Difference between revisions of "Team:Newcastle/Results/Chemotaxis"

Line 198: Line 198:
 
             <font size="2">Figure 4: Absorbance at 600 nm of four bacterial species (<i>A. brasilense</i>, <i>A.  caulinodans</i>, <i>H. seropedicae</i>, and <i>E. coli</i>) after 24 hours of growth when grown in liquid media containing different concentrations of naringenin. </font>
 
             <font size="2">Figure 4: Absorbance at 600 nm of four bacterial species (<i>A. brasilense</i>, <i>A.  caulinodans</i>, <i>H. seropedicae</i>, and <i>E. coli</i>) after 24 hours of growth when grown in liquid media containing different concentrations of naringenin. </font>
 
              
 
              
 +
            <p> </p>
 
             <p>All species successfully grew in the presence of 0-150 μM naringenin (Figure 4). However, it was noted that E. coli showed a reduced growth rate even at lower concentrations of naringenin. When the concentration of naringenin exceeded 100 μM, there exists greater flux in all species suggesting that naringenin begins to have a greater impact on some, but not all, bacteria. As such, naringenin concentrations of <100 μM were used as part of subsequent chemotaxis assays to avoid negatively impacting bacterial growth.  </p>
 
             <p>All species successfully grew in the presence of 0-150 μM naringenin (Figure 4). However, it was noted that E. coli showed a reduced growth rate even at lower concentrations of naringenin. When the concentration of naringenin exceeded 100 μM, there exists greater flux in all species suggesting that naringenin begins to have a greater impact on some, but not all, bacteria. As such, naringenin concentrations of <100 μM were used as part of subsequent chemotaxis assays to avoid negatively impacting bacterial growth.  </p>
 
             </div>
 
             </div>
Line 266: Line 267:
 
            
 
            
 
         <tr>
 
         <tr>
−
           <td><i>E. coli (DH5α)</i></td>
+
           <td><i>E. coli</i></td>
 
           <td>0</td>
 
           <td>0</td>
 
           <td>0</td>
 
           <td>0</td>
Line 276: Line 277:
 
     </table>
 
     </table>
 
      
 
      
 +
    <p> </p>
 
     <p>After 24 hours incubation at either 30 °C (A. caulinodans and H. seropedicae) or 37 °C (A. brasilense and E. coli), the number of colonies which grew on the LB agar plate was counted (Table 2). The results showed that of the four test bacterial species, only one was able to move into the capillary. This species was H. seropedicae which was able to move successfully into capillaries containing either the control (buffer solution) or the chemoattractant. This was demonstrated by the growth of colonies on LB agar from the contents of each capillary (Figure 5). Both methods of agar inoculation (spreading and pipetteing) lead to colony growth.</p>
 
     <p>After 24 hours incubation at either 30 °C (A. caulinodans and H. seropedicae) or 37 °C (A. brasilense and E. coli), the number of colonies which grew on the LB agar plate was counted (Table 2). The results showed that of the four test bacterial species, only one was able to move into the capillary. This species was H. seropedicae which was able to move successfully into capillaries containing either the control (buffer solution) or the chemoattractant. This was demonstrated by the growth of colonies on LB agar from the contents of each capillary (Figure 5). Both methods of agar inoculation (spreading and pipetteing) lead to colony growth.</p>
 
      
 
      
Line 306: Line 308:
 
<font size="2">Figure 6: Example of haemocytometer square at 40x objective containing <i>H. seropedicae</i> utilised for cell counting, cells along the bottom and/or right lines were not counted to avoid double counting. </font>
 
<font size="2">Figure 6: Example of haemocytometer square at 40x objective containing <i>H. seropedicae</i> utilised for cell counting, cells along the bottom and/or right lines were not counted to avoid double counting. </font>
  
 +
<p> </p>
 
<p>The CD:OD index was produced utilising data collected from a haemocytometer. A haemocytometer is a specialised microscopy slide of a known volume, it also contains a grid at the centre. By counting the number of cells in 16 squares at the top right (Figure 6) and performing a series of mathematical calculations (<a href="https://2018.igem.org/Team:Newcastle/Protocols">[x]</a>, we were able to determine cell density. By utilising a spectrophotometer, we were also able to take a reading of the absorbance (600 nm) and thus link the two together (Table 3).</p>
 
<p>The CD:OD index was produced utilising data collected from a haemocytometer. A haemocytometer is a specialised microscopy slide of a known volume, it also contains a grid at the centre. By counting the number of cells in 16 squares at the top right (Figure 6) and performing a series of mathematical calculations (<a href="https://2018.igem.org/Team:Newcastle/Protocols">[x]</a>, we were able to determine cell density. By utilising a spectrophotometer, we were also able to take a reading of the absorbance (600 nm) and thus link the two together (Table 3).</p>
 +
             
 +
<font size="2">Table 3: Cell density (cells.ml<sup>-1</sup>) of <i>A. brasilense</i>, <i>A.  caulinodans</i>, <i>H. seropedicae</i> and <i>E. coli</i> at different optical densities and as a ratio of OD<sub>600</sub>=1</font>             
 +
               
 +
                <table id="protocols">
 +
      <thead>
 +
        <tr>
 +
          <th>Species (Strain)</th>
 +
          <th>Optical Density</th>
 +
          <th>Cell Density</th>
 +
          <th>Colony Count (Control)</th>
 +
        </tr>
 +
      </thead>
 +
      <tbody>
 +
        <tr>
 +
          <td><i>A. caulinodans</i> (ORS571)</td>
 +
          <td>0</td>
 +
          <td>0</td>
 +
          <td>0</td>
 +
        </tr>
 +
        <tr>
 +
          <td><i>A. brasilense</i> (SP245)</td>
 +
  <td>0</td>
 +
          <td>0</td>
 +
          <td>0</td>
 +
       
 +
        </tr>
 +
        <tr>
 +
          <td><i>H. seropedicae</i> (Z67)</td>
 +
          <td>145.33</td>
 +
          <td>85.58</td>
 +
          <td>109.78</td>
 +
          </tr>
 +
         
 +
        <tr>
 +
          <td><i>E. coli (DH5α)</i></td>
 +
          <td>0</td>
 +
          <td>0</td>
 +
          <td>0</td>
 +
 +
          </tr>
 +
     
 +
    </table>
 +
   
 +
    <p>Utilising a haemocytometer to count cells also allows observations of cell morphology and behaviour. This was used to an advantage as it allowed us to explore whether our bacteria’s morphology aligns with the literature that was followed to produce the community model. It also enabled observations of motility which was important after theories that the species were no longer motile which is why only <i>H. seropedicae</i> showed movement into the capillary during the attempt to quantify chemotaxis. </p>
 
                 </div>
 
                 </div>
 
 

Revision as of 16:51, 15 October 2018

Alternative Roots

Exploring Bacterial Chemotaxis

Introduction Colony Morphology
Growth in LB Naringenin & Growth
Quantification Microscopy Chemotaxis on Agar
Scroll Down

Naringenin Chemotaxis

An Introduction

We examined how three species of free-living nitrogen-fixing bacteria respond to the presence of the flavonoid naringenin. The three species, Azorhizobium caulinodans (ORS571), Azospirillum brasilense (SP245), and Herbaspirillum seropedicae (Z67), were selected because they all have potential to form different types of interactions with plant roots. A. caulinodans has been shown to fix nitrogen both as a free-living microbe and when in symbiosis with the semi-aquatic leguminous tree Sesbania rostrata [1]. H. seropedicae is a root endophyte and has shown potential to colonise popular crops such as wheat and maize [2].

Bacteria Characterisation

Colony Morphology

Before commencing chemotaxis studies, we needed to understand the growth characteristics of the three free-living nitrogen-fixing bacteria to be used in our project. We first examined the colony morphology of these three species in the absence of any chemoattractants. Familiarisation with the bacteria allows identification of abnormal behaviour and contamination. For colony morphology, the size after a minimum of 24 hours and morphology (shape and pigmentation) was recorded (Table 1, Figs 1-3).

Table 1: Qualitative analysis of Azorhizobium caulinodans, Azospirillum brasilense, Herbaspirillum seropedicae colonies grown on solid media.
Species (Strain) Colony Pigmentation Colony Morphology Points of Interest
Azorhizobium caulinodans (ORS571) White Regular form, Typically raised, Entire margin Colonies rarely grow to a measurable size when grown at 30˚c on YEB media after 24 hours
Azospirillum brasilense (SP245) Orange/Pink Non-slimy, Regular and round form, Entire margins Both immature and dead colonies lack the orange/pink pigment, Colonies wrinkle with age
Herbaspirillum seropedicae (Z67) Cream/Light Green Circular or Irregular form (occasionally rhizoid), Raised elevation, Shiny Colonies took on a different morphology depending on how the media was innoculated; stab-innoculation lead to rhizoid form while spreading leads to circular/irregular form

Azorhizobium caulinodans (Figure 1): Colonies do not grow to a measurable size within 24 hours at 30 ˚C on Yeast Extract Broth agar. Colonies contain white pigmentation and are raised in elevation with an entire margin – a continuous, uninterrupted border of the colony. Colonies rarely grow larger than 2 mm whilst smaller colonies, which are much more numerous, could not be accurately measured.

Azospirillum brasilense (Figure 2): Colonies are distinguishable by their distinctive orange/pink pigmentation though both immature and dead colonies lack this pigmentation. Older colonies became ingrained into the agar, making them hard to remove without damaging the agar. Older colonies also began to wrinkle with time. The average diameter for a colony of this species after 24 hours incubation at 37 ˚C on Tryptone Soya Agar was 3 mm, making A. brasilense the fastest growing of our nitrogen-fixing bacteria. Young A. brasilense colonies were shiny, round and with entire margins. These young colonies may have some pigmentation near the centre as the colony matures. This is in contrast to older colonies which maintain a different phenotype; losing their shine and gaining the odd wrinkle. Wrinkling often leads to the loss of the round shape.

H. seropedicae (Figure 3): the colonies take different forms depending on how the plate is inoculated. If the plate is stab-inoculated, the colony takes a rhizoid appearance (Figure 3a). If the culture is spread across the plate, then it typically takes a circular or irregular form (Figure 3b). Colonies possess a green-cream pigmentation and are raised from the surface. Most colonies were shiny and typically 1.5 mm in diameter after 24 hours at 30 ˚C.

Growth Rates in Liquid Media

Sadiya's text

Effect of Naringenin on Growth Rate in Liquid Culture

Initial research for the Alternative Roots project noted that naringenin possesses antimicrobial properties, particularly towards E. coli [3] [link to the notebook data where you observed this]. As E. coli (DH5α) was to be used as both a control in our chemotaxis assays and as the organism in which our naringenin biosynthesis operon would first be assembled, it was deemed important to characterise the effect of increasing naringenin concentrations on growth rates of both our free-living nitrogen-fixing bacteria, and E. coli in LB medium. This was essential to guide the chemotaxis assays enabling an understanding of naringenin concentrations which would not have detrimental impacts upon the cell. If cell health is impaired, then there is potential for cell death to lead to the appearance of chemorepulsion. This is particularly problematic when applying the response index as a semi-quantitative measure of chemotactic response as the method utilises ratios between colony edges to determine the significance of chemotaxis [4].

Figure 4: Absorbance at 600 nm of four bacterial species (A. brasilense, A. caulinodans, H. seropedicae, and E. coli) after 24 hours of growth when grown in liquid media containing different concentrations of naringenin.

All species successfully grew in the presence of 0-150 μM naringenin (Figure 4). However, it was noted that E. coli showed a reduced growth rate even at lower concentrations of naringenin. When the concentration of naringenin exceeded 100 μM, there exists greater flux in all species suggesting that naringenin begins to have a greater impact on some, but not all, bacteria. As such, naringenin concentrations of <100 μM were used as part of subsequent chemotaxis assays to avoid negatively impacting bacterial growth.

Characterising Chemotactic Behaviour

Quantification Utilising Capillaries

To characterise chemotactic behaviour in response to naringenin, a quantitative approach is desirable. This allows for direct comparison of the strength of the response between different species. Results from a quantitative assay would also be better suited for our community model [link to modelling page] as it allows a ranking of bacterial responses to naringenin.

Table 2: Colony forming units of four bacterial species from capillaries containing 1 µl 100 µM naringenin or motility buffer solution (10 mM potassium phosphate, 0.1 mM EDTA, 10 mM glucose, pH 7.0) after 60 minutes open-end submersion in sterile conditions at room temperature/pressure. Values are mean cfu.μl-1. Difference between colony counts from capillaries containing naringenin or motility buffer was non-significant for all species (P>0.05).
Species (Strain) Colony Count (Naringenin) ± Standard Error Colony Count (Control) ± Standard Error Significant Difference
A. caulinodans (ORS571) 0 0 0 0 No
A. brasilense (SP245) 0 0 0 0 No
H. seropedicae (Z67) 145.33 85.58 109.78 117.44 No
E. coli 0 0 0 0 No

After 24 hours incubation at either 30 °C (A. caulinodans and H. seropedicae) or 37 °C (A. brasilense and E. coli), the number of colonies which grew on the LB agar plate was counted (Table 2). The results showed that of the four test bacterial species, only one was able to move into the capillary. This species was H. seropedicae which was able to move successfully into capillaries containing either the control (buffer solution) or the chemoattractant. This was demonstrated by the growth of colonies on LB agar from the contents of each capillary (Figure 5). Both methods of agar inoculation (spreading and pipetteing) lead to colony growth.

After counting colonies from the contents of both the control and naringenin capillaries, no significant difference between mean colony count of the two conditions was observed (P>0.05). The results therefore show no evidence for positive chemotaxis using this method. It should be considered, however, that H. seropedicae was the only species that demonstrated growth on agar, and therefore the only one able to enter the capillaries. We concluded that this methodology is not yet sufficiently optimised for our application and may be having a confounding effect upon chemotactic response. Further details of these potential factors can be found here:

Figure 5: a) Growth of H. seropedicae on 1 % LB agar inoculated with contents of a 1 µl capillary containing 100 µM naringenin after 60 minutes open-end submersion in bacterial solution. Plate was incubated for 24 hours at 30 °C. b) Growth of H. seropedicae on 1 % LB agar inoculated with contents of a 1 µl capillary containing motility buffer after 60 minutes open-end submersion in bacterial solution. Plates were incubated for 24 hours at 30 °C.

Microscopy Observations

An alternative method of observing chemotactic responses is through the use of microscopy. Brightfield microscopy allows direct observations of bacterial responses. This will allow comparisons of motility and morphology from our experimental data to that of the published literature that was used to underpin our first iteration of the community model [link to modelling]. Using microscopy enables the development of a cell density:optical density index (CD:OD index), a method of converting the two values. This index was also used in the community model to adapt the growth curve data collected during bacterial characterisation in standard laboratory conditions.

Figure 6: Example of haemocytometer square at 40x objective containing H. seropedicae utilised for cell counting, cells along the bottom and/or right lines were not counted to avoid double counting.

The CD:OD index was produced utilising data collected from a haemocytometer. A haemocytometer is a specialised microscopy slide of a known volume, it also contains a grid at the centre. By counting the number of cells in 16 squares at the top right (Figure 6) and performing a series of mathematical calculations ([x], we were able to determine cell density. By utilising a spectrophotometer, we were also able to take a reading of the absorbance (600 nm) and thus link the two together (Table 3).

Table 3: Cell density (cells.ml-1) of A. brasilense, A. caulinodans, H. seropedicae and E. coli at different optical densities and as a ratio of OD600=1
Species (Strain) Optical Density Cell Density Colony Count (Control)
A. caulinodans (ORS571) 0 0 0
A. brasilense (SP245) 0 0 0
H. seropedicae (Z67) 145.33 85.58 109.78
E. coli (DH5α) 0 0 0

Utilising a haemocytometer to count cells also allows observations of cell morphology and behaviour. This was used to an advantage as it allowed us to explore whether our bacteria’s morphology aligns with the literature that was followed to produce the community model. It also enabled observations of motility which was important after theories that the species were no longer motile which is why only H. seropedicae showed movement into the capillary during the attempt to quantify chemotaxis.

Fluorescein/OD600 and MEFL/particle analysis

The relatively poor growth and high fluorescence levels effectively cancelled each other out when readings were converted to Fluorescence per OD600 and MEFL per OD600 measurements, resulting in TD4 transformants producing the highest expression levels (figure 3.1). The high fluorescence and MEFL per OD600 reading for TD4 despite lowest growth suggests expression of TD4 is not fully representative of the relative promoter strength; expression levels are interdependent with growth rate, with higher growth rates expected to produce higher expression levels (Scott et al. 2010). Expected fluorescence levels based on relative promoter strength reported for the Anderson collection of promoters did not match entirely the results produced here. In particular, TD5 utilising the promoter J23104 was expected to be the second strongest but yielded only the fourth highest fluorescence reading of 22.14 and 21.42 for colonies 1 and 2 respectively. Similarly, the highest fluorescence reading was recorded by TD1 (expected strength: third), though this test device produced the widest range in fluorescence reading between the two colonies (r = 36.2), despite both colonies having the closest OD600 reading of any of the test devices (r = 0.007). In addition to the iGEM repository documentation for relative strengths of the Anderson promoter collection, previous literature has also demonstrated that the J23101 and J23104 promoters should have almost equal strength (He et al. 2017).

While TD5 appeared to underperform compared to promoter activity previously reported in the literature, subsequent sequencing of test devices revealed that colonies labelled as TD5 had in fact been transformed with the positive control device. This may have simply been the result of human error when pipetting or labelling over the process of the study. The consistence of TD5 underperformance across multiple replications of the study suggests that this occurred early on. The variation in expression visible is particularly alarming for the J23101 promoter, which has been proposed and utilised in the literature as a reference promoter to characterise relative strengths of other promoters as relative promoter units (Kelly et al. 2009).

Sources of variation in the InterLab study design

While the InterLab study is an effective way of gathering large sets of data surrounding the parts and protocols, it does not consider all sources of variation within datasets or the variability between data sets. As there are many factors which cause variability in microbial protein expression and productivity, some of these factors may be more favoured than others, leading to an overrepresentation of these in the metadata.

Competent cell protocols

Even prior to the main cell measurement protocol, irreproducibility had a significant effect in the preparation of competent cells and successful transformation. While the recommended CCMB80 and transformation protocols were followed exactly, successful expression of transformants were not guaranteed. It took 3 weeks of constant run-throughs within our lab to gain a transformation efficiency (TrE) of 5.05 x 106 before we could even start the Interlab. As such, the competent cell and transformation process was further investigated through a Bio-design Automation platform that used a design of experiments methodology to optimise transformation buffer (TB) composition. Utilising our recently acquired OT-2 liquid handling robot (Opentrons, USA), a robust automated competent cell preparation protocol was developed.

Recovery period

One factor that has been overlooked in the literature is the recovery period. Anecdotal evidence during all transformation protocols has indicated that the antibiotic resistance gene has a significant impact on how long the recovery time needs to be. For chloramphenicol, the widespread suggestion of a 1-hour recovery incubation for optimal TrE is quite simply incorrect, with a recovery incubation time of upwards of 2 hours being required for optimal TrE in our study. This is irrespective of volume, be it in 2 mL microcentrifuge tubes or 96 well plates. However, if using a 96 well plate format, this recovery period requires full optimisation due to its suboptimal OTR and KLa characteristics. With the additional growth inhibition of the majority of TB compositions, this recovery step needs to be fully optimised for optimal TrE. Super optimal broth with catabolite repression (SOC) is regularly used instead of SOB to enhance cells recovery, however as it includes glucose, this was not considered due to the potential inhibitory effects of increasing pH due to glucose metabolism (Islam et al. 2007; Losen et al. 2004; Marini et al. 2014).

Experimental observation found ampicillin (AMP) did not require longer incubation. Chloramphenicol’s requirement for increased recovery time can be explained through its mechanism of action, inhibiting protein synthesis via the inhibition of peptidyl transferase (Schifano et al. 2013; Wolfe and Hahn 1965). AMP however inhibits cell wall synthesis via inhibition of transpeptidase. Unlike AMP resistance, which involves the synthesis and excretion of either β-lactamase or penicillinase (Drawz and Bonomo 2010), chloramphenicol resistance is acquired by the synthesis of chloramphenicol acetyltransferase which is not readily excreted (Shaw 1983). This is beneficial for generating libraries as it decreases risk of satellite colonies, but the resistance mechanism may take longer to form and confer sufficient antibiotic resistance. As a result, for a further optimised protocol, using a plasmid that confers AMP resistance would be beneficial to minimise the protocol time requirement. This would also decrease the safety risk as while CAM is a known carcinogen, AMP is not.

Use of mut3GFP as a reporter

Even prior to the main cell measurement protocol, irreproducibility had a significant effect in the preparation of competent cells and successful transformation. While the recommended CCMB80 and transformation protocols were followed exactly, successful expression of transformants were not guaranteed. It took 3 weeks of constant run-throughs within our lab to gain a transformation efficiency (TrE) of 5.05 x 106 before we could even start the Interlab. As such, the competent cell and transformation process was further investigated through a Bio-design Automation platform that used a design of experiments methodology to optimise transformation buffer (TB) composition. Utilising our recently acquired OT-2 liquid handling robot (Opentrons, USA), a robust automated competent cell preparation protocol was developed.

Competent cell protocols

Even prior to the main cell measurement protocol, irreproducibility had a significant effect in the preparation of competent cells and successful transformation. While the recommended CCMB80 and transformation protocols were followed exactly, successful expression of transformants were not guaranteed. It took 3 weeks of constant run-throughs within our lab to gain a transformation efficiency (TrE) of 5.05 x 106 before we could even start the Interlab. As such, the competent cell and transformation process was further investigated through a Bio-design Automation platform that used a design of experiments methodology to optimise transformation buffer (TB) composition. Utilising our recently acquired OT-2 liquid handling robot (Opentrons, USA), a robust automated competent cell preparation protocol was developed.

Competent cell protocols

Even prior to the main cell measurement protocol, irreproducibility had a significant effect in the preparation of competent cells and successful transformation. While the recommended CCMB80 and transformation protocols were followed exactly, successful expression of transformants were not guaranteed. It took 3 weeks of constant run-throughs within our lab to gain a transformation efficiency (TrE) of 5.05 x 106 before we could even start the Interlab. As such, the competent cell and transformation process was further investigated through a Bio-design Automation platform that used a design of experiments methodology to optimise transformation buffer (TB) composition. Utilising our recently acquired OT-2 liquid handling robot (Opentrons, USA), a robust automated competent cell preparation protocol was developed.





InterLab

REFERENCES

1. Jousset, A., et al. (2009). "Predators promote defence of rhizosphere bacterial populations by selective feeding on non-toxic cheaters." The Isme Journal 3: 666

2. Jousset, A., et al. (2009). "Predators promote defence of rhizosphere bacterial populations by selective feeding on non-toxic cheaters." The Isme Journal 3: 666

3. Vanitha SC & Umesha S (2011) Pseudomonas fluorescens mediated systemic resistance in tomato is driven through an elevated synthesis of defense enzymes. Biologia Plantarum 55(2):317-322.

4. United Nations, Department of Economic and Social Affairs, Population Division (2017) World Population Prospects: The 2017 Revision, Key Findings and Advance Tables. https://population.un.org/wpp/Publications/Files/WPP2017_KeyFindings.pdf

5. Food and Agriculture Organization of the United Nations (2015) World Fertilizer Trends and Outlook to 2018. http://www.fao.org/3/a-i4324e.pdf

6. Usman MN, MG; Musa, I (2015) Effect of Three Levels of NPK Fertilizer on Growth Parameters and Yield of Maize-Soybean Intercrop. International Journal of Scientific and Research Publications 5(9).

7. Pfromm PH (2017) Towards sustainable agriculture: Fossil-free ammonia. Journal of Renewable and Sustainable Energy 9(3):034702.

8. Bitew YA, M (2017) Impact of Crop Production Inputs on Soil Health: A Review. Asian Journal of Plant Sciences 16(3):109-131.

9. Yang X-e, Wu X, Hao H-l, & He Z-l (2008) Mechanisms and assessment of water eutrophication. Journal of Zhejiang University. Science. B 9(3):197-209.

10. Carmichael WW (2001) Health Effects of Toxin-Producing Cyanobacteria: “The CyanoHABs”. Human and Ecological Risk Assessment: An International Journal 7(5):1393-1407.

11. New Partnership for Africa's Development (2013) Agriculture in Africa - Transformation and Outlook. http://www.un.org/en/africa/osaa/pdf/pubs/2013africanagricultures.pdf

12. Food and Agriculture Organization of the United Nations (2017) World Fertilizer Trends and Outlook to 2020. http://www.fao.org/3/a-i6895e.pdf

13. Bergey, D. H., et al. (1984). Bergey's manual of systematic bacteriology. Baltimore, MD, Williams & Wilkins.

14. Gómez-Lama Cabanás C, Schilirò E, Valverde-Corredor A, & Mercado-Blanco J (2014) The biocontrol endophytic bacterium Pseudomonas fluorescens PICF7 induces systemic defense responses in aerial tissues upon colonization of olive roots. Frontiers in Microbiology 5:427.

15. Gross, H. and J. Loper (2009). Genomics of secondary metabolite production by Pseudomonas spp.

16. Sharma SB, Sayyed RZ, Trivedi MH, & Gobi TA (2013) Phosphate solubilizing microbes: sustainable approach for managing phosphorus deficiency in agricultural soils. SpringerPlus 2:587.

17. Ruffner, B., et al. (2013). "Oral insecticidal activity of plant-associated pseudomonads." Environmental Microbiology 15(3): 751-763.

18. Jousset, A., et al. (2009). "Predators promote defence of rhizosphere bacterial populations by selective feeding on non-toxic cheaters." The Isme Journal 3: 666

19. Vanitha SC & Umesha S (2011) Pseudomonas fluorescens mediated systemic resistance in tomato is driven through an elevated synthesis of defense enzymes. Biologia Plantarum 55(2):317-322.

20. Maheshwari DK (2012) Bacteria in Agrobiology: Plant Probiotics (Springer Berlin Heidelberg).

21. Despommier D (2011) The vertical farm: Controlled environment agriculture carried out in tall buildings would create greater food safety and security for large urban populations. J fur Verbraucherschutz und Leb 6(2):233–236.

22.World Health Organization. (2018). Q&A: genetically modified food. [online] Available at: http://www.who.int/foodsafety/areas_work/food-technology/faq-genetically-modified-food/en/ [Accessed 13 Sep. 2018].