Journal of Biotechnology 179 (2014) 8–14
Contents lists available at ScienceDirect
Journal of Biotechnology journal homepage: www.elsevier.com/locate/jbiotec
Homeoviscous response of Clostridium pasteurianum to butanol toxicity during glycerol fermentation Keerthi P. Venkataramanan a,1 , Yogi Kurniawan b,1 , Judy J. Boatman c , Cassandra H. Haynes c , Katherine A. Taconi c , Lenore Martin d , Geoffrey D. Bothun b,∗ , Carmen Scholz e,∗∗ a
Biotechnology Science and Engineering Program, University of Alabama in Huntsville, 301 Sparkman Dr NW, MSB 333, Huntsville, AL 35899, USA Department of Chemical Engineering, University of Rhode Island, 16 Greenhouse Rd., 205 Crawford Hall, Kingston, RI 02881, USA c Department of Chemical and Materials Engineering, University of Alabama in Huntsville, 301 Sparkman Dr NW, EB 130, Huntsville, AL 35899, USA d Department of Cell and Molecular Biology, University of Rhode Island, 120 Flagg Rd., 393 CBLS, Kingston, RI 02881, USA e Department of Chemistry, University of Alabama in Huntsville, 301 Sparkman Dr NW, MSB 333, Huntsville, AL 35899, USA b
a r t i c l e
i n f o
Article history: Received 6 December 2013 Received in revised form 5 February 2014 Accepted 7 March 2014 Available online 15 March 2014 Keywords: Clostridium pasteurianum Glycerol Fermentation Butanol Lipid composition Homeoviscous response
a b s t r a c t Clostridium pasteurianum ATCC 6013 achieves high n-butanol production when glycerol is used as the sole carbon source. In this study, the homeoviscous membrane response of C. pasteurianum ATCC 6013 has been examined through n-butanol challenge experiments. Homeoviscous response is a critical aspect of n-butanol tolerance and has not been examined in detail for C. pasteurianum. Lipid membrane compositions were examined for glycerol fermentations with n-butanol production, and during cell growth in the absence of n-butanol production, using gas chromatography–mass spectrometry (GC–MS) and proton nuclear magnetic resonance (1 H-NMR). Membrane stabilization due to homeoviscous response was further examined by surface pressure–area (–A) analysis of membrane extract monolayers. C. pasteurianum was found to exert a homeoviscous response that was comprised of an increase lipid tail length and a decrease in the percentage of unsaturated fatty acids with increasing n-butanol challenge. This led to a more rigid or stable membrane that counteracted n-butanol fluidization. This is the first report on the changes in the membrane lipid composition during n-butanol production by C. pasteurianum ATCC 6013, which is a versatile microorganism that has the potential to be engineered as an industrial n-butanol producer using crude glycerol. © 2014 Published by Elsevier B.V.
1. Introduction Clostridium pasteurianum ATCC 6013 ferments glycerol anaerobically as the sole substrate, resulting in a mixture of butanol, ethanol, and 1,3-propanediol (PDO) along with acetate and butyrate (Biebl, 2001; Dabrock et al., 1992; Taconi et al., 2009; Venkataramanan et al., 2012). C. pasteurianum is of particular
∗ Corresponding author at: Department of Chemical Engineering, University of Rhode Island, 16 Greenhouse Rd., 205 Crawford Hall, Kingston, RI 02881, USA. Tel.: +1 401 874 9518; fax: +1 401 874 4689. ∗∗ Corresponding author at: s at: Department of Chemistry, University of Alabama in Huntsville, 301 Sparkman Dr NW, MSB 333, .Huntsville, AL 35899, USA. Tel.:+1 256 824 6188; fax: +1 256 824 6349. E-mail addresses: [email protected] (G.D. Bothun), [email protected] (C. Scholz). 1 Authors KPV and YK contributed equally and are co-first authors. http://dx.doi.org/10.1016/j.jbiotec.2014.03.017 0168-1656/© 2014 Published by Elsevier B.V.
interest for n-butanol production, as it has also been shown to ferment biodiesel-derived crude glycerol (Taconi et al., 2009; Venkataramanan et al., 2012). However, n-butanol is toxic to cells, as it partitions into the cell membrane and affects both the structural and functional integrity of the cell. n-Butanol has a log P value of 0.8 and is considered to be one of the most toxic fermentation products (Sardessai and Bhosle, 2002). When Clostridia are exposed to solvents, the solvents have a fluidizing effect on the phospholipid bilayer, which then causes the organism to alter the lipid composition of the bilayer. This response of bacteria to tolerate toxic solvents by altering the composition of the lipid bilayer is known as homeoviscous response (membrane viscosity is proportional to fluidity). To compensate for the fluidizing effects of n-butanol, Clostridia increase the ratio of saturated to unsaturated fatty acids (SFA/UFA) in the lipid membrane, thereby reducing the fluidity of the membrane and increasing n-butanol tolerance (Baer et al., 1987, 1989; Baut et al., 1994; Bowles and Ellefson, 1985;
K.P. Venkataramanan et al. / Journal of Biotechnology 179 (2014) 8–14
Heipieper et al., 1994; Isar and Rangaswamy, 2012; L˘az˘aroaie, 2009; Lepage et al., 1987; Liu and Qureshi, 2009; Ramos et al., 2002; Vollherbst-Schneck et al., 1984). The fluidity of the lipid membrane is indirectly proportional to the amount of saturated fatty acids in the tail of the lipid bilayer. Hence, the bacteria that tolerate more n-butanol have a much higher SFA/UFA ratio in the lipid bilayer (Liu and Qureshi, 2009). This has been observed to be an essential biophysical process of tolerating n-butanol in various n-butanolproducing organisms from the genus Clostridium (Baer et al., 1987, 1989; Baut et al., 1994; Isar and Rangaswamy, 2012; Lepage et al., 1987; Vollherbst-Schneck et al., 1984). It has been established that the composition and distribution of lipids in the cell membrane play an essential role not only in maintaining membrane stability, curvature and membrane fluidity but also in modulating protein function and insertion of membrane proteins into the membranes (Lingwood et al., 2009; Lopez-Montero et al., 2008; Simons and Ikonen, 1997). Membrane fluidity is also essential for maintaining the proper distribution and diffusion of embedded proteins in the membrane (Lenaz, 1987). Membrane lipid composition changes in response to alcohol toxicity with an increase in SFA/UFA has been observed in Clostridia, particularly C. acetobutylicum (Lepage et al., 1987) and C. thermocellum (Timmons et al., 2009). Among wild-type and ethanol-adapted (EA) C. thermocellum, the EA cells preserved the optimum level of fluidity in response to ethanol toxicity by increasing the fatty acids chain length of the lipid tails and SFA/UFA in the cell membrane; resulting in higher membrane rigidity. Contrastingly, an opposite effect (decreasing SFA/UFA) in response to alcohol toxicity has been observed in several other microorganisms including gram-positive species such as Lactobacillus heterohiochii, Lactobacillus homohiochii; and gram-negative species such as Escherichia coli and Saccharomyces cerevisiae (Beaven et al., 1982; Ingram, 1982; Uchida, 1974). Although, these responses include a decrease in SFA/UFA ratio, the elongation in fatty acid chains nonetheless drove the homeoviscous response. Thus, the homeoviscous response involves a change in the SFA/UFA ratio of fatty acids complemented by an increase in the length of the fatty acids in the membrane.n-Butanol has been shown to affect the membrane by increasing fluidity and hence reducing lipid ordering (Aguilar et al., 1996; Iiyama et al., 1992; Kurniawan et al., 2012; Lepage et al., 1987). Also, n-butanol can lead to the formation of interdigitated phases and phase separation (Cevc, 1982; Kurniawan et al., 2012). Overall, n-butanol can compromise cellular function of the membrane by altering cell fission, fusion, budding, vesicle formation and cell signaling (Cevc, 1982; Lobbecke and Ceve, 1995). The passive and active transport of substrates and products is also affected, along with the structure and function of integral membrane proteins. This can hinder the ability of the cell to maintain an internal pH and inhibits membrane-bound ATPases and the uptake of the carbon source (if present), which subsequently inhibits energy generation (Bowles and Ellefson, 1985). In this work the homeoviscous response of C. pasteurianum was examined through n-butanol challenge (added exogenous n-butanol) experiments during glycerol fermentation. Comparative experiments were conducted under conditions where there was no endogenous n-butanol production (reinforced clostridial media with glucose). Cell membrane composition, specifically lipid tail length and degree of lipid unsaturation, was determined from membrane extracts by gas chromatography–mass spectrometry (GC–MS) and proton nuclear magnetic resonance (1 H-NMR). Surface pressure–area isotherms were further conducted to link homeoviscous response to membrane rigidity. To our knowledge this is the first report of membrane lipid composition and homeoviscous response for C. pasteurianum.
9
2. Materials and methods 2.1. Materials All chemicals were purchased from Fisher Scientific or SigmaAldrich. The bacterial strain C. pasteurianum ATCC 6013 was purchased from American Type Culture Collection and the glycerol stock was maintained as described earlier (Venkataramanan et al., 2012). Deuterated chloroform (CDCl3 ) was obtained from Cambridge Isotope Laboratory. Synthetic lipids, 1,2Dipalmitoyl-sn-glycero-3-phosphocholine (DPPC, >99% purity) and 1,2-dioleoyl-sn-glycero-3-phosphocholine (DOPC, >99% purity) were obtained from Avanti Polar Lipids, Inc.
2.2. Fermentation and n-butanol challenge The effect of n-butanol on the bacterial growth, productivity, membrane composition was studied by adding butanol to Biebl media (Biebl, 2001) containing glycerol (25 g/L) as the sole carbon source (condition 1) or by adding n-butanol to reinforced clostridial media (RCM) containing glucose (5 g/L) (condition 2). All experiments were conducted at 37 ◦ C with 10% (v/v) inoculum pre-grown in RCM. C. pasteurianum cultures were initially grown in either Biebl or RCM media until mid-exponential phase growth (two days). These cultures were then split into thirds prior to the addition of exogenous n-butanol to provide replicate experiments. The effect of n-butanol was investigated by adding 0 to 1% (w/v) (0 g/L to 10 g/L) exogenous n-butanol to the media. Under condition 1, nbutanol concentrations reflect exogenous n-butanol and n-butanol produced endogenously by C. pasteurianum. In contrast, under condition 2, C. pasteurianum did not produce n-butanol when grown on glucose and the n-butanol concentrations reflected only exogenous n-butanol. The cells were allowed to continue growing for 24 h after exogenous n-butanol was added, at which point the fermentations were terminated and the cell membranes were extracted. Growth (optical density) was monitored using spectrophotometry absorbance measurement at 600 nm of the cultures and the substrate utilization and product formation was measured using HPLC as described earlier (Venkataramanan et al., 2012).
2.3. Extraction of the cell membrane Cell membranes were extracted using the modified protocol of Bligh and Dyer using dichloromethane/methanol mixtures (Bligh and Dyer, 1959). The cells were harvested (0.5 mL cell suspension) by centrifugation at 13,000 rpm for 15 min and the pellets were resuspended in 0.5 mL of sterile 1.0% (w/v) NaCl in a 10 mL glass sample tube with PTFE lined caps. To the resuspended pellets, 2 mL of dichloromethane/methanol mixture (1:2 v/v) was added and shaken vigorously for 15 min. Following a 2 h incubation at room temperature (RT), the resuspended pellets were centrifuged at 2500 rpm and the supernatant (S1) was collected in a fresh tube. The pellet was again resuspended in 0.5 mL 1.0% NaCl and 2 mL of a dichloromethane/methanol mixture (2:1 v/v) was added to the resuspended pellet and shaken vigorously for 15 min. Following a 2 h incubation, the samples were centrifuged at 2500 rpm and the supernatant (S2) was collected. Supernatants S1 and S2 were combined and 1 mL of dichloromethane and 1 mL of sterile 1.0% NaCl were added. The top phase (aqueous) was removed and the bottom (organic) phase was retained. The solvent was evaporated under a gentle nitrogen stream. Once a dry film was obtained, the headspace was flushed with nitrogen, capped tightly, and stored for further analysis.
10
K.P. Venkataramanan et al. / Journal of Biotechnology 179 (2014) 8–14
2.4. GC–MS analysis of lipid composition Membrane extracts were pooled together for each condition and the lipids were methylated for GC/MS analysis. The dried lipid samples were saponified using 1 mL of 3 N sodium hydroxide at 90 ◦ C for an hour and then cooled to RT. The excess sodium hydroxide was neutralized with 1.8 mL of 3.6 N hydrochloric acid at 90 ◦ C for 10 min and cooled to RT. The free fatty acids were extracted using 1 mL of hexane and diethyl ether (1:1 v/v). The organic phase was then separated into a dry round bottom flask. The fatty acid hydrolyzates were dried using a rotary evaporator and stored under Argon in desiccators. The dried fatty acid hydrolyzates were derivatized using a 5 mL borontrifluoride–methanol complex at 60 ◦ C for 5 min and then cooled to RT. To the cooled solution, 1 mL water and 1 mL hexane were added and the container shaken multiple times to ensure the transfer of esters into the non-polar solvent. The upper organic layer was removed and transferred into an Erlenmeyer flask containing 5 g of anhydrous sodium sulfate. The flask was incubated at RT to dry overnight. The sodium sulfate was filtered and the hexane solution was transferred into a round bottom flask and dried in a rotary evaporator and the samples were stored under Argon. For GC–MS analysis, samples were dissolved in 250 L of dichloromethane. Fatty acid methyl esters (FAME) derivatives were analyzed by a Shimadzu GC–MS QP 2010 system using a SHR5xLB silica capillary column (30 m × 0.25 mm ID, composed of 100% dimethyl polysiloxane). The compounds were identified by comparison with the data in the NIST libraries. The FAME distribution was determined as FAME (%) =
A
FAME
AFAME
(1)
where AFAME is the peak area associated with a single FAME species and AFAME is the total area of all FAME species. 2.5. NMR analysis The dry film of the membrane was dissolved in CDCl3 for analysis. Synthetic lipids, DPPC and DOPC were used as standards for NMR analysis. Various ratios of DPPC and DOPC (1:0, 3:1, 1:1, 1:3 and 0:1) in a total lipid concentration of 10 mM were used for calibration (Supplementary material and Fig. S1). The synthetic lipids were dried and dissolved in CDCl3 . All 1 H-NMR spectra were recorded on a VarianTM Unity Inova 500 (500 MHz) spectrometer equipped with a 5 mm triple resonance inverse detectable probe. The percentage of unsaturation in the membrane lipid samples was calculated from the integration ratios (I) of the olefinic hydrogen signals (at 5.31 ppm) as
1 H-NMR
% unsaturation =
Iolefinic . Itotal
(2)
2.6. Surface pressure–area (–A) isotherms –A isotherms were recorded using a Langmuir trough (model 102 M, Nima technology Ltd, UK) with a deposition area of 70 cm2 at 25 ◦ C, 37 ◦ C and at 50 ◦ C. A Wilhelmy plate connected to an electronic micro-balance was used to measure surface pressure with an accuracy of ± 1 N/m. The external water bath system was used to control the subphase temperature. Monolayers were obtained by spreading diluted solutions of lipid samples in chloroform at the air/water interface using a 5 L syringe. The chloroform was allowed to evaporate and the lipid films spread at air/water within 15 min. The system was the equilibrated for 15 min before being compressed at a speed of 10 cm2 /min. Mechanical properties of the membrane extract monolayer films were determined by the elastic modulus, Cs −1 , which is the
reciprocal of the area compressibility (Duncan and Larson, 2008). The elastic modulus corresponds to the elasticity of the monolayers under the compression force. Cs−1 values were calculated as
Cs−1 = −A
∂p ∂A
.
(3)
T
where, A is the area per membrane extract weight (cm2 /g) at surface pressure under isothermal conditions. 3. Results and discussion 3.1. Glycerol fermentation and n-butanol challenge The effect of n-butanol stress on the membrane composition was studied in a semi-defined Biebl media at 37 ◦ C, with glycerol as the sole carbon source (condition 1). The fermentation of glycerol is typically biphasic in nature, with an acidogenic (acetate, butyrate and 1,3-PDO) exponential phase followed by a solventogenic (nbutanol) late exponential and stationary phase. C. pasteurianum produces n-butanol mainly when glycerol is the carbon source. C. pasteurianum cultures do not undergo solventogenesis during the fermentation of glucose, as the fermentation is predominantly only in the acidogenic phase, resulting in butyric acid as the major fermentation product with no n-butanol formation (condition 2). Results for glycerol fermentation (condition 1) are shown in Fig. 1 during n-butanol challenge. Exogenous n-butanol was added at concentrations of 0 g/L (control), 2.5 g/L (0.0335 M), 5 g/L (0.067 M), 7.5 g/L (0.105 M) and 10 g/L (0.134 M) to cultures in midexponential phase growth, grown in static flasks. The endogenous concentration of butanol of at t = 0 was 1.65 g/L. After the addition of exogenous butanol, the total concentration at t = 0 was 1.65 g/L, 3.14 g/L, 5.8 g/L, 8.8 g/L and 11.53 g/L, respectively. The concentrations of exogenous n-butanol (0 to 10 g/L) were selected based on n-butanol produced during the consumptions of glycerol, which is between 5 g/L and 50 g/L (glycerol concentration) based on previous studies (Biebl, 2001; Dabrock et al., 1992; Jensen et al., 2012a,b; Taconi et al., 2009; Venkataramanan et al., 2012). The optical densities indicated that cell growth was nearly identical across cell cultures prior to the addition of n-butanol (identified as t = 0 h) (Fig. 1A). After 24 h, there was a clear reduction in optical density with increasing exogenous n-butanol concentrations indicative of cell growth inhibition. However, analysis of glycerol utilization indicates that the bacteria continued to consume glycerol despite this growth inhibition (Fig. 1B). The total n-butanol concentrations at 0 h and 24 h, which reflected the combination of endogenous and exogenous n-butanol within the media, are shown in Fig. 1C. After 24 h the total n-butanol concentration decreased with increasing exogenous n-butanol concentration. This is more clearly depicted in the change in product concentrations from 0 h to 24 h, [product], shown in Fig. 1D. Adding exogenous n-butanol did not significantly change the production of acetate, ethanol, or butyrate; but it did lead to decreases in the production of 1,3-propanediol and n-butanol. C. pasteurianum were able to tolerate exogenous n-butanol up to 2.5 g/L, but beyond that concentration the amount of exogenous n-butanol became intolerable and the ability to produce solvents decreased proportionately. For n-butanol, there is a negative change in concentration and, at 10 g/L exogenous n-butanol, the total concentration of butanol in the media at 24 h is less than that of the exogenous n-butanol added at 0 h. This apparent loss of n-butanol can be attributed to n-butanol partitioning from the media, where samples are collected for product analysis, into the cell membrane. Collectively, the fermentation results show that adding exogenous
K.P. Venkataramanan et al. / Journal of Biotechnology 179 (2014) 8–14
11
Fig. 1. Glycerol fermentation (25 g/L) by C. pasteurianum. (A) Cell growth determined by optical density as a function of fermentation time. Arrows at 2 and 3 days denote when exogenous n-butanol was added (0 h) and when the fermentation was terminated and membranes were extracted post-stress (24 h), respectively. (B) Glycerol utilization as a function of exogenous n-butanol concentration. (C) Total n-butanol concentration, [n-butanol], in the media (endogenous + exogenous) as a function of exogenous n-butanol concentration. (D) Change in product concentrations, [product] from 0 h to 24 h. Error bars for 24 h samples represent standard deviation of triplicate experiments. Lines are shown to guide the eye and depict trends.
n-butanol inhibited cell growth and solvent production. However, exogenous n-butanol did not inhibit glycerol consumption, which suggests that the bacteria utilized the substrate for maintenance functions (i.e. counteracting n-butanol toxicity) and energy production. 3.2. Homeoviscous response due to n-butanol toxicity Homeoviscous response with n-butanol challenge (condition 1, n-butanol production) was determined based on GC–MS analysis of C. pasteurianum membranes extracted after 24 h. Results were compared to the homeoviscous response with n-butanol challenge of C. pasteurianum grown in RCM (condition 2, no n-butanol production).
For comparative purposes and ease of viewing, only the GC–MS results at 0 g/L and 10 g/L exogenous n-butanol are shown in Fig. 2. Results at all exogenous n-butanol concentrations are provided in the Supporting information (Fig. S1). With glycerol fermentation, the homeoviscous response involved progressive increases in the lipid tail length and decreases in the percentage of unsaturated lipids with increasing exogenous n-butanol concentration (Figs. 2A and 3A, respectively). At 0 g/L n-butanol the membrane was composed of lipids with tail lengths less than C16 (palmitoyl), while at 10 g/L the tails lengths were greater than C16 , and predominately C18 (stearoyl) or higher (Fig. 2A). The longer tailed lipids (C19 to C22 ) were almost completely absent when no external n-butanol
Fig. 2. Membrane fatty acid distribution, determined from GC–MS as fatty acid methyl esters (FAME), at 0 g/L and 10 g/L exogenous n-butanol concentrations. (A) FAME distribution at 25 g/L glycerol, corresponding to fermentation results in Fig. 1, and (B) in RCM media containing 5 g/L glucose.
12
K.P. Venkataramanan et al. / Journal of Biotechnology 179 (2014) 8–14
Fig. 3. Percentage of unsaturated lipids, determined as FAME via GC–MS and 1 H-NMR, in C. pasteurianum membrane extracts as a function of exogenous n-butanol concentration during (A) glycerol fermentation and (B) in RCM media.
was present. In conjunction with increasing tail length, the percentage of unsaturated lipids determined by GC–MS and 1 HNMR decreased from with increasing n-butanol concentration (Fig. 3A). A markedly different response was observed when the bacteria were grown in RCM (condition 2) where n-butanol production was not possible. In this case there was minimal change in the lipid tail length from 0 g/L to 10 g/L exogenous n-butanol (Fig. 2B). Furthermore, the trend in percentage of unsaturated lipids was opposite of that for glycerol fermentation; increasing unsaturation with increasing n-butanol concentration (Fig. 3B). Comparing condition 1 (glycerol) and condition 2 (RCM) suggest that the homeoviscous response of C. pasteurianum to elevated n-butanol concentrations only occurs when n-butanol is being produced. Hence, biophysical changes in membrane structure alone, which would have occurred due to n-butanol partitioning in condition 2, were not sufficient to drive an increase in tail length based on the conditions examined, but did yield an increase in the degree of lipid unsaturation. While the effects of n-butanol on membrane composition have not been reported for C. pasteurianum, they have been reported for C. acetobutylicum fermenting glucose (Baer et al., 1987, 1989; Lepage et al., 1987; Vollherbst-Schneck et al., 1984). In this case, C. acetobutylicum has been shown to tolerate n-butanol through a homeoviscous response that predominantly involves an increase in the SFA/UFA ratio within the lipid membrane (Baer et al., 1987, 1989; Lepage et al., 1987; Vollherbst-Schneck et al., 1984). Lepage et al. (1987) have reported that the lipid composition of C. acetobutylicum consists primarily of fatty acids from
Fig. 4. Weight averaged fatty acid tail length as a function of the change in n-butanol concentration due to the addition of exogenous n-butanol.
C12 to C19 . The percentage of unsaturated fatty acids was close to 50% in the absence of n-butanol challenge, and reduced to 46.5% and 43.5% upon exposure to 4 g/L and 8 g/L n-butanol, respectively. At 8 g/L n-butanol, this reflects a 13% decrease in UFA. Comparatively, C. pasteurianum has a lower percentage of unsaturated lipids in its membrane (∼12%), but the decrease in UFA was ∼25% upon exposure to 10 g/L n-butanol. Increases
Fig. 5. (A) Surface pressure–area (–A) isotherms of reconstituted membrane monolayers at 37 ◦ C at air/water interfaces. (B) Elastic modulus determined from –A isotherms. The legends show exogenous n-butanol concentrations corresponding to the fermentation conditions from which the membranes were extracted.
K.P. Venkataramanan et al. / Journal of Biotechnology 179 (2014) 8–14
in lipid acyl tail length were also observed for C. pasteurianum and there is a clear relationship between the average tail length and the change in n-butanol production (Fig. 4). A decrease in n-butanol production due to n-butanol toxicity correlates with an increase tail length. Hence, increases in both the SFA/UFA ratio and lipid tail length comprise the homeoviscous response of C. pasteurianum. 3.3. –A isotherms of membrane extract monolayers The homeoviscous response observed for C. pasteurianum counteracted n-butanol fluidization (membrane disordering) via increased lipid tail length. Increasing tail length increases the attractive van der Waals interactions between lipids and leads to a more rigid, stable membrane. Membrane extract monolayers spread at air/water interfaces were examined to gain a better understanding of this process. Membrane lipids within the monolayers were oriented with the hydrophilic polar headgroups in water and the hydrophobic hydrocarbon tails extending into air. –A isotherms and elastic moduli of membrane monolayers are plotted in Fig. 5. The temperature was maintained at 37 ◦ C consistent with the fermentation temperature. Because the average lipid molecular weight could not be determined, the areas are expressed as area per mass of membrane extract (cm2 /g) rather than area per molecule, which is typically used for lipid monolayers. As the monolayers were compressed the surface pressure increased with decreasing area per mass due to greater lipid packing at the interface. There were no distinct plateau regions within the –A isotherms, suggesting that the monolayers did not undergo two-dimensional phase transitions at the conditions examined (Fig. 5A). Increasing n-butanol concentration shifted the isotherms to higher surface pressures, which can be attributed to the longer lipid tail lengths due to the homeoviscous response. Higher molecular weight lipids with longer tails occupy more area at the air/water interface. The monolayers also exhibited higher elastic moduli with increasing n-butanol concentration, reflecting a more rigid or less compressible membrane (Fig. 5B). 4. Conclusions This study is the first to report of the lipid composition in C. pasteurianum ATCC 6013, an anaerobic spore-forming bacteria, ferments glycerol as the sole substrate, resulting in a mixture of butanol, ethanol, and 1,3-propanediol (PDO) along with acetate and butyrateand the homeoviscous response of C. pasteurianum due to elevated n-butanol concentrations (n-butanol challenge). The analysis of the extracted membranes from cells exposed to various concentrations of n-butanol under two different conditions, (i) n-butanol production by glycerol fermentation (condition 1) and (ii) no n-butanol production in RCM (condition 2), resulted in two completely different responses with respect to the changes in lipid composition. Only when C. pasteurianum is producing n-butanol is it able to respond to elevated n-butanol concentrations. When stressed with exogenous n-butanol during n-butanol production, C. pasteurianum responded by increasing lipid tail length and the ratio of saturated to unsaturated fatty acids. Despite n-butanol toxicity, which led to cell growth inhibition and reduced solvent production, the bacteria continued to utilize glycerol to achieve this response. –A isotherms confirmed that the homeoviscous response increased the stability of the membrane to counteract n-butanol disordering. The homeoviscous response was similar to other n-butanol producing species of Clostridia, as well as other microbes that respond to tolerate toxic organic compounds. It is envisioned that these results can be coupled with proteomics,
13
functional genomics and metabolic engineering to determine the mechanism of homeoviscous response and for the future development of n-butanol-tolerant industrial strains. Acknowledgments This project was supported by NSF grant (CBET-0966846). We sincerely thank Dr. Vinka Oyanedel-Craver and Varun Kasaraneni at the University of Rhode Island for their assistance with GC–MS analysis. References Aguilar, L.F., Sotomayor, C.P., Lissi, E.A., 1996. Main phase transition depression by incorporation of alkanols in DPPC vesicles in the gel state: influence of the solute topology. Colloids Surf., A 108, 287–293. Baer, S.H., Blaschek, H.P., Smith, T.L., 1987. Effect of butanol challenge and temperature on lipid-composition and membrane fluidity of butanoltolerant Clostridium-acetobutylicum. Appl. Environ. Microbiol. 53, 2854– 2861. Baer, S.H., Bryant, D.L., Blaschek, H.P., 1989. Electron-spin resonance analysis of the effect of butanol on the membrane fluidity of intact-cells of Clostridiumacetobutylicum. Appl. Environ. Microbiol. 55, 2729–2731. Baut, F., Fick, M., Viriot, M.L., Andre, J.C., Engasser, J.M., 1994. Investigation of acetone–butanol–ethanol fermentation by fluorescence. Appl. Microbiol. Biotechnol. 41, 551. Beaven, M.J., Charpentier, C., Rose, A.H., 1982. Production and tolerance of ethanol in relation to phospholipid fatty-acyl composition in Saccharomyces-cerevisiae Ncyc 431. J. Gen. Microbiol. 128, 1447–1455. Biebl, H., 2001. Fermentation of glycerol by Clostridium pasteurianum—batch and continuous culture studies. J. Ind. Microbiol. Biotechnol. 27, 18–26. Bligh, E.G., Dyer, W.J., 1959. A rapid method of total lipid extraction and purification. Can. J. Biochem. Physiol. 37, 911–917. Bowles, L.K., Ellefson, W.L., 1985. Effects of butanol on Clostridium-acetobutylicum. Appl. Environ. Microbiol. 50, 1165–1170. Cevc, G., 1982. Water and membranes—the interdependence of their physicochemical properties in the case of phospholipid-bilayers. Stud. Biophys. 91, 45–52. Dabrock, B., Bahl, H., Gottschalk, G., 1992. Parameters affecting solvent production by Clostridium pasteurianum. Appl. Environ. Microbiol. 58, 1233–1239. Duncan, S.L., Larson, R.G., 2008. Comparing experimental and simulated pressurearea isotherms for DPPC. Biophys. J. 94, 2965–2986. Heipieper, H.J., Weber, F.J., Sikkema, J., Keweloh, H., Debont, J.A.M., 1994. Mechanisms of resistance of whole cells to toxic organic-solvents. Trends Biotechnol. 12, 409–415. Iiyama, S., Toko, K., Murata, T., Ichinose, H., Suezaki, Y., Kamaya, H., Ueda, I., Yamafuji, K., 1992. Cutoff effect of n-alkanols in an excitable model membrane composed of dioleyl phosphate. Biophys. Chem. 45, 91. Ingram, L.O., 1982. Regulation of fatty-acid composition in Escherichia-coli—a proposed common mechanism for changes induced by ethanol, chaotropic agents, and a reduction of growth temperature. J. Bacteriol. 149, 166–172. Isar, J., Rangaswamy, V., 2012. Improved n-butanol production by solvent tolerant Clostridium beijerinckii. Biomass Bioenergy 37, 9–15. Jensen, T.O., Kvist, T., Mikkelsen, M.J., Christensen, P.V., Westermann, P., 2012a. Fermentation of crude glycerol from biodiesel production by Clostridium pasteurianum. J. Ind. Microbiol. Biotechnol. 39, 709–717. Jensen, T.O., Kvist, T., Mikkelsen, M.J., Westermann, P., 2012b. Production of 1,3PDO and butanol by a mutant strain of Clostridium pasteurianum with increased tolerance towards crude glycerol. AMB Express 2, 44. Kurniawan, Y., Venkataramanan, K.P., Scholz, C., Bothun, G.D., 2012. n-Butanol partitioning and phase behavior in DPPC/DOPC membranes. J. Phys. Chem. B 116, 5919–5924. L˘az˘aroaie, M.M., 2009. Mechanisms involved in organic solvent resistance in Gramnegative bacteria. World Acad. Sci. Eng. Technol. 30, 643–653. Lenaz, G., 1987. Lipid fluidity and membrane protein dynamics. Biosci. Rep. 7, 823–837. Lepage, C., Fayolle, F., Hermann, M., Vandecasteele, J.P., 1987. Changes in membrane lipid-composition of Clostridium-acetobutylicum during acetone butanol fermentation—effects of solvents, growth temperature and Ph. J. Gen. Microbiol. 133, 103–110. Lingwood, D., Kaiser, H.J., Levental, I., Simons, K., 2009. Lipid rafts as functional heterogeneity in cell membranes. Biochem. Soc. Trans. 37, 955–960. Liu, S.Q., Qureshi, N., 2009. How microbes tolerate ethanol and butanol. New Biotechnol. 26, 117–121. Lobbecke, L., Ceve, G., 1995. Effects of short-chain alcohols on the phase-behavior and interdigitation of phosphatidylcholine bilayer-membranes. Biochim. Biophys. Acta, Biomembr. 1237, 59–69. Lopez-Montero, I., Arriaga, L.R., Monroy, F., Rivas, G., Tarazona, P., Velez, M., 2008. High fluidity and soft elasticity of the inner membrane of Escherichia coli
14
K.P. Venkataramanan et al. / Journal of Biotechnology 179 (2014) 8–14
revealed by the surface rheology of model Langmuir monolayers. Langmuir 24, 4065–4076. Ramos, J.L., Duque, E., Gallegos, M.-T., Godoy, P., Ramos-González, M.I., Rojas, A., Terán, W., Segura, A., 2002. Mechanisms of solvent tolerance in gram-negative bacteria. Annu. Rev. Microbiol. 56, 743–768. Sardessai, Y., Bhosle, S., 2002. Tolerance of bacteria to organic solvents. Res. Microbiol. 153, 263–268. Simons, K., Ikonen, E., 1997. Functional rafts in cell membranes. Nature 387, 569–572. Taconi, K.A., Venkataramanan, K.P., Johnson, D.T., 2009. Growth and solvent production by Clostridium pasteurianum ATCC® 6013TM utilizing biodiesel-derived crude glycerol as the sole carbon source. Environ. Prog. Sustainable Energy 28, 100–110.
Timmons, M.D., Knutson, B.L., Nokes, S.E., Strobel, H.J., Lynn, B.C., 2009. Analysis of composition and structure of Clostridium thermocellum membranes from wildtype and ethanol-adapted strains. Appl. Microbiol. Biotechnol. 82, 929–939. Uchida, K., 1974. Occurrence of saturated and mono-unsaturated fatty-acids with unusually long chains (C20–C30) in Lactobacillus heterohiochii, an alcoholophilic bacterium. Biochim. Biophys. Acta 348, 86–93. Venkataramanan, K.P., Boatman, J.J., Kurniawan, Y., Taconi, K.A., Bothun, G.D., Scholz, C., 2012. Impact of impurities in biodiesel-derived crude glycerol on the fermentation by Clostridium pasteurianum ATCC 6013. Appl. Microbiol. Biotechnol. 93, 1325–1335. Vollherbst-Schneck, K., Sands, J.A., Montenecourt, B.S., 1984. Effect of butanol on lipid composition and fluidity of Clostridium acetobutylicum ATCC 824. Appl. Environ. Microbiol. 47, 193–194.
Contents lists available at ScienceDirect
Journal of Biotechnology journal homepage: www.elsevier.com/locate/jbiotec
Homeoviscous response of Clostridium pasteurianum to butanol toxicity during glycerol fermentation Keerthi P. Venkataramanan a,1 , Yogi Kurniawan b,1 , Judy J. Boatman c , Cassandra H. Haynes c , Katherine A. Taconi c , Lenore Martin d , Geoffrey D. Bothun b,∗ , Carmen Scholz e,∗∗ a
Biotechnology Science and Engineering Program, University of Alabama in Huntsville, 301 Sparkman Dr NW, MSB 333, Huntsville, AL 35899, USA Department of Chemical Engineering, University of Rhode Island, 16 Greenhouse Rd., 205 Crawford Hall, Kingston, RI 02881, USA c Department of Chemical and Materials Engineering, University of Alabama in Huntsville, 301 Sparkman Dr NW, EB 130, Huntsville, AL 35899, USA d Department of Cell and Molecular Biology, University of Rhode Island, 120 Flagg Rd., 393 CBLS, Kingston, RI 02881, USA e Department of Chemistry, University of Alabama in Huntsville, 301 Sparkman Dr NW, MSB 333, Huntsville, AL 35899, USA b
a r t i c l e
i n f o
Article history: Received 6 December 2013 Received in revised form 5 February 2014 Accepted 7 March 2014 Available online 15 March 2014 Keywords: Clostridium pasteurianum Glycerol Fermentation Butanol Lipid composition Homeoviscous response
a b s t r a c t Clostridium pasteurianum ATCC 6013 achieves high n-butanol production when glycerol is used as the sole carbon source. In this study, the homeoviscous membrane response of C. pasteurianum ATCC 6013 has been examined through n-butanol challenge experiments. Homeoviscous response is a critical aspect of n-butanol tolerance and has not been examined in detail for C. pasteurianum. Lipid membrane compositions were examined for glycerol fermentations with n-butanol production, and during cell growth in the absence of n-butanol production, using gas chromatography–mass spectrometry (GC–MS) and proton nuclear magnetic resonance (1 H-NMR). Membrane stabilization due to homeoviscous response was further examined by surface pressure–area (–A) analysis of membrane extract monolayers. C. pasteurianum was found to exert a homeoviscous response that was comprised of an increase lipid tail length and a decrease in the percentage of unsaturated fatty acids with increasing n-butanol challenge. This led to a more rigid or stable membrane that counteracted n-butanol fluidization. This is the first report on the changes in the membrane lipid composition during n-butanol production by C. pasteurianum ATCC 6013, which is a versatile microorganism that has the potential to be engineered as an industrial n-butanol producer using crude glycerol. © 2014 Published by Elsevier B.V.
1. Introduction Clostridium pasteurianum ATCC 6013 ferments glycerol anaerobically as the sole substrate, resulting in a mixture of butanol, ethanol, and 1,3-propanediol (PDO) along with acetate and butyrate (Biebl, 2001; Dabrock et al., 1992; Taconi et al., 2009; Venkataramanan et al., 2012). C. pasteurianum is of particular
∗ Corresponding author at: Department of Chemical Engineering, University of Rhode Island, 16 Greenhouse Rd., 205 Crawford Hall, Kingston, RI 02881, USA. Tel.: +1 401 874 9518; fax: +1 401 874 4689. ∗∗ Corresponding author at: s at: Department of Chemistry, University of Alabama in Huntsville, 301 Sparkman Dr NW, MSB 333, .Huntsville, AL 35899, USA. Tel.:+1 256 824 6188; fax: +1 256 824 6349. E-mail addresses: [email protected] (G.D. Bothun), [email protected] (C. Scholz). 1 Authors KPV and YK contributed equally and are co-first authors. http://dx.doi.org/10.1016/j.jbiotec.2014.03.017 0168-1656/© 2014 Published by Elsevier B.V.
interest for n-butanol production, as it has also been shown to ferment biodiesel-derived crude glycerol (Taconi et al., 2009; Venkataramanan et al., 2012). However, n-butanol is toxic to cells, as it partitions into the cell membrane and affects both the structural and functional integrity of the cell. n-Butanol has a log P value of 0.8 and is considered to be one of the most toxic fermentation products (Sardessai and Bhosle, 2002). When Clostridia are exposed to solvents, the solvents have a fluidizing effect on the phospholipid bilayer, which then causes the organism to alter the lipid composition of the bilayer. This response of bacteria to tolerate toxic solvents by altering the composition of the lipid bilayer is known as homeoviscous response (membrane viscosity is proportional to fluidity). To compensate for the fluidizing effects of n-butanol, Clostridia increase the ratio of saturated to unsaturated fatty acids (SFA/UFA) in the lipid membrane, thereby reducing the fluidity of the membrane and increasing n-butanol tolerance (Baer et al., 1987, 1989; Baut et al., 1994; Bowles and Ellefson, 1985;
K.P. Venkataramanan et al. / Journal of Biotechnology 179 (2014) 8–14
Heipieper et al., 1994; Isar and Rangaswamy, 2012; L˘az˘aroaie, 2009; Lepage et al., 1987; Liu and Qureshi, 2009; Ramos et al., 2002; Vollherbst-Schneck et al., 1984). The fluidity of the lipid membrane is indirectly proportional to the amount of saturated fatty acids in the tail of the lipid bilayer. Hence, the bacteria that tolerate more n-butanol have a much higher SFA/UFA ratio in the lipid bilayer (Liu and Qureshi, 2009). This has been observed to be an essential biophysical process of tolerating n-butanol in various n-butanolproducing organisms from the genus Clostridium (Baer et al., 1987, 1989; Baut et al., 1994; Isar and Rangaswamy, 2012; Lepage et al., 1987; Vollherbst-Schneck et al., 1984). It has been established that the composition and distribution of lipids in the cell membrane play an essential role not only in maintaining membrane stability, curvature and membrane fluidity but also in modulating protein function and insertion of membrane proteins into the membranes (Lingwood et al., 2009; Lopez-Montero et al., 2008; Simons and Ikonen, 1997). Membrane fluidity is also essential for maintaining the proper distribution and diffusion of embedded proteins in the membrane (Lenaz, 1987). Membrane lipid composition changes in response to alcohol toxicity with an increase in SFA/UFA has been observed in Clostridia, particularly C. acetobutylicum (Lepage et al., 1987) and C. thermocellum (Timmons et al., 2009). Among wild-type and ethanol-adapted (EA) C. thermocellum, the EA cells preserved the optimum level of fluidity in response to ethanol toxicity by increasing the fatty acids chain length of the lipid tails and SFA/UFA in the cell membrane; resulting in higher membrane rigidity. Contrastingly, an opposite effect (decreasing SFA/UFA) in response to alcohol toxicity has been observed in several other microorganisms including gram-positive species such as Lactobacillus heterohiochii, Lactobacillus homohiochii; and gram-negative species such as Escherichia coli and Saccharomyces cerevisiae (Beaven et al., 1982; Ingram, 1982; Uchida, 1974). Although, these responses include a decrease in SFA/UFA ratio, the elongation in fatty acid chains nonetheless drove the homeoviscous response. Thus, the homeoviscous response involves a change in the SFA/UFA ratio of fatty acids complemented by an increase in the length of the fatty acids in the membrane.n-Butanol has been shown to affect the membrane by increasing fluidity and hence reducing lipid ordering (Aguilar et al., 1996; Iiyama et al., 1992; Kurniawan et al., 2012; Lepage et al., 1987). Also, n-butanol can lead to the formation of interdigitated phases and phase separation (Cevc, 1982; Kurniawan et al., 2012). Overall, n-butanol can compromise cellular function of the membrane by altering cell fission, fusion, budding, vesicle formation and cell signaling (Cevc, 1982; Lobbecke and Ceve, 1995). The passive and active transport of substrates and products is also affected, along with the structure and function of integral membrane proteins. This can hinder the ability of the cell to maintain an internal pH and inhibits membrane-bound ATPases and the uptake of the carbon source (if present), which subsequently inhibits energy generation (Bowles and Ellefson, 1985). In this work the homeoviscous response of C. pasteurianum was examined through n-butanol challenge (added exogenous n-butanol) experiments during glycerol fermentation. Comparative experiments were conducted under conditions where there was no endogenous n-butanol production (reinforced clostridial media with glucose). Cell membrane composition, specifically lipid tail length and degree of lipid unsaturation, was determined from membrane extracts by gas chromatography–mass spectrometry (GC–MS) and proton nuclear magnetic resonance (1 H-NMR). Surface pressure–area isotherms were further conducted to link homeoviscous response to membrane rigidity. To our knowledge this is the first report of membrane lipid composition and homeoviscous response for C. pasteurianum.
9
2. Materials and methods 2.1. Materials All chemicals were purchased from Fisher Scientific or SigmaAldrich. The bacterial strain C. pasteurianum ATCC 6013 was purchased from American Type Culture Collection and the glycerol stock was maintained as described earlier (Venkataramanan et al., 2012). Deuterated chloroform (CDCl3 ) was obtained from Cambridge Isotope Laboratory. Synthetic lipids, 1,2Dipalmitoyl-sn-glycero-3-phosphocholine (DPPC, >99% purity) and 1,2-dioleoyl-sn-glycero-3-phosphocholine (DOPC, >99% purity) were obtained from Avanti Polar Lipids, Inc.
2.2. Fermentation and n-butanol challenge The effect of n-butanol on the bacterial growth, productivity, membrane composition was studied by adding butanol to Biebl media (Biebl, 2001) containing glycerol (25 g/L) as the sole carbon source (condition 1) or by adding n-butanol to reinforced clostridial media (RCM) containing glucose (5 g/L) (condition 2). All experiments were conducted at 37 ◦ C with 10% (v/v) inoculum pre-grown in RCM. C. pasteurianum cultures were initially grown in either Biebl or RCM media until mid-exponential phase growth (two days). These cultures were then split into thirds prior to the addition of exogenous n-butanol to provide replicate experiments. The effect of n-butanol was investigated by adding 0 to 1% (w/v) (0 g/L to 10 g/L) exogenous n-butanol to the media. Under condition 1, nbutanol concentrations reflect exogenous n-butanol and n-butanol produced endogenously by C. pasteurianum. In contrast, under condition 2, C. pasteurianum did not produce n-butanol when grown on glucose and the n-butanol concentrations reflected only exogenous n-butanol. The cells were allowed to continue growing for 24 h after exogenous n-butanol was added, at which point the fermentations were terminated and the cell membranes were extracted. Growth (optical density) was monitored using spectrophotometry absorbance measurement at 600 nm of the cultures and the substrate utilization and product formation was measured using HPLC as described earlier (Venkataramanan et al., 2012).
2.3. Extraction of the cell membrane Cell membranes were extracted using the modified protocol of Bligh and Dyer using dichloromethane/methanol mixtures (Bligh and Dyer, 1959). The cells were harvested (0.5 mL cell suspension) by centrifugation at 13,000 rpm for 15 min and the pellets were resuspended in 0.5 mL of sterile 1.0% (w/v) NaCl in a 10 mL glass sample tube with PTFE lined caps. To the resuspended pellets, 2 mL of dichloromethane/methanol mixture (1:2 v/v) was added and shaken vigorously for 15 min. Following a 2 h incubation at room temperature (RT), the resuspended pellets were centrifuged at 2500 rpm and the supernatant (S1) was collected in a fresh tube. The pellet was again resuspended in 0.5 mL 1.0% NaCl and 2 mL of a dichloromethane/methanol mixture (2:1 v/v) was added to the resuspended pellet and shaken vigorously for 15 min. Following a 2 h incubation, the samples were centrifuged at 2500 rpm and the supernatant (S2) was collected. Supernatants S1 and S2 were combined and 1 mL of dichloromethane and 1 mL of sterile 1.0% NaCl were added. The top phase (aqueous) was removed and the bottom (organic) phase was retained. The solvent was evaporated under a gentle nitrogen stream. Once a dry film was obtained, the headspace was flushed with nitrogen, capped tightly, and stored for further analysis.
10
K.P. Venkataramanan et al. / Journal of Biotechnology 179 (2014) 8–14
2.4. GC–MS analysis of lipid composition Membrane extracts were pooled together for each condition and the lipids were methylated for GC/MS analysis. The dried lipid samples were saponified using 1 mL of 3 N sodium hydroxide at 90 ◦ C for an hour and then cooled to RT. The excess sodium hydroxide was neutralized with 1.8 mL of 3.6 N hydrochloric acid at 90 ◦ C for 10 min and cooled to RT. The free fatty acids were extracted using 1 mL of hexane and diethyl ether (1:1 v/v). The organic phase was then separated into a dry round bottom flask. The fatty acid hydrolyzates were dried using a rotary evaporator and stored under Argon in desiccators. The dried fatty acid hydrolyzates were derivatized using a 5 mL borontrifluoride–methanol complex at 60 ◦ C for 5 min and then cooled to RT. To the cooled solution, 1 mL water and 1 mL hexane were added and the container shaken multiple times to ensure the transfer of esters into the non-polar solvent. The upper organic layer was removed and transferred into an Erlenmeyer flask containing 5 g of anhydrous sodium sulfate. The flask was incubated at RT to dry overnight. The sodium sulfate was filtered and the hexane solution was transferred into a round bottom flask and dried in a rotary evaporator and the samples were stored under Argon. For GC–MS analysis, samples were dissolved in 250 L of dichloromethane. Fatty acid methyl esters (FAME) derivatives were analyzed by a Shimadzu GC–MS QP 2010 system using a SHR5xLB silica capillary column (30 m × 0.25 mm ID, composed of 100% dimethyl polysiloxane). The compounds were identified by comparison with the data in the NIST libraries. The FAME distribution was determined as FAME (%) =
A
FAME
AFAME
(1)
where AFAME is the peak area associated with a single FAME species and AFAME is the total area of all FAME species. 2.5. NMR analysis The dry film of the membrane was dissolved in CDCl3 for analysis. Synthetic lipids, DPPC and DOPC were used as standards for NMR analysis. Various ratios of DPPC and DOPC (1:0, 3:1, 1:1, 1:3 and 0:1) in a total lipid concentration of 10 mM were used for calibration (Supplementary material and Fig. S1). The synthetic lipids were dried and dissolved in CDCl3 . All 1 H-NMR spectra were recorded on a VarianTM Unity Inova 500 (500 MHz) spectrometer equipped with a 5 mm triple resonance inverse detectable probe. The percentage of unsaturation in the membrane lipid samples was calculated from the integration ratios (I) of the olefinic hydrogen signals (at 5.31 ppm) as
1 H-NMR
% unsaturation =
Iolefinic . Itotal
(2)
2.6. Surface pressure–area (–A) isotherms –A isotherms were recorded using a Langmuir trough (model 102 M, Nima technology Ltd, UK) with a deposition area of 70 cm2 at 25 ◦ C, 37 ◦ C and at 50 ◦ C. A Wilhelmy plate connected to an electronic micro-balance was used to measure surface pressure with an accuracy of ± 1 N/m. The external water bath system was used to control the subphase temperature. Monolayers were obtained by spreading diluted solutions of lipid samples in chloroform at the air/water interface using a 5 L syringe. The chloroform was allowed to evaporate and the lipid films spread at air/water within 15 min. The system was the equilibrated for 15 min before being compressed at a speed of 10 cm2 /min. Mechanical properties of the membrane extract monolayer films were determined by the elastic modulus, Cs −1 , which is the
reciprocal of the area compressibility (Duncan and Larson, 2008). The elastic modulus corresponds to the elasticity of the monolayers under the compression force. Cs−1 values were calculated as
Cs−1 = −A
∂p ∂A
.
(3)
T
where, A is the area per membrane extract weight (cm2 /g) at surface pressure under isothermal conditions. 3. Results and discussion 3.1. Glycerol fermentation and n-butanol challenge The effect of n-butanol stress on the membrane composition was studied in a semi-defined Biebl media at 37 ◦ C, with glycerol as the sole carbon source (condition 1). The fermentation of glycerol is typically biphasic in nature, with an acidogenic (acetate, butyrate and 1,3-PDO) exponential phase followed by a solventogenic (nbutanol) late exponential and stationary phase. C. pasteurianum produces n-butanol mainly when glycerol is the carbon source. C. pasteurianum cultures do not undergo solventogenesis during the fermentation of glucose, as the fermentation is predominantly only in the acidogenic phase, resulting in butyric acid as the major fermentation product with no n-butanol formation (condition 2). Results for glycerol fermentation (condition 1) are shown in Fig. 1 during n-butanol challenge. Exogenous n-butanol was added at concentrations of 0 g/L (control), 2.5 g/L (0.0335 M), 5 g/L (0.067 M), 7.5 g/L (0.105 M) and 10 g/L (0.134 M) to cultures in midexponential phase growth, grown in static flasks. The endogenous concentration of butanol of at t = 0 was 1.65 g/L. After the addition of exogenous butanol, the total concentration at t = 0 was 1.65 g/L, 3.14 g/L, 5.8 g/L, 8.8 g/L and 11.53 g/L, respectively. The concentrations of exogenous n-butanol (0 to 10 g/L) were selected based on n-butanol produced during the consumptions of glycerol, which is between 5 g/L and 50 g/L (glycerol concentration) based on previous studies (Biebl, 2001; Dabrock et al., 1992; Jensen et al., 2012a,b; Taconi et al., 2009; Venkataramanan et al., 2012). The optical densities indicated that cell growth was nearly identical across cell cultures prior to the addition of n-butanol (identified as t = 0 h) (Fig. 1A). After 24 h, there was a clear reduction in optical density with increasing exogenous n-butanol concentrations indicative of cell growth inhibition. However, analysis of glycerol utilization indicates that the bacteria continued to consume glycerol despite this growth inhibition (Fig. 1B). The total n-butanol concentrations at 0 h and 24 h, which reflected the combination of endogenous and exogenous n-butanol within the media, are shown in Fig. 1C. After 24 h the total n-butanol concentration decreased with increasing exogenous n-butanol concentration. This is more clearly depicted in the change in product concentrations from 0 h to 24 h, [product], shown in Fig. 1D. Adding exogenous n-butanol did not significantly change the production of acetate, ethanol, or butyrate; but it did lead to decreases in the production of 1,3-propanediol and n-butanol. C. pasteurianum were able to tolerate exogenous n-butanol up to 2.5 g/L, but beyond that concentration the amount of exogenous n-butanol became intolerable and the ability to produce solvents decreased proportionately. For n-butanol, there is a negative change in concentration and, at 10 g/L exogenous n-butanol, the total concentration of butanol in the media at 24 h is less than that of the exogenous n-butanol added at 0 h. This apparent loss of n-butanol can be attributed to n-butanol partitioning from the media, where samples are collected for product analysis, into the cell membrane. Collectively, the fermentation results show that adding exogenous
K.P. Venkataramanan et al. / Journal of Biotechnology 179 (2014) 8–14
11
Fig. 1. Glycerol fermentation (25 g/L) by C. pasteurianum. (A) Cell growth determined by optical density as a function of fermentation time. Arrows at 2 and 3 days denote when exogenous n-butanol was added (0 h) and when the fermentation was terminated and membranes were extracted post-stress (24 h), respectively. (B) Glycerol utilization as a function of exogenous n-butanol concentration. (C) Total n-butanol concentration, [n-butanol], in the media (endogenous + exogenous) as a function of exogenous n-butanol concentration. (D) Change in product concentrations, [product] from 0 h to 24 h. Error bars for 24 h samples represent standard deviation of triplicate experiments. Lines are shown to guide the eye and depict trends.
n-butanol inhibited cell growth and solvent production. However, exogenous n-butanol did not inhibit glycerol consumption, which suggests that the bacteria utilized the substrate for maintenance functions (i.e. counteracting n-butanol toxicity) and energy production. 3.2. Homeoviscous response due to n-butanol toxicity Homeoviscous response with n-butanol challenge (condition 1, n-butanol production) was determined based on GC–MS analysis of C. pasteurianum membranes extracted after 24 h. Results were compared to the homeoviscous response with n-butanol challenge of C. pasteurianum grown in RCM (condition 2, no n-butanol production).
For comparative purposes and ease of viewing, only the GC–MS results at 0 g/L and 10 g/L exogenous n-butanol are shown in Fig. 2. Results at all exogenous n-butanol concentrations are provided in the Supporting information (Fig. S1). With glycerol fermentation, the homeoviscous response involved progressive increases in the lipid tail length and decreases in the percentage of unsaturated lipids with increasing exogenous n-butanol concentration (Figs. 2A and 3A, respectively). At 0 g/L n-butanol the membrane was composed of lipids with tail lengths less than C16 (palmitoyl), while at 10 g/L the tails lengths were greater than C16 , and predominately C18 (stearoyl) or higher (Fig. 2A). The longer tailed lipids (C19 to C22 ) were almost completely absent when no external n-butanol
Fig. 2. Membrane fatty acid distribution, determined from GC–MS as fatty acid methyl esters (FAME), at 0 g/L and 10 g/L exogenous n-butanol concentrations. (A) FAME distribution at 25 g/L glycerol, corresponding to fermentation results in Fig. 1, and (B) in RCM media containing 5 g/L glucose.
12
K.P. Venkataramanan et al. / Journal of Biotechnology 179 (2014) 8–14
Fig. 3. Percentage of unsaturated lipids, determined as FAME via GC–MS and 1 H-NMR, in C. pasteurianum membrane extracts as a function of exogenous n-butanol concentration during (A) glycerol fermentation and (B) in RCM media.
was present. In conjunction with increasing tail length, the percentage of unsaturated lipids determined by GC–MS and 1 HNMR decreased from with increasing n-butanol concentration (Fig. 3A). A markedly different response was observed when the bacteria were grown in RCM (condition 2) where n-butanol production was not possible. In this case there was minimal change in the lipid tail length from 0 g/L to 10 g/L exogenous n-butanol (Fig. 2B). Furthermore, the trend in percentage of unsaturated lipids was opposite of that for glycerol fermentation; increasing unsaturation with increasing n-butanol concentration (Fig. 3B). Comparing condition 1 (glycerol) and condition 2 (RCM) suggest that the homeoviscous response of C. pasteurianum to elevated n-butanol concentrations only occurs when n-butanol is being produced. Hence, biophysical changes in membrane structure alone, which would have occurred due to n-butanol partitioning in condition 2, were not sufficient to drive an increase in tail length based on the conditions examined, but did yield an increase in the degree of lipid unsaturation. While the effects of n-butanol on membrane composition have not been reported for C. pasteurianum, they have been reported for C. acetobutylicum fermenting glucose (Baer et al., 1987, 1989; Lepage et al., 1987; Vollherbst-Schneck et al., 1984). In this case, C. acetobutylicum has been shown to tolerate n-butanol through a homeoviscous response that predominantly involves an increase in the SFA/UFA ratio within the lipid membrane (Baer et al., 1987, 1989; Lepage et al., 1987; Vollherbst-Schneck et al., 1984). Lepage et al. (1987) have reported that the lipid composition of C. acetobutylicum consists primarily of fatty acids from
Fig. 4. Weight averaged fatty acid tail length as a function of the change in n-butanol concentration due to the addition of exogenous n-butanol.
C12 to C19 . The percentage of unsaturated fatty acids was close to 50% in the absence of n-butanol challenge, and reduced to 46.5% and 43.5% upon exposure to 4 g/L and 8 g/L n-butanol, respectively. At 8 g/L n-butanol, this reflects a 13% decrease in UFA. Comparatively, C. pasteurianum has a lower percentage of unsaturated lipids in its membrane (∼12%), but the decrease in UFA was ∼25% upon exposure to 10 g/L n-butanol. Increases
Fig. 5. (A) Surface pressure–area (–A) isotherms of reconstituted membrane monolayers at 37 ◦ C at air/water interfaces. (B) Elastic modulus determined from –A isotherms. The legends show exogenous n-butanol concentrations corresponding to the fermentation conditions from which the membranes were extracted.
K.P. Venkataramanan et al. / Journal of Biotechnology 179 (2014) 8–14
in lipid acyl tail length were also observed for C. pasteurianum and there is a clear relationship between the average tail length and the change in n-butanol production (Fig. 4). A decrease in n-butanol production due to n-butanol toxicity correlates with an increase tail length. Hence, increases in both the SFA/UFA ratio and lipid tail length comprise the homeoviscous response of C. pasteurianum. 3.3. –A isotherms of membrane extract monolayers The homeoviscous response observed for C. pasteurianum counteracted n-butanol fluidization (membrane disordering) via increased lipid tail length. Increasing tail length increases the attractive van der Waals interactions between lipids and leads to a more rigid, stable membrane. Membrane extract monolayers spread at air/water interfaces were examined to gain a better understanding of this process. Membrane lipids within the monolayers were oriented with the hydrophilic polar headgroups in water and the hydrophobic hydrocarbon tails extending into air. –A isotherms and elastic moduli of membrane monolayers are plotted in Fig. 5. The temperature was maintained at 37 ◦ C consistent with the fermentation temperature. Because the average lipid molecular weight could not be determined, the areas are expressed as area per mass of membrane extract (cm2 /g) rather than area per molecule, which is typically used for lipid monolayers. As the monolayers were compressed the surface pressure increased with decreasing area per mass due to greater lipid packing at the interface. There were no distinct plateau regions within the –A isotherms, suggesting that the monolayers did not undergo two-dimensional phase transitions at the conditions examined (Fig. 5A). Increasing n-butanol concentration shifted the isotherms to higher surface pressures, which can be attributed to the longer lipid tail lengths due to the homeoviscous response. Higher molecular weight lipids with longer tails occupy more area at the air/water interface. The monolayers also exhibited higher elastic moduli with increasing n-butanol concentration, reflecting a more rigid or less compressible membrane (Fig. 5B). 4. Conclusions This study is the first to report of the lipid composition in C. pasteurianum ATCC 6013, an anaerobic spore-forming bacteria, ferments glycerol as the sole substrate, resulting in a mixture of butanol, ethanol, and 1,3-propanediol (PDO) along with acetate and butyrateand the homeoviscous response of C. pasteurianum due to elevated n-butanol concentrations (n-butanol challenge). The analysis of the extracted membranes from cells exposed to various concentrations of n-butanol under two different conditions, (i) n-butanol production by glycerol fermentation (condition 1) and (ii) no n-butanol production in RCM (condition 2), resulted in two completely different responses with respect to the changes in lipid composition. Only when C. pasteurianum is producing n-butanol is it able to respond to elevated n-butanol concentrations. When stressed with exogenous n-butanol during n-butanol production, C. pasteurianum responded by increasing lipid tail length and the ratio of saturated to unsaturated fatty acids. Despite n-butanol toxicity, which led to cell growth inhibition and reduced solvent production, the bacteria continued to utilize glycerol to achieve this response. –A isotherms confirmed that the homeoviscous response increased the stability of the membrane to counteract n-butanol disordering. The homeoviscous response was similar to other n-butanol producing species of Clostridia, as well as other microbes that respond to tolerate toxic organic compounds. It is envisioned that these results can be coupled with proteomics,
13
functional genomics and metabolic engineering to determine the mechanism of homeoviscous response and for the future development of n-butanol-tolerant industrial strains. Acknowledgments This project was supported by NSF grant (CBET-0966846). We sincerely thank Dr. Vinka Oyanedel-Craver and Varun Kasaraneni at the University of Rhode Island for their assistance with GC–MS analysis. References Aguilar, L.F., Sotomayor, C.P., Lissi, E.A., 1996. Main phase transition depression by incorporation of alkanols in DPPC vesicles in the gel state: influence of the solute topology. Colloids Surf., A 108, 287–293. Baer, S.H., Blaschek, H.P., Smith, T.L., 1987. Effect of butanol challenge and temperature on lipid-composition and membrane fluidity of butanoltolerant Clostridium-acetobutylicum. Appl. Environ. Microbiol. 53, 2854– 2861. Baer, S.H., Bryant, D.L., Blaschek, H.P., 1989. Electron-spin resonance analysis of the effect of butanol on the membrane fluidity of intact-cells of Clostridiumacetobutylicum. Appl. Environ. Microbiol. 55, 2729–2731. Baut, F., Fick, M., Viriot, M.L., Andre, J.C., Engasser, J.M., 1994. Investigation of acetone–butanol–ethanol fermentation by fluorescence. Appl. Microbiol. Biotechnol. 41, 551. Beaven, M.J., Charpentier, C., Rose, A.H., 1982. Production and tolerance of ethanol in relation to phospholipid fatty-acyl composition in Saccharomyces-cerevisiae Ncyc 431. J. Gen. Microbiol. 128, 1447–1455. Biebl, H., 2001. Fermentation of glycerol by Clostridium pasteurianum—batch and continuous culture studies. J. Ind. Microbiol. Biotechnol. 27, 18–26. Bligh, E.G., Dyer, W.J., 1959. A rapid method of total lipid extraction and purification. Can. J. Biochem. Physiol. 37, 911–917. Bowles, L.K., Ellefson, W.L., 1985. Effects of butanol on Clostridium-acetobutylicum. Appl. Environ. Microbiol. 50, 1165–1170. Cevc, G., 1982. Water and membranes—the interdependence of their physicochemical properties in the case of phospholipid-bilayers. Stud. Biophys. 91, 45–52. Dabrock, B., Bahl, H., Gottschalk, G., 1992. Parameters affecting solvent production by Clostridium pasteurianum. Appl. Environ. Microbiol. 58, 1233–1239. Duncan, S.L., Larson, R.G., 2008. Comparing experimental and simulated pressurearea isotherms for DPPC. Biophys. J. 94, 2965–2986. Heipieper, H.J., Weber, F.J., Sikkema, J., Keweloh, H., Debont, J.A.M., 1994. Mechanisms of resistance of whole cells to toxic organic-solvents. Trends Biotechnol. 12, 409–415. Iiyama, S., Toko, K., Murata, T., Ichinose, H., Suezaki, Y., Kamaya, H., Ueda, I., Yamafuji, K., 1992. Cutoff effect of n-alkanols in an excitable model membrane composed of dioleyl phosphate. Biophys. Chem. 45, 91. Ingram, L.O., 1982. Regulation of fatty-acid composition in Escherichia-coli—a proposed common mechanism for changes induced by ethanol, chaotropic agents, and a reduction of growth temperature. J. Bacteriol. 149, 166–172. Isar, J., Rangaswamy, V., 2012. Improved n-butanol production by solvent tolerant Clostridium beijerinckii. Biomass Bioenergy 37, 9–15. Jensen, T.O., Kvist, T., Mikkelsen, M.J., Christensen, P.V., Westermann, P., 2012a. Fermentation of crude glycerol from biodiesel production by Clostridium pasteurianum. J. Ind. Microbiol. Biotechnol. 39, 709–717. Jensen, T.O., Kvist, T., Mikkelsen, M.J., Westermann, P., 2012b. Production of 1,3PDO and butanol by a mutant strain of Clostridium pasteurianum with increased tolerance towards crude glycerol. AMB Express 2, 44. Kurniawan, Y., Venkataramanan, K.P., Scholz, C., Bothun, G.D., 2012. n-Butanol partitioning and phase behavior in DPPC/DOPC membranes. J. Phys. Chem. B 116, 5919–5924. L˘az˘aroaie, M.M., 2009. Mechanisms involved in organic solvent resistance in Gramnegative bacteria. World Acad. Sci. Eng. Technol. 30, 643–653. Lenaz, G., 1987. Lipid fluidity and membrane protein dynamics. Biosci. Rep. 7, 823–837. Lepage, C., Fayolle, F., Hermann, M., Vandecasteele, J.P., 1987. Changes in membrane lipid-composition of Clostridium-acetobutylicum during acetone butanol fermentation—effects of solvents, growth temperature and Ph. J. Gen. Microbiol. 133, 103–110. Lingwood, D., Kaiser, H.J., Levental, I., Simons, K., 2009. Lipid rafts as functional heterogeneity in cell membranes. Biochem. Soc. Trans. 37, 955–960. Liu, S.Q., Qureshi, N., 2009. How microbes tolerate ethanol and butanol. New Biotechnol. 26, 117–121. Lobbecke, L., Ceve, G., 1995. Effects of short-chain alcohols on the phase-behavior and interdigitation of phosphatidylcholine bilayer-membranes. Biochim. Biophys. Acta, Biomembr. 1237, 59–69. Lopez-Montero, I., Arriaga, L.R., Monroy, F., Rivas, G., Tarazona, P., Velez, M., 2008. High fluidity and soft elasticity of the inner membrane of Escherichia coli
14
K.P. Venkataramanan et al. / Journal of Biotechnology 179 (2014) 8–14
revealed by the surface rheology of model Langmuir monolayers. Langmuir 24, 4065–4076. Ramos, J.L., Duque, E., Gallegos, M.-T., Godoy, P., Ramos-González, M.I., Rojas, A., Terán, W., Segura, A., 2002. Mechanisms of solvent tolerance in gram-negative bacteria. Annu. Rev. Microbiol. 56, 743–768. Sardessai, Y., Bhosle, S., 2002. Tolerance of bacteria to organic solvents. Res. Microbiol. 153, 263–268. Simons, K., Ikonen, E., 1997. Functional rafts in cell membranes. Nature 387, 569–572. Taconi, K.A., Venkataramanan, K.P., Johnson, D.T., 2009. Growth and solvent production by Clostridium pasteurianum ATCC® 6013TM utilizing biodiesel-derived crude glycerol as the sole carbon source. Environ. Prog. Sustainable Energy 28, 100–110.
Timmons, M.D., Knutson, B.L., Nokes, S.E., Strobel, H.J., Lynn, B.C., 2009. Analysis of composition and structure of Clostridium thermocellum membranes from wildtype and ethanol-adapted strains. Appl. Microbiol. Biotechnol. 82, 929–939. Uchida, K., 1974. Occurrence of saturated and mono-unsaturated fatty-acids with unusually long chains (C20–C30) in Lactobacillus heterohiochii, an alcoholophilic bacterium. Biochim. Biophys. Acta 348, 86–93. Venkataramanan, K.P., Boatman, J.J., Kurniawan, Y., Taconi, K.A., Bothun, G.D., Scholz, C., 2012. Impact of impurities in biodiesel-derived crude glycerol on the fermentation by Clostridium pasteurianum ATCC 6013. Appl. Microbiol. Biotechnol. 93, 1325–1335. Vollherbst-Schneck, K., Sands, J.A., Montenecourt, B.S., 1984. Effect of butanol on lipid composition and fluidity of Clostridium acetobutylicum ATCC 824. Appl. Environ. Microbiol. 47, 193–194.
