ARTICLE The Chemical Formula of a Magnetotactic Bacterium Mohit Naresh,1 Sayoni Das,1 Prashant Mishra,1 Aditya Mittal2 1

Department of Biochemical Engineering and Biotechnology, Indian Institute of Technology Delhi, Hauz Khas, New Delhi 110016, India 2 Kusuma School of Biological Sciences, Indian Institute of Technology Delhi, Hauz Khas, New Delhi 110016, India; telephone: 011-91-11-26591052; fax: 011-91-11-26582037; e-mail: [email protected] Received 12 August 2011; revision received 11 November 2011; accepted 30 November 2011 Published online 14 December 2011 in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/bit.24403

ABSTRACT: Elucidation of the chemical logic of life is one of the grand challenges in biology, and essential to the progress of the upcoming field of synthetic biology. Treatment of microbial cells explicitly as a ‘‘chemical’’ species in controlled reaction (growth) environments has allowed fascinating discoveries of elemental formulae of a few species that have guided the modern views on compositions of a living cell. Application of mass and energy balances on living cells has proved to be useful in modeling of bioengineering systems, particularly in deriving optimized media compositions for growing microorganisms to maximize yields of desired bio-derived products by regulating intra-cellular metabolic networks. In this work, application of elemental mass balance during growth of Magnetospirillum gryphiswaldense in bioreactors has resulted in the discovery of the chemical formula of the magnetotactic bacterium. By developing a stoichiometric equation characterizing the formation of a magnetotactic bacterial cell, coupled with rigorous experimental measurements and robust calculations, we report the elemental formula of M. gryphiswaldense cell as CH2.06O0.13N0.28Fe1.74  103. Remarkably, we find that iron metabolism during growth of this magnetotactic bacterium is much more correlated individually with carbon and nitrogen, compared to carbon and nitrogen with each other, indicating that iron serves more as a nutrient during bacterial growth rather than just a mineral. Magnetotactic bacteria have not only invoked some interest in the field of astrobiology for the last two decades, but are also prokaryotes having the unique ability of synthesizing membrane bound intracellular organelles. Our findings on these unique prokaryotes are a strong addition to the limited repertoire, of elemental compositions of living cells, aimed at exploring the chemical logic of life. Biotechnol. Bioeng. 2012;109: 1205–1216. ß 2011 Wiley Periodicals, Inc. Correspondence to: A. Mittal Contract grant sponsor: IIT Delhi Contract grant sponsor: Department of Science & Technology, Govt. of India Contract grant number: SR/FTP/ETA-29 Contract grant sponsor: Department of Biotechnology, Govt. of India Contract grant number: BT/PR7837/BRB/10/503/2006 Additional Supporting Information may be found in the online version of this article.

ß 2011 Wiley Periodicals, Inc.

KEYWORDS: magnetosome; biomineralization; synthetic biology; bioprocess; calculations; biosynthesis

Introduction Magnetotactic bacteria, discovered by Blakemore (1975), are aquatic prokaryotes which grow at ambient/ mesophilic conditions. These bacteria have a unique feature of consuming soluble salts of iron from growth media to form intracellular chains of single magnetic crystals (30–40 nm) of magnetite (Fe3O4) and gregite (Fe3S4) (Bazylinski et al., 1995; Blakemore et al., 1979; Faivre et al., 2007; Faivre and Schu¨ler, 2008; Mann et al., 1990). These intracellular nano-crystals, each encapsulated by its own biological membrane about 3–4 nm thick (Gorby et al., 1988; Grunberg et al., 2004) are known as magnetosomes. The bio-derived magnetosomes are eco-friendly, non-toxic, and exhibit high degree of uniformity for crystals shape, size, and orientation (elegantly discussed in Faivre and Schu¨ler, 2008; and see Naresh et al., 2009, 2011 for specific examples). Further, magnetotactic bacteria provide an excellent experimental system for understanding the bio-mineralization process on nano-scales (Arakaki et al., 2010; Prozorov et al., 2007; Staniland et al., 2008), especially from an energetic perspective using bioprocess principles (Naresh et al., 2010). On the application side, the biologically derived magneto-functional inorganic nanocrystals of magnetite have been used as carriers for enzymes and in immunoassay methods (Matsunaga and Kamiya, 1987; Matsunaga et al., 1996a; Matsunaga et al., 2003; Nakamura et al., 1991) including those involving nucleic acids (Takeyama et al., 1995; Yoza et al., 2002), antibodies (Lang et al., 2007;

Biotechnology and Bioengineering, Vol. 109, No. 5, May, 2012

1205

Wacker et al., 2007), and targeted delivery of anticancer drugs (Matsunaga et al., 1997; Rockenberger et al., 1999; Sun et al., 2008). However, a key question is why do magnetotactic bacteria synthesize magnetosomes? The answer to this question has so far appeared only in terms of an evolutionary hypothesis of magnetoaerotaxis— the bacteria align with the Earth’s magnetic field to reduce their search space for the right oxygen environment from three-dimensions to one-dimension. However, it is also possible that a metabolic coupling of iron uptake and utilization with growth-related processes has evolved closely in these bacteria. In order to gain insights into this possibility, a thorough investigation is required in terms of overall metabolism of these bacteria. Energy required for metabolism and growth for any cell is derived from the breakdown of various compositions of sources rich in carbon (C), hydrogen (H), oxygen (O), and nitrogen (N) present in cell growth medium. Evaluation of the chemical basis for the utilization of nutrients for the cell growth has evolved an approach based on the mass balance that allows determination of the ‘‘chemical formula’’ of a cell in terms of stoichiometric ratios of carbon, hydrogen, oxygen, and nitrogen (CaHbOgNd). Utilization of this chemical formula leads to not only an in-depth understanding of metabolism in microbial species, but also plays crucial role in our ability to scale-up the growth of microbes to achieve production of desired products (Bailey and Ollis, 1986; Doran, 1995; Fogler, 2005; Shuler and Kargi, 2001). Therefore, it is expected that elucidation of the chemical formula for a magnetotactic bacterium will be a significant step in optimizing cell growth conditions for maximizing magnetosomal production, as is well established for a variety of other microbial species (Bailey and Ollis, 1986; Doran, 1995; Stockar and Liu, 1999). Further, a comparison of the chemical formula for a magnetotactic bacterium with that of other microbial species may provide valuable insights into the stoichiometric mechanisms leading to bio-mineralization. In this work, Magnetospirillum gryphiswaldense has been grown in defined medium under controlled oxygen supply in a bioreactor to understand the chemical basis and the process optimization for magnetosomes production. Using principles of elemental mass balance to determine chemical formula of M. gryphiswaldense, we report the discovery of some interesting biochemical aspects of metabolism, like nutrient concentrations and correlation coefficients of nutrients, essential for the growth of M. gryphiswaldense and magnetosome productivity. An extremely useful and interesting discovery is that of close coupling of carbon and iron consumption by these bacterial cells. Our findings now reconcile data from different groups reporting lower cell and magnetosomal productivity (cells and magnetosomes produced per unit time) in low-carbon medium (Faivre et al., 2007) and, high cell and magnetosomal productivity with carbon-rich medium (Staniland et al., 2007).

1206

Biotechnology and Bioengineering, Vol. 109, No. 5, May, 2012

Materials and Methods Bacterial Strain, Culture Medium, and Inoculum Preparation M. gryphiswaldense strain MSR-1 (DSM 6361), DSMZ (Germany) was used for all experiments. At all stages of growth culture purity was confirmed with gram staining, spirillum morphology, and high motility under optical microscope. Transmission electron microscopy (TEM) was carried out to confirm the presence of magnetosomes inside the cell by using CM-10 TEM, (Philips, Eindhoven, Netherlands) as described earlier (Naresh et al., 2009, 2011). For cultivation of M. gryphiswaldense, liquid medium contained (per liter): 35 mM potassium L-lactate (Fluka, Sigma–Aldrich, Deisenhofen, Germany) as main carbon source. Other media constituents included (per liter) 2.38 g Hepes, 3 g soya peptone, 0.1 g yeast extract (Himedia, Mumbai, India), 0.34 g sodium nitrate, 0.1 g potassium dihydrogen phosphate, (Merck, Darmstadt, Germany), and 100 mM Ferric citrate (Loba Chemie, Mumbai, India). One microliter trace element solution (Sharma et al., 2007) prepared afresh was added in distilled water and pH of the culture medium was adjusted to 7 with freshly prepared sodium hydroxide solution. Nitrogen was flushed through the final boiled medium which was then sealed with butyl rubber stoppers and autoclaved at 1218C and 15 psig for 20 min. Distilled water was used for all solutions. After autoclaving 500 mL flasks containing 100 mL volume of medium were flushed with sterile nitrogen. 0.15 g magnesium sulfate heptahydrate and 1 g Na-thioglycolate (Merck) were added after cooling the autoclaved medium. 10% (v/v) sterile air was introduced inside flask via syringe through rubber stopper retained by plastic cap with small hole for collecting samples. The medium flasks were inoculated with 10% (v/v) cells growing in exponential phase from the standard flask. Growth in flasks was monitored at 288C and agitation speed of 100 rpm in incubator shaker.

Cultivation of M. gryphiswaldense in Bioreactor and Analytical Methods Bacterial culturing was done in a 3 L bioreactor (Bioengineering AG, Wald, Switzerland) with 2 L working volume having in-situ autoclaving facility. Hepes, trace element mixture and yeast extract were omitted from flask cultivation medium. Other medium components concentration was kept similar as for flask growth except iron supplementation in the form of ferric citrate (150 mM) in distilled water. All culture medium components were autoclaved at 1218C for 30 min. The pH was set at 7 and controlled by the addition of either 1 M H2SO4 or 1 M NaOH. The dissolved oxygen (DO) probe was calibrated with sterile nitrogen and air. After autoclaving, sterile nitrogen was flushed during cooling and MgSO4.7H2O was added to the cooled medium. Filter sterilized air was purged

to culture medium during cultivation of magnetotactic bacteria to keep dissolved oxygen concentration to 6% (v/v). Bioreactor was inoculated with 10% (v/v) inoculum from exponential phase of the growth flask. During growth agitation speed was kept in the range of 150–300 rpm and temperature was kept constant at 288C. Cell growth was measured at an absorbance of 565 nm (OD565) by spectrophotometer (Helios Epsilon, ThermoSpectronic, Mumbai, India). The concentration of lactate in supernatant was measured by high performance liquid chromatography (Agilent 1100 series, Waldbronn, Germany). Analysis was done using Aminex HPX-87 H (BioRad, Gladesville, New South Wales, Australia) column at 408C with a mobile phase of 0.01 N H2SO4 with a flow rate 0.5 mL m1. The eluted acid was detected at an absorbance of 210 nm. The concentration of nitrate ion was measured by colorimetric method, nitration of salicylic acid under highly acidic conditions shows maximum absorbance at 410 nm in basic solution (Catalado et al., 1975). Total nitrogen was estimated by Kjeldahl method (FOSS Tecator, Sweden). Cell count, dry cell weight, and residual iron concentration was measured as explained earlier (Naresh et al., 2009, 2011). CO2 gas evolved was analyzed using a gas chromatograph (AIMIL-NUCON, Series 5700, New Delhi, India) equipped with a 6 feet long Poropak Q column and a Thermal Conductivity Detector (TCD). The conditions maintained for CO2 detection were: injector temperature 808C; oven temperature 508C; detector temperature 808C; TCD Current 80 mA. Standardization of the instrument was done using a standard calibration mixture gas containing carbon dioxide in methane and nitrogen (EDT Research, Birmingham, England). For quantification of Fe3O4 it is assumed that all Fe3þ is consumed for magnetite formation. Therefore if d moles of Fe3þ are consumed than d/3 moles of Fe3O4 are formed. To validate the assumption, after 40 h magnetosomes have been separated by lysing the cell (followed by centrifugation for 10 min at 10,000 rpm and washing with buffer five times) and dried under vacuum overnight. So the left out non-volatile crystals has been weighed. For first run 129 mM of iron has been consumed and the weight of crystals estimated as 10 mg/L. Thus Fe3O4 has been found to be 20 mg in 2 L, which is equivalent to 129 mM of iron that has been supplied in the 2 L reactor. Hence for the calculation, moles of magnetite formed are calculated with amount of iron consumed during growth of bacterial cell for all set of experiments.

Elemental (Carbon, Hydrogen, Nitrogen) Analysis Using CHN Analyzer For independent determination of elemental compositions by using standard CHN analysis, bacterial cells were cultured in 500 mL flasks. During the experiments, micro-aerobic conditions were maintained and hypodermic syringes were used for inoculation and sample collection. After 48 h, cells were harvested in their exponential phase and washed three

times (by centrifugation and resuspension) using phosphate buffer saline (PBS), pH 7.2. The wet cell weight was quantified by the difference in weight of a tube before and after adding the cell pellet. The dry cell weight was taken after keeping the wet cell pellet at 808C till it reached a constant weight. For CHN analysis, the dry cell pellet (in powder form) was provided to the Shriram Institute of Industrial Research (University Road, Delhi 110007, India). Utilizing a CHNS Vario III analyzer, and analytical conditions of combustion tube temperature of 1,1508C, reduction tube temperature of 8508C, helium as carrier gas, flow of 200 mL/min. and pressure of 1.25 psi, the Shriram Institute of Industrial Research measured and reported the results for carbon, hydrogen, and nitrogen in % by mass for 15 mg of sample.

Results Figure 1A shows the kinetics of nutrients carbon, nitrate, total nitrogen, and iron consumed for the magnetosomal synthesis and growth of M. gryphiswaldense represented by cell count and dry cell weight of biomass. The experiments were performed in a 3 L bioreactor in triplicates. Nutrient consumption and cell growth were monitored continuously at 4 h intervals. It is clear that nitrate was consumed earlier in the lag phase of the bacterial growth compared to ammonia represented by total nitrogen. Carbon, total nitrogen, and iron were consumed rapidly during exponential phase from 12 h onwards till 28 h. In Figure 1A, 20 h indicates the inflection point of the growth profile of M. gryphiswaldense for the consumption of nutrients and cell count. At the deceleration of exponential phase and the initiation of stationary phase (28 h onwards) substantial iron and total nitrogen were found to be unconsumed, whereas carbon was nearly exhausted at the later phase of growth of bacterial cell. M. gryphiswaldense phenotype was confirmed by gram staining and spirillum morphology, magnetic response, TEM images and a visible black pellet showing evidence of magnetosomes (Fig. 1B–F). Magnetosomes extracted from cell lyses in an inverted tube responds to external magnetic field (Fig. 1G).

Determination of Chemical Formula for M. gryphiswaldense Material balance is mostly used for various biochemical processes for cell growth and product formation. In our work, the data obtained during the bioreactor runs were utilized to investigate variables like consumption rate of organic substrate and nitrogen with controlled oxygen environment with subsequent biomass and product formation. For determining the chemical formula of magnetotactic bacterium M. gryphiswaldense, a stoichiometric equation for oxidation of carbon source with oxygen and nitrogen into biomass (M. gryphiswaldense) was developed

Naresh et al.: Chemical Formula of M. gryphiswaldense Biotechnology and Bioengineering

1207

Figure 1. Growth and characterization of M. gryphiswaldense for magnetosome formation: (A) Growth and nutrients consumed profile during growth of M. gryphiswaldense in 3 L bioreactor: lactate (*), total nitrogen (*), nitrate (&), iron (Fe3þ; ^), cell count (^), and dry cell weight (DCW; D). All data are shown as mean  SD of three independent set of experiments. B: Gram staining of M. gryphiswaldense. C: Cell deposition at the site of external magnet attached to test tube. D: TEM image of M. gryphiswaldense showing intracellular magnetosome. E: Magnified TEM image of aligned magnetosomal chain. F: Cell pellet of M. gryphiswaldense. G: Magnetic clustering of magnetosomes. [Color figure can be seen in the online version of this article, available at http://wileyonlinelibrary.com/bit] as shown below (Equation 1):

N-balance : b1 þ b2 ¼ ed

(5)

Fe-balance : d ¼ ev

(6)

þ 3þ aC3 H6 O3 þ b1 NO 3 þ b2 NH4 þ cO2 þ dFe

! eðCa Hb Og Nd Fev Þ þ f H2 O þ gCO2

(1)

In Equation (1), a represents the moles of the carbon source (lactate) as organic substrate for energy requirements of cell and biomass formation, b1 are the moles of nitrate and b2 are the moles of ammonia used as nitrogen source, c are the moles of oxygen (from the air supplied by maintaining dissolved oxygen), d are the moles of ferric citrate used for iron source in the growth medium, e are the ‘‘moles’’ of cells formed, f are the moles of water and g are the moles of carbon dioxide. C, H, O, N, Fe denotes the elemental composition of the biomass. a, b, g, d, and v are the empirical coefficients for the biomass (chemical formula of the cell). Thus, to estimate the chemical formula of the cell, stoichiometric calculations are done using elemental mass balance using a calculation basis of 1 L of total volume of culture broth (i.e., units of calculation used are in moles per liter for each ‘‘reacting’’ species). Stoichiometric coefficients a, b1, b2, c, and d of Equation (1) are measured quantities (i.e., estimated using analytical techniques, see Materials and Methods section) during the cultivation of M. gryphiswaldense in bioreactor. Therefore, elemental mass balance is applied to Equation (1) as follows:

1208

C-balance : 3a ¼ ea þ g

(2)

H-balance : 6a þ 4b2 ¼ eb þ 2f

(3)

O-balance : 3a þ 3b1 þ 2c ¼ ea þ f þ 2g

(4)

Biotechnology and Bioengineering, Vol. 109, No. 5, May, 2012

If ‘‘y’’ is the dry cell weight (in grams) obtained in 1 L, then overall mass balance of a cell gives e¼

y 12a þ b þ 16g þ 14d þ 55v

(7)

Now, the stoichiometric coefficients a, b1, b2 c, d, e, and g are known (measured) in Equation (1), along with y in Equation (7) (see details below). Therefore solving the linear Equations (2–7) simultaneously, we obtain the stoichiometric coefficient f and empirical coefficients a, b, g, d, and v as follows: 3a  g e

(8)

4b2  6b1  4c þ 4g þ 2eg e

(9)

a¼

b¼

g¼

y  36a þ 8g  8b1  18b2 þ 4c  55d e

d¼

b e

(10)

(11)

d e

(12)

empirical coefficients w.r.t. carbon. Thus, the normalized empirical formula (with respect to carbon) for M. gryphiswaldense is determined to be:

6a þ 4b2  eb 2

(13)

CH2:06 O0:13 N0:28 Fe1:74X103

v¼

f ¼

As mentioned above, a, b1, b2 c, d, e, g are known experimental parameters. Figure 2 shows values of these parameters at each time point during the growth of cells, calculated from the data shown in Figure 1. Data corresponding to Figure 2 are shown in the upper half of Table I. Also shown in the upper half of Table I are the measured values of dry cell weight, that is, y (also plotted in Fig. 1). Figure 3 shows the values of a, b, g, d, v, and f obtained for every time point after the onset of exponential phase (see supplementary material for a detailed worked example for the 12 h timepoint). It is clear from Figure 3 that values of the empirical coefficients of M. gryphiswaldense vary minimally with time (with the exception of d; see Discussion section). Further, the stoichiometric coefficient ( f) shows the formation of water molecules form during the growth of M. gryphiswaldense. Thus, Table II shows the summarization the results obtained in Table I and Figure 3, that is, the experimentally derived values of empirical coefficients in the chemical formula of M. gryphiswaldense. The results are shown as mean  SD, along with the ranges for each of the empirical coefficients. The findings of Table II lead to the absolute empirical formula of magnetotactic bacterium M. gryphiswaldense as: Cð2:310:16ÞX1010 Hð4:680:66ÞX1010 Oð2:930:76ÞX109 Nð6:562:24ÞX109 Feð4:000:43ÞX107

(A)

Generally, the chemical formulae of cells are expressed (and reported; see Discussion section) by normalizing the

(B)

Here, it is interesting to note that much before the evolution of a formally distinct definition for ‘‘Systems Biology,’’ application of elemental mass balance to system parameters had been applied to derive the chemical formulae of a significant number of microbial systems (Abbott and Clamen, 1973; Bailey and Ollis, 1986; Bauer and Ziv, 1976; Doran, 1995; Harrison, 1967; Herbert, 1976; Kok and Roels, 1980; Mayberry et al., 1968; Shimizu et al., 1978; Stouthamer, 1977; Stockar and Liu, 1999; van Dijken and Harder, 1975; Wang et al., 1976). Table III shows a summary of chemical formulae of a variety of microbial systems and a clear comparison of the chemical formula of the magnetotactic bacterium: Magnetospirillum gryphiswaldense can be made. As done in several of the classical studies shown in Table III, we also counter-checked the elemental ratios (particularly C:H, see below) in our bacterial cells using a CHNS analyzer. Note that while CHNS analyzers are excellent tools for providing empirical formulae for organic compounds, for whole cell stoichiometries (with absolute elemental coefficients being 9–10 orders of magnitude higher than those for organic compounds) results from these analyzers are meaningful only in terms of obtaining the ratios of the elements (rather than absolute stoichiometries). Further, it has been un-ambiguously demonstrated that diagenesis (and to some extent metamorphism) leads to loss of proteinaceous nitrogen content in bio-mineral samples—that is, there is a selective loss of nitrogen-rich matter (proteins) in thermal processing of biological samples with mineral content (Meyers, 1997). Thus, we were primarily interested in the relative carbon and hydrogen elemental contents in the bacterial cells. The CHNS analysis provided

Figure 2. Stoichiometric coefficients of Equation (1) for determining empirical coefficients of M. gryphiswaldense: (A) Consumption of lactate (a, *) sodium nitrate (b1, &), and ammonia (b2, &) consumed during cell growth. B: Molecular oxygen (c, *), iron (d, D) consumed during cell growth and formation of CO2 (g, ^) during cell growth. All data are shown as mean  SD of three independent set of experiments.

Naresh et al.: Chemical Formula of M. gryphiswaldense Biotechnology and Bioengineering

1209

Table I.

Average experimental stoichiometric and calculated empirical coefficients for the growth of M. gryphiswaldense. Experimental values (measured, expressed as moles unless specified otherwise)

Time (h) 0 4 8 12 16 20 24 28 32 36 40



(K ) Lactate (a) 0 1.08  103 3.59  103 6.97  103 1.28  102 2.03  102 2.97  102 3.38  102 3.44  102 3.47  102 3.52  102

Nitrate (b1) 0 8.47  104 2.12  103 3.36  103 3.36  103 3.36  103 3.36  103 3.36  103 3.36  103 3.36  103 3.36  103

Ammonia (b2) 0 6.67  104 1.67  103 3.67  103 6.67  103 9.67  103 1.17  102 1.37  102 1.33  102 1.37  102 1.33  102

Oxygen from air (c) 0 8.38  104 6.79  104 7.79  104 9.46  104 1.00  103 8.83  104 7.92  104 6.54  104 7.04  104 6.46  104

(Fe) Citrate (d) 0 2.42  106 8.75  106 2.35  105 5.02  105 9.13  105 1.16  104 1.23  104 1.24  104 1.25  104 1.25  104

Cell count (e) 14

2.30  10 4.93  1014 1.96  1013 6.58  1013 1.09  1012 1.95  1012 2.89  1012 3.10  1012 3.27  1012 3.33  1012 3.48  1012

CO2 (g)

DCW (y) (g)

— 2.19  105 1.37  103 5.60  103 9.52  103 1.75  102 2.60  102 2.79  102 2.95  102 3.01  102 3.14  102

5.67  102 7.92  102 2.33  101 3.57  101 6.40  101 9.50  101 1.17 1.34 1.38 1.43 1.43

Calculated values Time (h) 0 4 8 12 16 20 24 28 32 36 40

a

b

g

d

v

f

— 6.62  1010 4.96  1010 2.33  1010 2.67  1010 2.25  1010 2.21  1010 2.39  1010 2.26  1010 2.22  1010 2.14  1010

— 6.10  1010 2.73  1010 3.08  1010 4.84  1010 5.19  1010 4.64  1010 4.83  1010 4.88  1010 5.04  1010 4.92  1010

— 2.90  1010 2.15  1010 5.07  109 5.48  109 4.31  109 1.27  109 1.12  109 1.67  109 2.38  109 2.13  109

— 2.99  1010 2.04  1010 1.07  1010 9.31  109 6.76  109 5.20  109 5.49  109 5.11  109 5.12  109 4.81  109

— 4.88  107 4.71  107 3.55  107 4.60  107 4.69  107 4.04  107 3.99  107 3.81  107 3.75  107 3.61  107

— 6.02  103 1.18  102 1.81  102 2.53  102 2.95  102 4.53  102 5.37  102 5.01  102 4.74  102 4.65  102

the following results: C ¼ 28.1 (% by mass), H ¼ 4.7 (% by mass), and N ¼ 3.5 (% by mass). Thus, C:H was calculated as (28.1/12.0):(4.7/1.0) where denominators are the molecular weights of the individual elements. This yielded C:H ¼ 1:2.007, a figure remarkably close to 1:2.06 obtained in Equation (B) above. As expected, due to loss of proteinaceous content, the CHNS analysis provided a higher value of C:N (28.1/12.0):(3.5/14.0) ¼ 1:0.107 compared to 1:0.28 obtained in Equation (B). In fact, our results provide further experimental support to loss of proteinaceous content during collection and processing of biomineralizing samples analyzed using CHN analyzers discussed extensively by Meyers (1997). In our case, loss of proteinaceous content would not be expected to affect the C:H ratio only if proteins have a C:H ratio similar to C:H ratio of whole cells; thus loss of protein material would lead to loss of absolute C and H material but the measured C:H ratio would still remain the same (i.e., relatively C:H remain the same). Remarkably, proteins have a C:H ratio of 1:2, and C:H ratio measured for the bacterial cells by two different methods in this work is also 1:2. Therefore, Equation (B) represents an accurate formula for magnetotactic bacterial cells, supported by the near exact agreement of C:H data acquired by two independent methods. It is interesting to note from Table III that M. gryphiswaldense has a higher C:O content compared to other microbial species. At this point of time, this result (higher C:O indicating lower relative oxygen content) is best

1210

Biotechnology and Bioengineering, Vol. 109, No. 5, May, 2012

supported by (and simultaneously explains) the extremely fastidious nature of magnetotactic bacterial cells in terms of their highly specific micro-aerobic nature (in fact, the natural presence of intra-cellular nanomagnets is attributed to evolutionary pressures on these bacteria due to specific low-oxygen requirements). Further investigations, that are beyond the scope of this work, will allow better understanding of lower relative oxygen content of these bacterial cells, including utilization of a CHNO analyzer to directly measure C:O cellular stoichiometry (note that we had access only to a CHNS analyzer and not a CHNO analyzer).

Nutrient Consumption of M. gryphiswaldense Figure 4A shows that carbon consumption was constant w.r.t. cell growth up to the end of exponential phase. Figure 4B and C shows that the specific nitrogen consumption rate increases during initial phase of the growth and nitrogen required per cell decreases. While nitrate was consumed in the early phase, nitrogen present in terms of ammonia, presumably utilized for synthesis of amino acids and proteins, was consumed later. There is clearly an increase in nitrogen consumption during the initial phase than the later phase of the growth of bacterial cell (Fig. 4B). Figure 4D shows that the specific iron consumption rate clearly increases during exponential phase

Figure 3. Chemical formula coefficients for M. gryphiswaldense: (A) a, *; (B) b, &; (C) g, *; (D) d, ^, and (E) v, ^ derived from stoichiometric coefficients from Equation (1) and (F) f (D) is the number of moles of water formed during cell growth. All data are shown as mean  SD of three independent set of experiments.

implying more iron assimilation per cell during the exponential phase. Classically, carbon to nitrogen consumption ratios have been considered to be key factors in determining cell metabolism for growing microbial cultures. Therefore, we calculated the correlation coefficient (CC) for specific consumption rate of lactate (C), total nitrogen (N),

Table II.

and iron (Fe). We found that CC–C:N 0.2583  0.2674; Fe:N 0.6970  0.1422; Fe:C 0.7789  0.1603. The fact that iron consumption is more correlated with nitrogen and carbon, compared to carbon and nitrogen themselves was a surprising and important finding (Discussion section). To further confirm this result rigorously, it was essential to

Empirical coefficients for magnetotactic bacterium formula during the cultivation of M. gryphiswaldense. Chemcial formula coefficient

Mean  SD Range

Stoichiometric coefficient

a (1010)

b (1010)

g (109)

d (109)

v (107)

f (102)

2.31  0.16 2.14–2.67

4.68  0.66 3.08–5.19

2.93  1.76 1.11–5.48

6.56  2.24 4.81–10.71

4.00  0.43 3.55–4.69

3.95  1.32 1.81–5.37

Naresh et al.: Chemical Formula of M. gryphiswaldense Biotechnology and Bioengineering

1211

Table III.

Chemical formulae of microorganisms.

Microorganisms Bacteria Bacteria Candida utilis C. utilis C. utilis C. utilis Klebsiella aerogenes K. aerogenes K. aerogenes K. aerogenes Saccharomyces cerevisiae S. cerevisiae S. cerevisiae Paracoccus denitrificans P. denitrificans Escherichia coli Pseudomonas C12B Aerobacter aerogenes Magnetospirillum gryphiswaldense

Chemical formula

Ref.

CH1.666O0.27N0.20 CH2.0O0.50N0.27 CH1.83O0.54N0.10 CH1.87O0.56N0.20 CH1.83O0.46N0.19 CH1.87O0.56N0.20 CH1.75O0.43N0.22 CH1.73O0.43N0.24 CH1.75O0.47N0.17 CH1.73O0.43N0.24 CH1.64O0.52N0.16 CH1.83O0.56N0.17 CH1.81O0.51N0.17 CH1.81O0.51N0.20 CH1.51O0.46N0.19 CH1.77O0.49N0.24 CH2.00O0.52N0.23 CH1.83O0.55N0.25 CH2:06 O0:13 N0:28 Fe1:74103

Abbott and Clamen (1973) van Dijken and Harder (1975) Herbert (1976)

check the relationship between carbon and nitrogen consumption with magnetosomal synthesis. Since the media composition used in this study is well established to yield >95% cells with magnetosomal synthesis, which is directly correlated to cell growth (Staniland et al., 2010), it was important to relate the consumption of the elements to magnetosomal synthesis. Thus, carbon and nitrogen

Harrison (1967) Kok and Roels (1980) Wang et al. (1976) Stouthamer (1977) Shimizu et al. (1978) Bauer and Ziv (1976) Mayberry et al. (1968) Current study

consumption was plotted as a function of cell count. Figure 5A and B shows that both nutrients have a linear correlation with cell count. Carbon was consumed rapidly in the lag phase and then consumed linearly with cell growth in the exponential phase (Fig. 5C). Similar trend was observed for nitrogen and iron (Fig. 5D and E), thus confirming implications of the CCs found above.

Figure 4. Specific consumption rate (Q) during cell growth: (A) Lactate (*), (B) total nitrogen (D), (C) nitrate (*), and (D) iron (^). All data are shown as mean  SD of three independent set of experiments.

1212

Biotechnology and Bioengineering, Vol. 109, No. 5, May, 2012

Figure 5. Carbon, nitrogen, and iron uptake for magnetosome formation and cell growth: Linear correlation of (A) carbon (*) and (B) total nitrogen (D) consumed for magnetite formation with r2 ¼ 0.9874 and 0.9914, respectively, during cell growth. C: Carbon (*), (D) total nitrogen (D), and (E) iron (^) consumption during cell growth. The dashed lines indicate the stationary phase region (marked by ‘‘S’’) in the growth curve. All data are shown as mean  SD of three independent set of experiments.

Discussion Magnetotactic bacteria are obligate micro-aerophiles and need narrow range of dissolve oxygen for growth as well as magnetosome formation. For producing magnetosomes from magnetotactic bacteria, oxygen and its concentration are crucial (Blakemore et al., 1985; Heyen and Schu¨ler, 2003; Schu¨ler and Baeuerlein, 1998). In this study, maintenance of the right oxygen concentrations led to a serendipitous result—production of not only maximum cell mass (biomass) but also a very high magnetosomal yield, compared to several previous results (Heyen and Schu¨ler, 2003; Liu et al., 2010; Yang et al., 2001)

obtained by different groups, on different strains of magnetotactic bacteria, during batch cultivation of M. gryphiswaldense. While maintenance of critical levels of dissolved oxygen was not only crucial for our stoichiometric calculations leading to elucidation of the chemical formula of a magnetotactic bacterium, we have also obtained an increase of 40% magnetosomal productivity in a bioreactor under the controlled oxygen tension. In spite of the well-documented ability of intracellular magenotosmal formation by magnetotactic bacteria, the complete mechanism(s) of magnetosomal synthesis have not been elucidated. In terms of uptake of soluble iron, it has been reported that Magnetospirillum magneticum (AMB-1)

Naresh et al.: Chemical Formula of M. gryphiswaldense Biotechnology and Bioengineering

1213

and Magnetospirillum magnetotacticum (MS-1) produce hydroxamate phenolate and catechol (Calugay et al., 2003, 2006) and hydroxamate (Paoletti and Blakemore, 1986) siderophores, respectively. These siderophores bind to Fe3þ and assist in transporting iron inside the cells. However, so far no siderophores have been detected and reported in M. gryphiswaldense cultures. So how do these bacteria regulate iron metabolism leading to formation of intracellular iron-oxide crystals? It has been reported that empty vesicles are formed prior to magnetite formation in cell (Komeili et al., 2004) and intracellular magnetosomal chains with the help of proteins MamJ and MamK (Scheffel and Schu¨ler, 2007). However, is iron alone responsible for regulating pathway(s) involved in magnetosome formation, as indicated by studies done primarily at a genetic level? It is important to investigate the effects of nutrient consumption on cell physiology in terms of growth and magnetosomal synthesis, to be able to attain mechanistic information at a metabolic level. Thus, we have rigorously quantified and studied specific consumption rate (Q) of carbon, hydrogen, nitrogen and iron. Figure 4 highlights the chemical basis of nutrients consumed for growth and magnetosome formation. Carbon, consumed at a constant rate till the exponential phase, is used for energy generation and biosynthesis (Fig. 4A). Nitrate is consumed in the early phase of growth (Fig. 4C) and later nitrogen (Fig. 4B) in terms of ammonia present in culture medium (presumably used for synthesis of amino acids and proteins). Interestingly, nitrate concentration in culture medium is related to cell growth and nitrate reductase activity for magnetosome formation for the strain MS-1 (Matsunaga et al., 1996b; Taoka et al., 2003). Further, Suzuki et al. (2006) have suggested that activation of the nitrogen operon was similar to iron uptake system (feo) at a genetic level, though they did not investigate the implications at the metabolic level. Our results establish the metabolic coupling between iron and nitrogen consumption, thereby explaining the findings of Suzuki et al. (2006) who concluded that iron somehow activates the transporter systems at the transcriptional level, but not how. The reason for increase in nitrogen consumption during the initial phase rather than later (Fig. 4B) could be the following. During the lag and accelerating log phase there is a substantial demand of nitrogen to form proteins for assembly of magnetosome vesicles and, to transport and incorporate iron for magnetosome synthesis. Once the vesicles are formed, proteins allow nucleation and deposition of iron to form magnetosomes. The concentration of carbon as an energy source plays a crucial role during the magnetosomal formation. The transport process for magnetite synthesis by Fe3þ is energy dependent (i.e., requires cellular energy since iron is not internalized by passive diffusion) with a high velocity/ transfer-rate across the cell (Schu¨ler and Baeuerlein, 1996). In a recent study, it was shown that magnetosome formation stunts bacterial growth and the energy required for cell growth has to compromise with magnetosome formation

1214

Biotechnology and Bioengineering, Vol. 109, No. 5, May, 2012

(Naresh et al., 2010). Another study shows that the number of magnetosomes in magnetosomal chain is cell length dependent (Staniland et al., 2010). Thus, a high substrate concentration (i.e., of iron) is must to achieve Vmax for the iron transporters. For increasing number of magnetosomes, high iron concentration is required increasing the transfer rate of iron inside the cell. The CC value shows that iron consumption has higher correlation with carbon than nitrogen with carbon. The CC of C, N, and Fe shows that iron is not only limited to magnetosome formation but it also plays vital role in the overall metabolic activity of the cell. A linear correlation between consumption of carbon and nitrogen with cell growth (and hence magnetosome formation, Fig. 5A and B) confirms this. Carbon is essential energy requirements for cell growth and maintenance, and, for synthesis of magnetosomes. Hence carbon being the limiting source, in spite of availability of nitrogen and iron in the growth medium (Fig. 1A), the latter are not consumed in the later phase of the growth (Fig. 5D and E). Interestingly, our findings directly provide a link between two elegant reports, but with different/conflicting observations regarding time of magnetosomal synthesis. On one hand, Faivre et al. (2007) reported formation of magnetosomes in over 5 h. On the other hand, Staniland et al. (2007) reported formation of magnetosomes in just 30 min. Our findings in this work directly resolve the two different observations. While Faivre et al. (2007) used low-carbon content medium, Staniland et al. (2007) used much higher carbon content medium. Thus, when carbon was limited, not only was cell growth lower, but even magnetosomal synthesis was slower in contrast to when carbon was not limited, magnetosomes were synthesized much faster. We are grateful to Manish Sharma, Center for Applied Research in Electronics, IIT Delhi for assistance with acquisition of TEM data. M. N. acknowledges financial stipend support from IIT Delhi. A. M. acknowledges the financial support from Department of Science & Technology, Govt. of India via Grant # SR/FTP/ETA-29, and Department of Biotechnology, Govt. of India via Grant # BT/PR7837/BRB/ 10/503/2006. We are also grateful to (at least) three anonymous reviewers and the handling editor for their invaluable inputs leading to a much improved readability of the manuscript. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

References Abbott BJ, Clamen A. 1973. The relationship of substrate, growth rate, and maintenance coefficient to single cell protein production. Biotechnol Bioeng 15:117–127. Arakaki A, Masuda F, Amemiya Y, Tanaka T, Matsunaga T. 2010. Control of the morphology and size of magnetite particles with peptides mimicking the Mms6 protein from magnetotactic bacteria. J Colloid Interf Sci 343:65–70. Bailey JE, Ollis DF. 1986. Biochemical engineering fundamentals. Singapore: McGraw-Hill Science/Engineering/Math. 928. Bauer S, Ziv E. 1976. Dense growth of aerobic bacteria in a bench-scale fermentor. Biotechnol Bioeng 18:81–94. Bazylinski DA, Frankel RB, Heywood BR, Mann S, King JW, Donaghay PL, Hanson AK. 1995. Controlled biomineralization of magnetite (Fe3O4)

and gregite (Fe3S4) in a magnetotactic bacterium. Appl Environ Microbiol 61:3232–3239. Blakemore RP. 1975. Magnetotactic bacteria. Science 190:377–379. Blakemore RP, Maratea D, Wolfe RS. 1979. Isolation and pure culture of a freshwater magnetic spirillum in chemically defined medium. J Bacteriol 140:720–729. Blakemore RP, Short KA, Bazylinski Rosenblatt C, Frankel RB. 1985. Microaerobic conditions are required for magnetite formation within Aquaspirillum magnetotacticum. Geomicrobiol J 4:53–71. Calugay RJ, Miyashita H, Okamura Y, Matsunaga T. 2003. Siderophore production by the magnetic bacterium Magnetospirillum magneticum AMB-1. FEMS Microbiol Lett 218:371–375. Calugay RJ, Takeyama H, Mukoyama D, Fakuda Y, Suzuki T, et al. 2006. Catechol siderophore excretion by magnetotactic bacterium Magnetospirillum magneticum AMB-1. J Biosci Bioeng 101:445–447. Catalado DA, Haroon M, Schrader LE, Youngs VL. 1975. Rapid colorimetric determination of nitrate in plant tissues by nitration of salicylic acid. Comm Soil Sci Plant Ana 6:71–80. Doran PM. 1995. Bioprocess engineering principles. Academic Press. 439. Faivre D, Schu¨ler D. 2008. Magnetotactic bacteria and magnetosomes. Chem Rev 108:4875–4898. Faivre D, Boettger L, Matzanke B, Schu¨ler D. 2007. Intracellular magnetite biomineralization in bacteria proceeds via a distinct pathway involving membrane-bound ferritin and ferrous iron. Angewandte Chemie 119:8647–8652. Fogler HS. 2005. Elements of chemical reaction engineering, 4th edition. Englewood cliffs, New Jersey: Prentice Hall. 1080. Gorby YA, Beveridge TJ, Blakemore RP. 1988. Characterization of the bacterial magnetosome membrane. J Bacteriol 170:834–841. Grunberg K, Muller EC, Otto A, Reszka R, Linder D, et al. 2004. Biochemical and proteomic analysis of the magnetosome membrane in Magnetospirillum gryphiswaldense. Appl Environ Microbiol 70:1040– 1050. Harrison JS. 1967. Aspects of commercial yeast production. Process Biochem 2:41–45. Herbert D. 1976. Stoichiometric aspects of microbial growth. In: Dean ACR, Ellwood DC, Evans CGT, Melling J, editors. Continuous Culture 6: Applications and new fields. Chichester, England: Ellis Horwood. p 1–30. Heyen U, Schu¨ler D. 2003. Growth and magnetosome formation by microaerophilic Magnetospirillum strains in an oxygen-controlled fermentor. Appl Microbiol Biotechnol 61:536–544. Kok HED, Roels JA. 1980. Method for the statistical treatment of elemental and energy balances with application to steady-state continuousculture growth of Saccharomyces cerevisiae CBS 426 in the respiratory region. Biotechnol Bioeng 22:1097–1104. Komeili A, Vali H, Beveridge TJ, Newman DK. 2004. Magnetosome vesicles are present before magnetite formation and MamA is required for their activation. Proc Natl Acad Sci USA 101:3839–3844. Lang C, Schu¨ler D, Faivre D. 2007. Synthesis of magnetite nanoparticles for bio- and nanotechnology: Genetic engineering and biomimetics of bacterial magnetosomes. Macromol Biosci 7:144–151. Liu Y, Li GR, Guo FF, Jiang W, Li Y, Li LJ. 2010. Large-scale production of magnetosomes by chemostat culture of Magnetospirillum gryphiswaldense at high cell density. Microb Cell Factories 9:99 (doi: 10.1186/ 1475-2859-9-99). Mann S, Sparks NHC, Frankel RB, Bazylinski DA, Jannasch HW. 1990. Biomineralization of ferrimagnetic greigite (Fe3S4) and iron pyrite (FeS2) in a magnetotactic bacterium. Nature 343:258–261. Matsunaga T, Kamiya S. 1987. Use of magnetic particles isolated from magnetotactic bacteria for enzyme immobilization. Appl Microbiol Biotechnol 26:328–332. Matsunaga T, Kawasaki M, Yu X, Tsujimura N, Nakamura N. 1996a. Chemiluminescence enzyme immunoassay using bacterial magnetic particles. Anal Chem 68:3551–3554. Matsunaga T, Tsujimura N, Kamiya S. 1996b. Enhancement of magnetic particle production by nitrate and succinate fed-batch culture of Magnetospirillum sp. AMB-1. Biotechnol Tech 10:495–500.

Matsunaga T, Higashi Y, Tsujimura N. 1997. Drug delivery by magnetoliposomes containing bacterial magnetic particles. Cellular Eng 2: 7–11. Matsunaga T, Ueki F, Obata K, Tajima H, Tanaka T, Takeyama H, Goda Y, Fujimoto S. 2003. Fully automated immunoassay system of endocrine disrupting chemicals using monoclonal antibodies chemically conjugated to bacterial magnetic particles. Anal Chim Acta 475:75–83. Mayberry WR, Prochazka GJ, Payne WJ. 1968. Factors derived from studies of aerobic growth in minimal media. J Bacteriol 96:1424–1426. Meyers PA. 1997. Organic geochemical proxies of paleoceanographic, paleolimnologic, and paleoclimatic processes. Org Geochem 27:213– 250. Nakamura N, Hashimoto K, Matsunaga T. 1991. Immunoassay method for the determination of immunoglobin G using bacterial magnetic particles. Anal Chem 63:268–272. Naresh M, Gopinadhan K, Sekhar S, Juneja P, Sharma M, Mittal A. 2009. NSOM/HRTEM characterization of biologically derived cubo-octahedral nanomagnets. IEEE T Magn 45:4861–4864. Naresh M, Hasija V, Sharma M, Mittal A. 2010. Synthesis of cellular organelles containing nano-magnets stunts growth of magnetotactic bacteria. J Nanosci Nanotechnol 10:4135–4144. Naresh M, Sharma M, Mittal A. 2011. Intracellular magneto-spatial organization of magnetic organelles inside intact bacterial cells. J Biomed Nanotechnol 7:572–577. Paoletti LC, Blakemore RP. 1986. Hydroxamate production by Aquaspirillum magnetotacticum. J Bacteriol 167:73–76. Prozorov T, Mallapragada SK, Narasimhan B, Wang L, Palo P, et al. 2007. Protein-mediated synthesis of uniform superparamagnetic magnetite nanocrystals. Adv Funct Mater 17:951–957. Rockenberger J, Scher EC, Alivisatos AP. 1999. A new nonhydrolytic singleprecursor approach to surfactant-capped nanocrystals of transition metal oxides. J Am Chem Soc 121:11595–11596. Scheffel A, Schu¨ler D. 2007. The acidic repetitive domain of the Magnetospirillum gryphiswaldense MamJ protein displays hypervariability but is not required for magnetosome chain assembly. J Bacteriol 189:6437– 6446. Schu¨ler D, Baeuerlein E. 1996. Iron-limited growth kinetics of iron uptake in Magnetospirillum gryphiswaldense. Arch Microbiol 166:301–307. Schu¨ler D, Baeuerlein E. 1998. Dynamics of iron uptake and Fe3O4 biomineralization during aerobic and microaerobic growth of Magnetospirillum gryphiswaldense. J. Bacteriol 180:159–162. Sharma M, Naresh M, Mittal A. 2007. Morphological changes in magnetotactic bacteria in presence of magnetic fields. J Biomed Nanotechnol 3:75–80. Shimizu T, Furuki T, Waki T, Ichikawa K. 1978. Metabolic characteristics of denitrification by Paracoccus denitrificans. J Ferment Technol 56:207– 213. Shuler M, Kargi F. 2001. Bioprocess engineering: Basic concepts, 2nd edition. Englewood Cliffs, New Jersey: Prentice Hall. p 576. Staniland S, Ward B, Harrison A, Laan GVD, Telling N. 2007. Rapid magnetosome formation shown by real-time x-ray magnetic circular dichroism. Proc Natl Acad Sci USA 104:19524–19528. Staniland S, Williams WYN, Telling N, Laan GV, Harrison A, Ward B. 2008. Controlled cobalt doping of magnetosomes in vivo. Nat Nanotechnol 3:158–162. Staniland SS, Moisescu C, Benning LG. 2010. Cell division in magnetotactic bacteria splits magnetosome chain in half. J Basic Microbiol 50:1–5. Stockar UV, Liu J-S. 1999. Does microbial life always feed on negative entropy? Thermodynamic analysis of microbial growth. Biochimica et Biophysica Acta 1412:191–211. Stouthamer AH. 1977. Theoretical calculations on the influence of the inorganic nitrogen source on parameters for aerobic growth of microorganisms. Antonie van Leeuwenhoek 43:351–367. Sun JB, Duan JH, Dai SL, Ren J, Guo L, et al. 2008. Preparation and anti-tumor efficiency evaluation of Doxorubicin-loaded bacterial magnetosomes: Magnetic nanoparticles as drug carriers isolated from Magnetospirillum gryphiswaldense. Biotechnol Bioeng 101:1313– 1320.

Naresh et al.: Chemical Formula of M. gryphiswaldense Biotechnology and Bioengineering

1215

Suzuki T, Okamura Y, Calugay RJ, Takeyama H, Matsunaga T. 2006. Global gene expression analysis of iron-inducible genes in Magnetospirillum magneticum AMB-1. J Bacteriol 188:2275–2279. Takeyama H, Yamazawa A, Nakamura C, Matsunaga T. 1995. Application of bacterial magnetic particles as novel DNA carriers for ballistic transformation of a marine cyanobacterium. Biotechnol Tech 9:355– 360. Taoka A, Yoshimatsu K, Kanemori M, Fukumori Y. 2003. Nitrate reductase from the magnetotactic bacterium Magnetospirillum magnetotacticum MS-1: Purification and sequence analyses. Can J Microbiol 49:197– 206. van Dijken JPv, Harder W. 1975. Growth yields of microorganisms on methanol and methane. A theoretical study. Biotechnol Bioeng 17: 15–30.

1216

Biotechnology and Bioengineering, Vol. 109, No. 5, May, 2012

Wacker R, Ceyhan B, Alhorn P, Schu¨ler D, Lang C, Niemeyer CM. 2007. Magneto immuno-PCR: A novel immunoassay based on biogenic magnetosome nanoparticles. Biochem Biophys Res Commun 357: 391–396. Wang HY, Mou DG, Swartz JR. 1976. Thermodynamic evaluation of microbial growth. Biotechnol Bioeng 18:1811–1814. Yang CD, Takeyama H, Tanaka T, Matsunaga T. 2001. Effects of growth medium composition, iron sources and atmospheric oxygen concentrations on production of luciferase—Bacterial magnetic particle complex by a recombinant Magnetospirillum magneticum AMB-1. Enzyme Microbiol Tech 29:13–19. Yoza B, Matsumoto M, Matsunaga T. 2002. DNA extraction using bacterial magnetic particles in the presence of amino silane compound. J Biotechnol 94:217–224.

Supplementary Material To understand the calculation for the stoichiometric formula in detail, let us look at the time point of 12 h. At time 12 h, lactate consumed between 0 to 12h is 0.892 g Æ moles of lactate (C3H6O3) are given by 0.892/(molecular weight of potassium lactate) , i.e. a (Equation 1) = 6.97 x 10-3. Similarly, b1 = 3.36 x 10-3 moles, b2 = 1.37 x 10-2 moles, and, d = 2.35x10-5 moles at t = 12 h. Dissolved oxygen (DO) in the reactor medium was controlled at 6% (vol. by vol.). Air was purged into the reactor rapidly to achieve a DO of 6%. On consumption of oxygen (by the bacterial cells) leading to a decrease in DO value to 0.1% in the culture medium, air was rapidly purged into the reactor again till a DO value of 6% was achieved. The total number of air purging cycles was recorded. Further, calibration of DO (in %) resulting from air purging was done by considering that DO = 100% was achieved when oxygen concentration in the reactor medium was 8 ppm (mg/L). Thus, by simply multiplying the number of air purging cycles with the difference of 6% and 0.1% DO up to 12 h gave the moles of oxygen consumed (per liter), i.e. c = 7.79 x 10-4 moles. The moles of carbon dioxide produced during the bioreactor run at various time points was calculated using the experimentally determined relation n(t) = 1.52 x 10-14N(t) – 4.29 x 10-4, where n(t) = moles of carbon dioxide at time t hours and N = number of bacterial cells/Liter at time t hours. At 12 h, cell count was found to be 3.97x1011 cells/Liter. This gave a value of e = 6.58 x 10-13 moles, and g = 5.60 x 10-3 moles. Finally, at 12 h, dry cell weight was measured giving the value of y = 0.357 g/Liter. Thus, at 12 h, having experimentally determined a, b1, b2 c, d, e, g, Equations 8 – 13 were used to obtain the values of DEJGZ and f. Similarly, for all time points, the values of DEJGZ and f were obtained, as shown in the lower half of Table I. Here, it is important to note that the exponential phase for cell growth starts from t = 12 h onwards (see Figure 1). Prior to that (i.e. lag phase), application of Equation

1 is not reliable since bacterial cells are known to be acclimatizing to culture medium conditions in the lag phase. Therefore, the calculations for empirical coefficients in the chemical formula of M. gryphiswaldense are reliable only after the lag phase.