Team:Lethbridge/Results

From 2010.igem.org

(Difference between revisions)
(Catechol Degradation)
(Method)
 
(12 intermediate revisions not shown)
Line 130: Line 130:
=<font color="white">Catechol Degradation=
=<font color="white">Catechol Degradation=
-
The focus of our <html><a href="https://2010.igem.org/Team:Lethbridge/Project"><font color="#00DC00">project</font></a></html> is to decrease the toxicity of tailing pond water through bioremediation. We are specifically interested in the <html><a href="https://2010.igem.org/Team:Lethbridge/Project/Catechol_Degradation"><font color="#00DC00">degradation of the toxic molecule catechol</font></a></html> into 2-hydroxymuconate semialdyhyde (2-HMS); a bright yellow substrate that can be metabolized by the cell. This conversion is accomplished by a protein called catechol 2,3-dioxygenase (xylE).
+
The focus of our <html><a href="https://2010.igem.org/Team:Lethbridge/Project"><font color="#00DC00">project</font></a></html> is to decrease the toxicity of tailings pond water through bioremediation. We are specifically interested in the <html><a href="https://2010.igem.org/Team:Lethbridge/Project/Catechol_Degradation"><font color="#00DC00">degradation of the toxic molecule catechol</font></a></html> into 2-hydroxymuconate semialdyhyde (2-HMS); a bright yellow substrate that can be metabolized by the cell. This conversion is accomplished by a protein called catechol 2,3-dioxygenase (xylE).
==<font color="white">Characterization of Catechol 2,3-Dioxygenase==
==<font color="white">Characterization of Catechol 2,3-Dioxygenase==
Line 200: Line 200:
Catechol 2,3-dioxygenase contains an iron molecule in its active site. Our hypothesis was that the iron molecule is oxidized after converting a single catechol molecule to 2-HMS; rendering the catechol 2,3-dioxygenase inactive. We also want to test how catechol 2,3-dioxygenase would behave <i>in vitro</i> compared to <i>in vivo</i>.
Catechol 2,3-dioxygenase contains an iron molecule in its active site. Our hypothesis was that the iron molecule is oxidized after converting a single catechol molecule to 2-HMS; rendering the catechol 2,3-dioxygenase inactive. We also want to test how catechol 2,3-dioxygenase would behave <i>in vitro</i> compared to <i>in vivo</i>.
<br><br>
<br><br>
-
To test this hypothesis we grew <i>E. coli</i> containing BBa_K118021 in 500 mL of LB media with ampicillin. <i>E.coli</i> containing the pUC19 plasmid were also grown in 500 mL of LB media with ampicillin to act as a negative control throughout the course of the experiment. The cells were grown to a final optical density of 4.55 AU (measured at 600 nm). The cells were first spun down at 3800 rcf for 5 minutes. The supernatant was decanted and the cells were re-suspended in 40 mL M9 minimal media and incubated with 10 mg of lysozyme for 10 minutes. This cell extract was spun down at 10000 rcf for 30 minutes. The supernatant was taken off and spun at 30000 rcf (S30) for 1 hour. The supernatant of the S30 samples was divided. Half the samples were flash frozen in liquid nitrogen and stored at -80<sup>o</sup>C. The second half was spun at 100000 rcf for 45 minutes.  
+
To test this hypothesis we grew <i>E. coli</i> containing BBa_K118021 in 500 mL of LB media with ampicillin. <i>E.coli</i> containing the pUC19 plasmid were also grown in 500 mL of LB media with ampicillin to act as a negative control throughout the course of the experiment. The cells were grown to a final optical density of 4.55 AU (measured at 600 nm). The cells were first spun down at 3800 rcf for 5 minutes. The supernatant was decanted and the cells were re-suspended in 40 mL M9 minimal media and incubated with 10 mg of lysozyme for 10 minutes. This cell extract was spun down at 10000 rcf for 30 minutes. The supernatant was taken off and spun at 30000 rcf (S30) for 1 hour. The supernatant of the S30 samples was divided. Half the samples were flash frozen in liquid nitrogen and stored at -80<sup>o</sup>C. The second half was spun at 100000 rcf for 45 minutes (S100).  
<br><br>
<br><br>
-
In order to measure the 2-HMS production we needed to determine the concentration of protein per volume of sedimentation material. To determine the total protein concentration of the S30 and S100 extracts a Bradford assay was conducted using a standard curve of BSA. Concentrations of the negative controls (<i>Escherichia coli</i> DH5&alpha; hosting the pUC19 plasmid) for each of the S30 and S100 samples were also determined. Absorbance readings were taken at a 595 nm wavelength and concentration reported in µg/mL.
+
In order to measure the 2-HMS production we needed to determine the concentration of protein per volume of cell extract. To determine the total protein concentration of the S30 and S100 extracts a Bradford assay was conducted using a standard curve of BSA. Concentrations of the negative controls (<i>Escherichia coli</i> DH5&alpha; hosting the pUC19 plasmid) for each of the S30 and S100 samples were also determined. Absorbance readings were taken at a 595 nm wavelength and concentration reported in µg/mL.
<br><br>
<br><br>
===<font color="white">Results of the Bradford Assay===
===<font color="white">Results of the Bradford Assay===
<hr>
<hr>
-
We determined the concentrations of S30, S30 negative control, S100, and S100 negative control to be 779 µg/mL, 104.5 µg/mL, 519.3 µg/mL, and 100.1 µg/mL respectively.
 
<br><br>
<br><br>
-
Protein concentrations of S30 and S100 extracts of xylE cells and pUC19 cells were normalized in 1mL of either 20mM Tris pH 8.0 or double distilled water, and incubated in a 0.05 mM catechol solution. The subsequent production of 2-HMS was observed as a function of time. This observation was measured by recording absorbance readings at 375 nm over a ten minute period immediately after the introduction of catechol.  
+
<table><table border="3">
 +
<tr><td>Sample<td>Concentration
 +
<tr><td>S30<td>779&micro;g/mL
 +
<tr><td>S30 Control<td>104.5&micro;g/mL
 +
<tr><td>S100<td>519.3&micro;g/mL
 +
<tr><td>S100 Control<td>100.1&micro;g/mL
 +
</table>
 +
 
 +
<br><br>
 +
Protein mass of S30 and S100 (and S30 Control/S100 Control) extracts were normalized to between 2&micro;g and 10&micro;g in 1mL of either 20mM Tris pH 8.0 or double distilled water. Catechol was added to normalized cell extract at a final concentration of 0.05 mM. The subsequent production of 2-HMS was observed as a function of time by recording absorbance at 375 nm over the ten minute period immediately following the addition of catechol.  
<br><br>
<br><br>
===<font color="white">Results===
===<font color="white">Results===
<hr>
<hr>
-
We did not observe a difference in the catechol production in samples containing Tris vs. samples containing d<sub>2</sub>H<sub>2</sub>O. <font color="red" size="+2">Justin and Adam to fill in!</font>
+
Figure 3 displays that catechol 2,3-dioxygenase can degrade catechol <i>in vitro</i>, producing 2-HMS. We also observe that when approximately twice the amount of S30 extract is added to an excess catechol solution, production of 2-HMS is increased approximately twofold. This suggests that the enzyme (rather than substrate) is the limiting component in catechol degradation by catechol 2,3-dioxygenase. This result corresponds with our hypothesis that the catechol 2,3-dioxygenase active site iron is reduced upon catalysis, rendering our enzyme inactive. The samples in Figure 3 were carried out in double distilled water. However, duplicates of the readings were measured in a buffer (20mM Tris pH 8.0). We do not observe a significant difference in 2-HMS production in samples containing buffer compared to samples containing double distilled water.  
<br><br>
<br><br>
<html>
<html>
Line 223: Line 231:
<tr>
<tr>
<th>
<th>
-
<img src="https://static.igem.org/mediawiki/2010/2/2d/UofLcatecholfigure4.jpg" width="450"/>
+
<img src="https://static.igem.org/mediawiki/2010/b/b7/Catecholgood.jpg" width="450"/>
</th>
</th>
</tr>
</tr>
Line 230: Line 238:
<tr><th colspan="1"  align="left">
<tr><th colspan="1"  align="left">
<p>
<p>
-
<font color="white">Figure 3.  <font color="red" size="+2">Justin and Adam to fill in!</font>
+
<font color="white">Figure 3.  The production of 2-hydroxymuconate semialdehyde by different concentrations of S30 extract. Green and red lines are 2&micro;g and 4&micro;g respectively of S30 cell extract from cells expressing catechol 2,3-dioxygenase. The blue line is 4&micro;g of control S30 cell extract.
</th>
</th>
</tr>
</tr>
Line 241: Line 249:
===<font color="white">Conclusion===
===<font color="white">Conclusion===
<hr>
<hr>
-
<font color="red" size="+2">Justin and Adam to fill in!</font>
+
We have successfully show that our <i>E. coli</i> cells carrying the K118021 biobrick is capable of converting catechol into 2-HMS. Moreover, the reduction Figure 3 shows a reduction of 2-HMS concentration in the solution. It is more than likely, considering that 2-HMS can be metabolized by the cell, that the soluble metabolic machinery of <i>E. coli</i> is metabolizing 2-HMS into its breakdown products. This is a feature of the system that we intend to exploit in our remediation project. <br>
 +
 
 +
With this in mind, it is inefficient to utilize a single turnover enzyme such as catechol 2,3-dioxygenase in an industrial setting. We have identified a protein that will reduce the iron that is oxidized in the conversion of catechol to 2-HMS. The gene coding for this enzyme (<i>xylT</i>) codes for a ferredoxin protein, and is located on the same operon as <i>xylE</i> in <i>Pseudomonas putida</i><sup>1</sup></i>. <br>
 +
 
 +
We intend to introduce ferredoxin to our bioremediation project next year with the aim of increasing catechol 2,3-dioxygenase efficiency by allowing it to catalyze multiple turnovers of catechol to 2-HMS<br>
<br><br>
<br><br>
 +
 +
===<font color="white">References===
 +
<hr>
 +
<sup>1</sup>Polissi A., Harayama S. (1993) <b>In vivo reactivation of catechol 2,3-dioxygenase mediated by a chloroplast-type ferredoxin: a bacterial strategy to expand the substrate specificity of aromatic degradative pathways.</b>. <i>EMBO J.</i>, <b>12</b>(8); 3339-3347
=<font color="white">Compartmentalization Parts=
=<font color="white">Compartmentalization Parts=

Latest revision as of 02:03, 28 October 2010




Check out these important project links!


Contents

Catechol Degradation

The focus of our project is to decrease the toxicity of tailings pond water through bioremediation. We are specifically interested in the degradation of the toxic molecule catechol into 2-hydroxymuconate semialdyhyde (2-HMS); a bright yellow substrate that can be metabolized by the cell. This conversion is accomplished by a protein called catechol 2,3-dioxygenase (xylE).

Characterization of Catechol 2,3-Dioxygenase

Characterized Parts


BBa_K118021

Hypothesis


The addition of the catechol to a suspension of cells expressing catechol 2,3-dioxygenase will result in the production of 2-HMS.

Method


BBa_K118021 was transformed into Escherichia coli DH5α (E. coli) cells using our transformation protocol. We confirmed that the cells hosted the plasmid containing the BBa_118021 part and we begun experiments to characterize catechol degredation.

In our first experiment we grew a 5 mL culture of our engineered E. coli cells in M9 minimal media overnight. The cells were spun down at 14000 rfc for 2 minutes. We raised the catechol concentration of this solution to 100 mM, the solution immediately turned bright yellow.

Figure 1. Left: M9 media containing 100 mM catechol. The solution contains E. coli cells hosting the pUC19 plasmid. Right: M9 media containing 100 mM catechol. The solution contains E. coli cells hosting part BBa_118021. The yellow colour suggests the production of 2-HMS.

In our second experiment we wanted to measure the absorbance of 2-hydroxymuconate semialdehyde over time. We grew a 5 mL culture of our engineered E. coli cells in M9 media overnight. 1 mL of the cell solution was removed for analysis. Catechol was added to a final concentration of 100mM in the cell suspension. Immediately the formation of 2-HMS was tracked. The formation of 2-HMS can easily be tracked, as it absorbs light at 375 nm.

Figure 2. Production of 2-HMS over time.

Catechol 2,3-dioxygenase contains an iron molecule in its active site. Our hypothesis was that the iron molecule is oxidized after converting a single catechol molecule to 2-HMS; rendering the catechol 2,3-dioxygenase inactive. We also want to test how catechol 2,3-dioxygenase would behave in vitro compared to in vivo.

To test this hypothesis we grew E. coli containing BBa_K118021 in 500 mL of LB media with ampicillin. E.coli containing the pUC19 plasmid were also grown in 500 mL of LB media with ampicillin to act as a negative control throughout the course of the experiment. The cells were grown to a final optical density of 4.55 AU (measured at 600 nm). The cells were first spun down at 3800 rcf for 5 minutes. The supernatant was decanted and the cells were re-suspended in 40 mL M9 minimal media and incubated with 10 mg of lysozyme for 10 minutes. This cell extract was spun down at 10000 rcf for 30 minutes. The supernatant was taken off and spun at 30000 rcf (S30) for 1 hour. The supernatant of the S30 samples was divided. Half the samples were flash frozen in liquid nitrogen and stored at -80oC. The second half was spun at 100000 rcf for 45 minutes (S100).

In order to measure the 2-HMS production we needed to determine the concentration of protein per volume of cell extract. To determine the total protein concentration of the S30 and S100 extracts a Bradford assay was conducted using a standard curve of BSA. Concentrations of the negative controls (Escherichia coli DH5α hosting the pUC19 plasmid) for each of the S30 and S100 samples were also determined. Absorbance readings were taken at a 595 nm wavelength and concentration reported in µg/mL.

Results of the Bradford Assay




SampleConcentration
S30779µg/mL
S30 Control104.5µg/mL
S100519.3µg/mL
S100 Control100.1µg/mL



Protein mass of S30 and S100 (and S30 Control/S100 Control) extracts were normalized to between 2µg and 10µg in 1mL of either 20mM Tris pH 8.0 or double distilled water. Catechol was added to normalized cell extract at a final concentration of 0.05 mM. The subsequent production of 2-HMS was observed as a function of time by recording absorbance at 375 nm over the ten minute period immediately following the addition of catechol.

Results


Figure 3 displays that catechol 2,3-dioxygenase can degrade catechol in vitro, producing 2-HMS. We also observe that when approximately twice the amount of S30 extract is added to an excess catechol solution, production of 2-HMS is increased approximately twofold. This suggests that the enzyme (rather than substrate) is the limiting component in catechol degradation by catechol 2,3-dioxygenase. This result corresponds with our hypothesis that the catechol 2,3-dioxygenase active site iron is reduced upon catalysis, rendering our enzyme inactive. The samples in Figure 3 were carried out in double distilled water. However, duplicates of the readings were measured in a buffer (20mM Tris pH 8.0). We do not observe a significant difference in 2-HMS production in samples containing buffer compared to samples containing double distilled water.

Figure 3. The production of 2-hydroxymuconate semialdehyde by different concentrations of S30 extract. Green and red lines are 2µg and 4µg respectively of S30 cell extract from cells expressing catechol 2,3-dioxygenase. The blue line is 4µg of control S30 cell extract.

Conclusion


We have successfully show that our E. coli cells carrying the K118021 biobrick is capable of converting catechol into 2-HMS. Moreover, the reduction Figure 3 shows a reduction of 2-HMS concentration in the solution. It is more than likely, considering that 2-HMS can be metabolized by the cell, that the soluble metabolic machinery of E. coli is metabolizing 2-HMS into its breakdown products. This is a feature of the system that we intend to exploit in our remediation project.

With this in mind, it is inefficient to utilize a single turnover enzyme such as catechol 2,3-dioxygenase in an industrial setting. We have identified a protein that will reduce the iron that is oxidized in the conversion of catechol to 2-HMS. The gene coding for this enzyme (xylT) codes for a ferredoxin protein, and is located on the same operon as xylE in Pseudomonas putida1</i>.

We intend to introduce ferredoxin to our bioremediation project next year with the aim of increasing catechol 2,3-dioxygenase efficiency by allowing it to catalyze multiple turnovers of catechol to 2-HMS


References


1Polissi A., Harayama S. (1993) In vivo reactivation of catechol 2,3-dioxygenase mediated by a chloroplast-type ferredoxin: a bacterial strategy to expand the substrate specificity of aromatic degradative pathways.. EMBO J., 12(8); 3339-3347

Compartmentalization Parts

One of the sub-projects for the bioremediation of the tailings ponds is to create synthetic microcompartments that we can then use to isolate various pathways within an Escherichia coli cell. To do this we need to have a microcompartment as well as a means to characterize the compartment so that the system can be optimized. Here are the experiments we have performed so far towards the characterization of the microcomparments.

Placement of Oligoarginine Tail on Proteins

Characterized Parts


BBa_K249004

BBa_K249005

BBa_K331030

BBa_K331031

Hypothesis


The placement of an oligoarginine sequence at the N-terminus of a protein will destabilize the protein in vivo.

Introduction


The long term goal of our team is to utilize an oligoarginine tail to specifically target enzymes into a microcompartment composed of modified lumazine synthase subunits. While conducting background research on the project, we came upon data originally reported by Bachmair et al.1 suggesting that the identity of the amino acid at the N-terminus of a protein is related to its half-life, and mostly notably, that arginine residues at the are destabilizing. This data suggests that by placing an arginine at the N-terminus of a protein to be targeted into a lumazine synthase microcompartment would cause degradation of our protein before it can be moved into the microcompartment.

We chose to investigate the how the placement of an oligoarginine sequence affects the stability of the protein to which it is fused.

Method


In order to further characterize the C-terminal and N-terminal oligoarginine tag (BioBricks BBa_K249005 and BBa_K249004 respectively) and investigate the effect their placement on protein stability, yellow fluorescent proteins (YFP) with the oligoarginine fused to either the C-terminus (BBa_K331023) or N-terminus (BBa_K331022) (and preceded by a ribosomal binding site – BBa_B0034) were synthesized. We used our Red/White 3-Antibiotic assembly method to add a tetracycline repressible promoter (BBa_R0040) for constitutive expression of the fusion protein. This addition generated BioBricks BBa_K331031 and BBa_K331030 for the C-terminal tagged and N-terminal tagged YFP respectively.

The BioBrick containing plasmid was transformed into Escherichia coli DH5α cells. These cells were grown to an OD600 of approximately 0.7, and diluted 1:10 with MilliQ H2O immediately prior to analysis by fluorescent spectroscopy.

This dilution of cells was excited at 517 nm, and the emission spectra was read from 522 nm to 650 nm. Fluorescence at 524 nm (emission maxima of YFP) of control cells (Escherichia coli DH5α), N-terminal tagged, and C-terminal tagged YFP were compared.

Results


N-terminal tagged YFP did not have substantially more fluorescence than control cells. Cells expressing C-terminal tagged YFP had ten times more fluorescence than control cells and cells expressing N-terminal tagged YFP. Lethbridge NvsC-terminalOligoArgBlackfINAL.png

Conclusion


Our results showing that the placement of arginine residues at the N-terminus of our YFP results in no observable fluorescence over control cells are consistent with the general N-terminal rule reported by Bachmair et al. Assuming that transcription of this K331030 and K331031 are equivalent, these data suggest that the N-terminal oligoarginine is reducing the half-life of the protein to which it is fused, ie YFP.

Reference


1Bachmair A., Finley D., Varshavsky A. (1986), In Vivo Half-Life of a Protein Is a Function of Its Amino-Terminal Residue. Science 234. 4773 179-186.

Characterization of C-terminal Oligo Arginine Cyan Fluorescent Protein

Characterized Parts


BBa_K331033

Hypothesis


Light will be observed at 476 nm when cells containing BBa_K331033 are illuminated with light at 439 nm as a result of the excitation and subsequent fluorescence of Cyan Fluorescent Protein (CFP).

Introduction


Prior to introducing a oligoarginine tail onto the catechol-2,3-dioxygenase protein with the intention of directing the fusion protein into a lumazine synthase microcompartment, there must be some confirmation that oligoarginine tail is indeed causing the protein to which it is attached to localize within the microcompartment. We chose to utilize a technique known as fluorescent resonance energy transfer1 (FRET) to verify localization. In FRET, when a dye pair comes into close proximity to one another, one dye (the donor) will transfer some of its energy to the second dye (the acceptor). This will cause the acceptor to fluoresce without direct excitation. The dye pair we have chose is CFP as the donor and yellow fluorescent protein (YFP) as the acceptor.

Here we investigate the ability of BBa_K331033 to produce CFP and emit light at its characteristic emission wavelength.

Method


In order to characterize the tetracycline repressible CFP (BioBrick BBa_K331033), cyan fluorescent protein (CFP-BBa_E0020) with an oligoarginine tag fused to the C-terminus (BBa_K331025) (and preceded by a ribosomal binding site – BBa_B0034) was synthesized. We used our Red/White 3-Antibiotic assembly method to add a tetracycline repressible promoter (BBa_R0040) for constitutive expression of the fusion protein. This construction yielded BioBrick BBa_K331033.

The BioBrick containing plasmid was transformed into Escherichia coli DH5α cells. These cells were grown to an OD600 of approximately 0.7 in LB media, and diluted 1:10 with MilliQ H2O immediately prior to analysis by fluorescent spectroscopy.

This dilution of cells was excited at 439 nm, and the emission spectra was read from 444 nm to 650 nm.

Results


Fluorescence at 476 nm was observed. This peak, along with a shoulder occurring at approximately 510 nm is consistent with results obtained by McRae1 et al. in their rapid purification of ECFP.

UofLcfpfigureblack.jpg


Conclusion


The BioBrick BBa_K331033 that we constructed generates CFP in the absence of tetracycline, as expected.

Future Directions


We will assembly this into a single biobrick which can differentially express CFP/YFP and lumazine synthase for confirmation (through FRET) that the oligoarginine tail can be utilized to localize proteins within a modified lumazine synthase microcompartment.

Reference


1Förster T. (1948), Zwischenmolekulare Energiewanderung und Fluoreszenz. Annalen der Physik, 437: 55-75.

2McRae S.R., Brown C.L., Bushell G.R. (2005), Rapid Purification of EGFP, EYFP, and ECFP with high yield and purity. Protein Expression and Purification 41. 1 121-127.

Magnetic Nanoparticles Parts

One of the sub-projects for the bioremediation of the tailings ponds is to reduce heavy metals to create magnetic nanoparticles that can then be removed from the pond. To do this we need to be able to produce Mms6 (the iron reducing protein) and show that it can successfully produce the nanoparticles within the Escherichia coli cell. He is the work we have accomplished so far to characterize Mms6.

Characterization of Mms6

Characterized Parts


BBa_K249019

Hypothesis


Induction of Mms6 protein overexpression in Escherichia coli DH5α cells will cause a change in growth.

Introduction


A parallel project we are undertaking in oilsands remediation is in the clean-up of heavy metal contaminants utilizing the Mms6 protein expressed by Magnetiospirilum magneticum1. The Mms6 protein is capable of reducing aqueous iron into magnetite nanoparticles. The magnetite nanoparticles can be removed with much greater ease than aqueous iron.

We chose to investigate how the presence of the Mms6 protein affected Escherichia coli DH5α, its bacterial host cell.

Method


In order to characterize the effect of Mms6 production on Escherichia coli DH5α cells, we utilized BioBrick BBa_K249019, a part submitted by the 2009 Lethbridge iGEM team. BBa_K249019 is the Mms6 coding region (BBa_K249016) with a ribosomal binding side (BBa_B0030) and a double terminator (BBa_B0015) under the control of a lactose inducible promoter ( BBa_R0010).

We inoculated two 500 mL cultures of LB media with Escherichia coli DH5α cells containing BBa_K249019 and grew the cells until they reached approximately 0.6 OD600 units. At this point, expression of the Mms6 protein was induced in one of the 500 mL cultures by the addition of 1 mM Isopropyl β-D-1-thiogalactopyranoside (IPTG). One milliliter of cells from each sample was removed at one hour intervals and the optical density at 600 nm was recorded and plotted.

Results


The addition of IPTG to cells containing the BBa_K249019 BioBrick caused a slowdown of cell growth after 2 hours, and a subsequent plateau at approximately one-half of the cells density that uninduced cells reached after three hours post-induction.

UofLmms6growth.jpg


Conclusion


The slowdown in cell growth as a result of production of Mms6 protein could be attributed to a number of factors.

One potential explanation could be a significant diversion of cellular resources to the production of Mms6 protein. This is unlikely, as subsequent analysis of cell fractions with SDS-PAGE showed no increase in banding at the expected molecular weight.

Another explanation could be that the Mms6 protein is toxic to the cells, and causes them to die. This possibility is being explored in collaboration with the Calgary team by using their troubleshooting kit.

References


1Arakaki 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. Journal of Colloid and Interface Science. 343:65-60.