
understand how they behave independent of the knowledge of
planktonic systems.
The lipid productivities of 0.18–0.45 g/m2/d for the algae bio-
films grown at the lab scale are comparable to the canola/rapeseed
oil value cited by Mata et al. (2010) at 0.25 g/m2/d. However, it is
important to note that terrestrial crops are constrained by area of
land for growth, and that the value cited by Mata et al. (2010) is
on a per area of land basis. Conversely, the per unit area of algae
biofilms cited in the (algae) research above is on substratum area
which can have orders of magnitude more surface area per area
of land due to vertical orientation of PBR’s. Therefore, it is believed
that bioreactors based on biofilms that have significantly higher
productivities than crops, with the added advantage of improved
dewatering and downstream processing, can be developed.
4. Conclusions
S. obliquus and N. palea biofilms do not accumulate more lipids
when exposed to nutrient deficient conditions, but rather, growth
kinetics are severely compromised, and/or the biomass sloughs off
into suspension. N. palea forms more robust biofilms with higher
biomass and lipid productivities than S. obliquus. The observed lin-
ear growth curves indicate some growth limitation, likely associ-
ated with light attenuation and/or the transport of nutrients
across the film. Algae biofilm lipid productivities compare very
favorably to those of conventional terrestrial crops, while simulta-
neously having several other potential advantages over these con-
ventional biofuel sources.
343
Acknowledgements
The authors would like to thank the Natural Science and Engi-
neering Research Council of Canada (NSERC) for financial support
in the form of a Strategic Grant, and a Canadian Graduate Scholar-
ship (CGS D). Additional thanks to the Ontario Ministry of Training
College and Universities for an Ontario Graduate Scholarship. Spe-
cial acknowledgement and thanks to Pond Biofuels Inc. for helping
develop some lab technologies used in these experiments.
References
Ames, G.F., 1968. Lipids of Salmonella typhilmurium and Escherichia coli: structure
and metabolism. Journal of Bacteriology 95 (3), 833–843.
Barowitzka, M.A., 1992. Algal biotechnology products and processes – matching
science and economies. Journal of Applied Phycology 4, 267–279.
Besemer, K., Singer, G., Limberger, R., Chlup, A.K., Hochedlinger, G., Hodl, I., Baranyi,
C., Battin, T.J., 2007. Biophysical control on community succession in stream
biofilms. Applied & Environmental Microbiology 73 (15), 4966–4974.
Chisti, Y., 2007. Biodiesel from microalgae. Biotechnology Advances 25, 294–306.
Christenson, L., Sims, R., 2011. Production and harvesting of microalgae for
wastewater treatment, biofuels, and bioproducts. Biotechnology Advances 29,
686–702.
Christenson, L., Sims, R., 2012. Rotating algal biofilm reactor and spool harvester for
wastewater treatment with biofuels by-products. Biotechnology and
Bioengineering 109 (7), 1674–1684.
Congestri, R., Di Pippo, F.D., De Phillippis, R., Buttino, I., Paradossi, G., Albertano, P.,
2006. Seasonal succession of phototrophic biofilms in an Italian wastewater
treatment plant: biovolume, spatial structure and exopolysaccharides. Aquatic
Microbial Ecology 45, 301–312.
Davies, D.G., 1999. Regulation of matrix polymer in biofilm formation and
dispersion. In: Wingender, J., Neu, T.R., Flemming, H.-C. (Eds.), Microbial
Extracellular Polymeric Substances: Characterization, Structure and Function.
Spring, Berlin.
Decho, A.W., 1999. Chemical communication within microbial biofilms: chemotaxis
and quorum sensing in bacterial cells. In: Wingender, J., Neu, T.R., Flemming, H.-
C. (Eds.), Microbial Extracellular Polymeric Substances: Characterization,
Structure and Function. Spring, Berlin.
Flemming, H.C., Neu, T.R., Wozniak, D.J., 2007. The EPS matrix: the house of biofilm
cells. Journal of Bacteriology 189 (22), 7945–7947.
Gordon, J.M., Polle, J.E.W., 2007. Ultrahigh bioproductivity from algae. Applied
Microbiology and Biotechnology 76, 969–975.
Gudin, C., Therpenier, C., 1986. Bioconversion of solar energy into organic chemicals
by microalgae. Advanced Biotechnology Processes 6, 73–110.
Hu, Q., Sommerfeld, M., Jarvis, E., Ghirardi, M., Posewitz, M., Seibert, M., Darzins, A.,
2008. Microalgal triacylglycerols as feedstocks for biofuel production:
perspectives and advances. The Plant Journal 54, 621–639.
Irving, T.E., Allen, D.G., 2011. Species and material considerations in the formation
and development of microalgal biofilms. Applied Microbiology and
Biotechnology 92 (2), 283–294.
Johnson, M.B., Wen, Z., 2010. Development of an attached microalgal growth system
for biofuel production. Applied Microbiology & Biotechnology 85, 525–534.
Laspidou, C.S., Rittmann, B.E., 2002. A unified theory for extracellular polymeric
substances, soluble microbial products, and active and inert biomass. Water
Research 36, 2711–2720.
Li, Y., Horsman, M., Wu, N., Lan, C.Q., Dubois-Calero, N., 2008. Biofuels from
microalgae. Biotechnology Progress 24, 815–820.
Madigan, M.T., Martinko, J.M., Dunlap, P.V., Clark, D.P., 2009. Brock: Biology of
Microorganisms. Pearson Benjamin Cummings Inc.
Mata, T.M., Martins, A.A., Caetano, N.S., 2010. Microalgae for biodiesel production
and other applications: a revew. Renewable and Sustainable Energy Reviews 14,
217–232.
Neu, T.R., Lawrence, J.R., 1999. In situ characterization of extracellular polymeric
substances (EPS) in biofilm systems. In: Wingender, J., Neu, T.R., Flemming, H.-
C. (Eds.), Microbial Extracellular Polymeric Substances: Characterization,
Structure and Function. Spring, Berlin.
Opute, F.I., 1974. Studies on fat accumulation in Nitzschia palea kutz. Annals of
Botany 38, 889–902.
Ozkan, A., Kinney, K., Katz, L., Berberoglu, H., 2012. Reduction of water and energy
requirement of algae cultivation using an algae biofilm photobioreactor.
Bioresource Technology 114, 542–548.
Piorreck, M., Baasch, K.-H., Pohl, P., 1984. Biomass production, total protein,
chlorophylls, lipids and fatty acids of freshwater green and blue-green algae
under different nitrogen regimes. Phytochemistry 23 (2), 207–216.
Rodolfi, L., Zittelli, G.C., Bassi, N., Padovani, G., Biondi, N., Bonini, G., Tredici, M.R.,
2009. Microalgae for oil: strain selection, induction of lipid synthesis and
outdoor mass cultivation in a low-cost photobioreactor. Biotechnology and
Bioengineering 102, 100–112.
Roessler, P.G., 1988. Effects of silicon deficiency on lipid composition and
metabolism in the diatom Cyclotella cryptica. Journal of Phycology 24, 394–400.
Shifrin, N., Chisholm, S.W., 1981. Phytoplankton lipids: interspecific differences and
effects of nitrate, silicate and light-dark cycles. Journal of Phycology 17, 374–
384.
344 P.J. Schnurr et al. / Bioresource Technology 136 (2013) 337–344
Sicko-Goad, L., Anderson, N.A., 1991. Effect of growth and light/dark cycles on
diatom lipid content and composition. Journal of Phycology 27, 710–718.
Smid, I., Salfinger, M., 1994. Microbial identification by computer-aided gas-liquid
chromatography. Diagnostic Microbiology & Infectious Disease 19 (2), 81–88.
Spilling, K., Tamminen, T., Andersen, T., Kremp, A., 2010. Nutrient kinetics modeled
from time series of substrate depletion and growth: dissolved silicate uptake of
Baltic Sea spring diatoms. Marine Biology 157, 427–436.
Sriharan, S., Bagga, D., Nawaz, M., 1991. The effects of nutrient and temperature on
biomass, growth, lipid production, and fatty acid composition of Cyclotella
cryptica Reimann, Lewin and Guillard. Applied Biochemistry & Biotechnology 28
(29), 317–326.
Uduman, N., Qi, Y., Danquah, M.K., 2010. Dewatering of microalgal cultures: a major
bottleneck to algae based fuels. Journal of Renewable and Sustainable Energy 2,
012701.
View publication stats
Wang, C.Y., Fu, C.C., Liu, Y.C., 2007. Effects of using light-emitting diodes on the
cultivation of Spirulina platensis. Biochemical Engineering Journal 37, 21–25.
Wetzel, R.G., 1993. Microcommunities and microgradients: linking nutrient
regerenation, microbial mutualism, and high sustained aquatic primary
production. Netherlands Journal of Aquatic Ecology 27 (1), 3–9.
Wingender, J., Neu, T.R., Flemming, H.-C., 1999. What are bacterial extracellular
polymeric substances. In: Wingender, J., Neu, T.R., Flemming, H.-C. (Eds.),
Microbial Extracellular Polymeric Substances: Characterization, Structure and
Function. Spring, Berlin.
Wolfaardt, G.M., Lawrence, J.R., Korber, D.R., 1999. Function of EPS. In: Wingender,
J., Neu, T.R., Flemming, H.-C. (Eds.), Microbial Extracellular Polymeric
Substances: Characterization Structure and Function. Spring, Berlin.