Team:Lambert GA/Contributions

OVERVIEW

Improvement

Characterizing biobricks is essential to designing and reproducing experiments. Although each part in the Registry of Standard Biological Parts has detailed description, there is insufficient characterization of varying strengths of promoters and ribosomal binding sites (RBSs).

Last year, Lambert iGEM experienced overexpression, or leakiness, in our composite part BBa_K255000 due to a strong promoter, BBa_J23100. As a result, Lambert iGEM decided to change last year’s strong promoter to a medium strength promoter, BBa_J23106, for this year’s part. We ran numerous trials to test a matrix of weak, medium, and strong promoter and RBS strengths, characterizing the Anderson series promoters (BBa_J23119, BBa_J23106, BBa_J23113, BBa_J23115, and the RBSs BBa_B0034, BBa_B0032, BBa_B0031, and BBa_B0035) by analyzing enzymatic activity with Miller Units. We measured β-galactosidase enzymatic activity using these nine combinations and positive control.

Background

Promoters and ribosomal binding sites (RBSs) are critical to cellular transcription and translation. Our testing system, the Lac operon, is inducible, turning on when a certain molecule is present and off when absent. The LacZ gene in the Lac operon codes for β-galactosidase (β-gal), an enzyme that normally breaks down lactose into glucose and galactose. A similar molecule, ONPG (ortho-Nitrophenyl-β-galactoside), was used in place of lactose because when broken down, its products, galactose and o-nitrophenol (ONP), produce a yellow color. We used a plate reader to measure absorbance at 400 nm for the intensity of yellow color and at 600 nm for cell density. Lambert iGEM used the Miller unit formula to standardize amounts of β-gal activity for each promoter/RBS combination. A higher value of Miller units correlates to higher enzymatic activity and therefore, greater expression.

The expression of LacZ is affected by a combination of the promoter strength and RBS translation rate.

Combination Key
Strain # Promoter Registry # RBS Registry # Relative strength
Promoter/RBS
2-R BBa_J23115 BBa_B0035 Reference/Reference
2-1 BBa_J23113 BBa_B0031 Weak/Weak
2-2 BBa_J23113 BBa_B0032 Weak/Medium
2-3 BBa_J23113 BBa_B0034 Weak/Strong
2-4 BBa_J23106 BBa_B0031 Medium/Weak
2-5 BBa_J23106 BBa_B0032 Medium/Medium
2-6 BBa_J23106 BBa_B0034 Medium/Strong
2-7 BBa_J23119 BBa_B0031 Strong/Weak
2-8 BBa_J23119 BBa_B0032 Strong/Medium
2-9 BBa_J23119 BBa_B0034 Strong/Strong

WORKFLOW

The tuning workflow makes the invisible visible.

Protocol

Materials

    • 640 μL of 0.1 M IPTG (Isopropyl-β-D-thiogalactoside)
    • 640 μL of ampicillin
    • 320 ml of sterile LB media
    • 320 ml of sterile M9 media
    • 8 ml of 1X ONPG (ortho-Nitrophenyl-β-galactoside)
    • 40 mg into 40 mL of dH20 → 0.01% solution

Streak Plates

  1. Culture tubes of bacteria strains are ordered from Biobuilder Carolina Laboratories.
  2. Sterilize inoculating loop and top of opened bacteria tube by passing through Bunsen burner flame.
  3. Allow the inoculating loop to cool for a few seconds before inserting into a bacterial culture tube and scraping off a bit of bacteria.
  4. Streak onto the corresponding plate.
  5. Repeat steps a through d for each plate/bacterial strain.
  6. Incubate for 24 hrs at 37°C.

Liquid Cultures

  1. Add 320 μL of amp and 320 μL of IPTG to 320 mL of LB, then aliquot 4 mL of this mixture to each liquid culture tube (80 tubes total).
  2. Add 1 colony from the desired bacterial strain plate to 20 μl of MilliQ water, making sure to pipette up and down to the first stop until the colony is dissolved.
  3. Pipette 5 μL of bacterial mixture into each corresponding liquid culture tube.
  4. Incubate at 37°C for 24 hours, shaking.

Subcultures

  1. Add 320 μL of amp and 320 μL of IPTG to 320 mL of M9 and mix.
  2. Add 4 mL of mixture to each liquid culture tube.
  3. Take 200 μL of cells from the LB liquid cultures and add to the corresponding M9 liquid culture tube.
  4. Incubate at 37°C for 4 hours, shaking.

Reactions

  1. Add 200 μL of cells from the subcultures into small culture tubes.
  2. Create blanks using 200ul of M9 instead of cells.
  3. Add 100μL ONPG into each culture tube.
  4. Add 100μL of ONPG into the blanks as well.
  5. Let all tubes, including blanks, sit for 24 hours in the incubator at 37°C.

Testing

  1. Pipette 200μL of the cell and ONPG mixture for each bacterial strain into its own well in a 96 well plate.
  2. Pipette the blank into one well.
  3. Measure absorbance for each well at 420 nm and 600 nm.

Formula

    • 420 nm - measures enzyme activity (ONP’s yellow color)
    • 1000 - constant
    • t - time in minutes
    • v - volume of cells in mL
    • 600 nm - measures optical density of cells

Lambert iGEM calculated a standardized amount of β-galactosidase (β-gal) activity for each combination of promoter/ribosomal binding site (RBS) using a formula by Jeffrey Miller, the creator of the protocol used by Lambert iGEM, in “Experiments in Molecular Genetics” [1]. Absorbance readings for each combination were taken at 600 nm to account for optical density and at 420 nm to measure the strength of yellow color (o-nitrophenol) produced by the enzyme. The variable, t, is the time in minutes that the enzyme β-gal was given to break down o-nitrophenyl-β-D-galactoside (ONPG), and v is the volume in milliliters of cells taken from the liquid culture.

CHARACTERIZED PARTS

Work Stream Name Type Description Designer Length
✓ Improved BBa_K2550000 Composite BBa_J23100 Toehold Ribosome Switch with LacZ expression Janet Standeven, Abby Bell, Megan Hong, Yan Zhang, Courtney Taing, Ashtyn Cauffman 3218
✓ Improved BBa_K2550001 Composite T7 promoter with a trigger sequence to match the toehold sequence of part BBa_K2550000 Yan Zhang, Janet Standeven, Abby Bell, Megan Hong 152
✓ Modeling BBa_J10050 Composite Weak promoter + Weak RBS + LacZa.GFP Fusion Natalie Kuldell 1017
✓ Modeling BBa_J10051 Composite Weak promoter + Strong RBS + LacZa.GFP Fusion Natalie Kuldell 1015
✓ Modeling BBa_J10052 Composite Weak promoter + Medium RBS + LacZa.GFP Fusion Natalie Kuldell 1016
✓ Modeling BBa_J10053 Composite Medium promoter + Weak RBS + LacZa.GFP Fusion Natalie Kuldell 1017
✓ Modeling BBa_J10054 Composite Medium promoter + Medium RBS + LacZa.GFP Fusion Natalie Kuldell 1016
✓ Modeling BBa_J10055 Composite Medium promoter + Strong RBS + LacZa.GFP Fusion Natalie Kuldell 1015
✓ Modeling BBa_J10056 Composite Strong promoter + Weak RBS + LacZa.GFP Fusion Natalie Kuldell 1017
✓ Modeling BBa_J10057 Composite Strong promoter + Medium RBS + LacZa.GFP Fusion Natalie Kuldell 1016
✓ Modeling BBa_J10058 Composite Strong promoter + Strong RBS + LacZa.GFP Fusion Natalie Kuldell 1015
✓ Modeling BBa_J23115 Promoter Constitutive Promoter Family Member John Anderson 35
✓ Modeling BBa_B0035 Ribosomal Binding Site RBS (B0030 derivative) Jason Kelly 14

RESULTS

Liquid Cultures

Combinations 1-9 and R show a prominent yellow color from the ONPG broken down by β-galactosidase to produce ONP. As the promoter strength increases, the yellow color becomes more visible. Combinations 1 and 2 have a clearer color, indicating a weak promoter.

Combinations 1-6 show differences in color, indicating different levels of expression.

The expression of combinations 7-9 are compared to expression of the reference combination.

Miller Units for each Promoter/ RBS Combination

Strain Number

Promoter Part

RBS Part

Promoter/ RBS Description

Miller Units (8 replicates)

1

BBa_J23113

BBa_B0031

Weak/ Weak

5.201

3.358

7.308

6.319

7.467

8.548

5.97

6.054

2

BBa_J23113

BBa_B0032

Weak/ Medium

4.945

5.182

5.149

5.37

5.953

7.462

5.002

2.759

3

BBa_J23113

BBa_B0034

Weak/ Strong

5.268

5.939

5.96

4.329

10.852

10.259

7.233

3.264

4

BBa_J23106

BBa_B0031

Medium/ Weak

5.367

6.092

6.728

5.9

6.437

7.607

7.19

3.247

5

BBa_J23106

BBa_B0032

Medium/ Medium

5.062

5.492

5.798

5.551

6.114

6.646

5.953

8.249

6

BBa_J23106

BBa_B0034

Medium/ Strong

5.332

6.473

6.064

5.695

5.959

7.727

6.409

6.974

7

BBa_J23119

BBa_B0031

Strong/ Weak

6.295

6.46

6.032

6.108

10.367

8.507

6.217

9.444

8

BBa_J23119

BBa_B0032

Strong/ Medium

5.538

6.343

6.536

6.101

9.34

8.407

7.525

8.454

9

BBa_J23119

BBa_B0034

Strong/ Strong

5.97

7.835

13.293

7.711

7.735

13.38

8.64

8.353

Reference

BBa_J23115

BBa_B0035

Reference/ Reference

5.636

 7.034

 6.191

 6.612

 7.485

 7.290

 6.974

 9.421


Data Distribution

The distribution of results from 10 trials are shown.


CONCLUSION

Lambert iGEM’s Tuning experiment supports the change in promoter strength for this year’s toehold construct. Last year, Lambert iGEM utilized the BBa_J23100, a strong promoter that causes overexpression. While constructing this year’s Toehold Biosensor for C. elegans, we decided to use a different promoter to prevent this leakiness. Testing supported our claim that the use of a medium promoter and strong RBS produced the most consistent expression. As a result, this year’s toehold design used the moderate promoter, BBa_J23106, rather than the strong promoter, BBa_J23100.

Using the data from Lambert iGEM’s Tuning experiment, a model was created to predict protein expression based on promoter and RBS strength by inputting promoter and RBS sequences. To represent how the strength of the promoter impacts expression, we built a multivariate linear regression of the data.

As seen in Trial Conclusions below, Lambert iGEM’s Tuning experiment encountered many issues with variability, similar to other iGEM teams conducting enzymatic experiments. To combat this, our lab decided to remove the lysing step from the original Biobuilder protocol, as it causes the number of lysed cells and the exposed enzymes to vary greatly. Additionally, our lab decided to allow the experiment to run longer to allow ONPG to cross the cell membrane and react within the cell. These findings, along with our adaptations, can aid other labs and iGEM Teams to combat variability in enzymatic expression experiments.

Individual Trial Conclusions

Trial 1

We followed protocol from Carolina Science’s Biobuilder iTune Device lab for our first trial. However, after conducting further research, we recognized that the Miller unit formula used to calculate β-gal activity did not account for cell debris from lysing the cells. Cell debris and light scatter caused inaccurate calculations of enzymatic activity in this trial.

To resolve this, our research determined that cell debris can be measured at 550 nm and that there is a different Miller unit formula for 420, 550, and 600 absorbance measurements. We also decided to cool our liquid culture tubes on ice to inhibit further growth of the cells and precisely time the reactions. Trial 2 consists of improvements to the initial protocol, which is further described below.

Trial 2

When conducting this trial, we experienced contamination, so we implemented sterilized cuvette caps in subsequent trials.

Normally, the distinct yellow color of the reaction comes from ONP, a component of ONPG after it is broken down by β-gal. However, in this trial, there was minimal to no yellow color production in comparison to Trial 1. To solve this issue, we looked into the efficacy of reactants used. We recognized that Isopropyl-β-D-thiogalactoside (IPTG) was being continuously frozen and thawed, potentially causing degradation. IPTG is the inducer of the Lac operon, so degradation would lead to minimal transcription of LacZ and low amounts β-gal enzymes in the solution. Thus, there would be little to no yellow color during the reaction. Our lab made a new stock of IPTG for subsequent trials.

Trial 3

After collecting results from Trial 3, the software team began to design our model and suggested to use SEM bars to account for any significant differences between the measurements, so we increased the sample size by conducting another trial with double the amount of samples from the previous trial. SEM bars were calculated using the s/√n formula, where s is the standard deviation and n is the sample size.

Trial 4

Our spectrophotometer yielded inaccurate results - reactions that were clearly yellow were measured as if they had no yellow color. Following Trial 4, we ensured that our spectrophotometer was measuring accurately by recalibrating it to our blanks multiple times and allowing the machine time to warm up before use.

Trial 5

The measurement of the results were negative, which could be due to error as tested cuvettes should not have less expression than the blank. However, results from samples 4-6 were positive and were used for model data.

To account for error, each experimenter tested one of each of the triplicates for strains 1 through 9 and R, instead of triplicates for 3 strains. We made new batches of IPTG and ONPG to ensure that degradation of the two reagents did not cause the errors seen in Trial 5.

Trial 6

To reduce variability between the reactions each experimenter would test 1-9 and one R, allowing us to take the averages of each triplicate of 3 combinations. We followed the same protocol as the previous trial and measured the reactions. However, the results showed that only one experimenter had expression in the nine combinations, while the other two sets did not. This was unlikely to happen since each experimenter was testing the combinations using the same reagents. There was only “yellowness” in 9 out of the 27 cuvettes that were supposed to have expression.

We troubleshooted and came to the conclusion that vortexing to lyse cells was the only variable not uniform, causing variability. This provides reasoning for why the other experimenters did not have expression in their reactions. Our hardware team advised us with vortexing and suggested that we wait until the RPM of the vortex reached the optimal speed before timing 10 seconds, allowing all the cell membranes to be broken and exposing the enzymes to ONPG.

We also decided to increase the volume of cells from 100 ul to 500 ul in the next trial to produce more expression. This change was accounted for in the formula, as well as in our protocol. The cuvette for 600nm would receive 500ul of cells and 2200ul of MQ water, and the 420/550nm cuvette would receive the same amount of reagents but just 500ul of cells.

Trial 7

We modified our protocol due to the errors in trial 6 and decided to use the same vortex at the same speed and waited for the RPM of the vortex to reach full speed before counting 10 seconds. We also used 500ul of cells this trial to optimize expression. Other than these two changes, the protocol remained the same. Each experimenter had expression in our results; however, the Miller units were significantly smaller than trial 6, which was expected because of the larger volume of cells in the denominator of our Miller unit calculations. We still experienced significant variability of enzymatic activity and expression between the strains of promoter/RBS combinations.

We discussed our results and concluded that we did not have enough ONPG for the increased number of cells. For the next trial, we decided to use 100 ul of cells so that the β-gal enzymes would not run out of ONPG.

Our results all had expression; however, the data was scattered, prompting us to collaborate with Georgia Tech and use their plate reader for more accurate and reliable data. Using a plate reader would allow us to read up to 96 reactions at one time and hopefully reduce variability.

Trial 8

Lysing cells was the last thing in our protocol that could possibly cause the significant variability between replicates we were experiencing. We consulted with Dr. Mark Styczynski, who informed us that ONPG can diffuse across the cell membrane. We created a new protocol, detailed above, where we allowed ONPG to diffuse into the cell instead of exposing the enzymes to ONPG through lysing. We could not find literature on how long ONPG took to diffuse across the cell membrane, so we let one set of reactions continue to diffuse and timed it. After 24 hours, the reaction was yellow, showing that ONPG did diffuse across the membrane and the enzyme was expressed.

Trial 9

We tried our new protocol in triplicate, and experienced reduced variability. However, we wanted to ensure that our spectrophotometer was not affecting the variability of our results, so we contacted Georgia Tech to request access to their plate reader.

Trial 10

Before the start of this trial, we contacted Dr. Styczynski once more. He informed us that sodium bicarbonate and sodium carbonate were superfluous to our reactions, so we omitted them from our experiment. In this trial, we ran 8 replicates of each bacterial strain to increase sample size. We followed the protocol detailed above and transported the final tubes to Georgia Tech to measure with their plate reader.

REFERENCES

[1] Beta-Galactosidase Assay (A better Miller). (2012, August 27). OpenWetWare, . Retrieved September 25, 2019 from https://openwetware.org/mediawiki/ index.php?title=Beta-Galactosidase_Assay_(A_better_Miller)&oldid=620416

[2] Biobuilder. (2017). iTune Device. Retrieved from https://biobuilder.org/wp-content/ uploads/2017/10/Ch7_iTuneDevice_BioBuilderTextbook.pdf

[3] Biobuilder. (2018). iTune Device. Retrieved from https://biobuilder.org/wp-content/ uploads/2018/08/carolina_itune_teacher.pdf

[4] Borralho, T., Chang, Y. S. H., Jain, P., Lalani, M., & Parghi, K. (1970, January 1). [PDF] Lactose Induction of the lac operon in Escherichia coli B 23 and its effect on the o-nitrophenyl ß-galactoside Assay - Semantic Scholar. Retrieved from https://www.semanticscholar.org/paper/Lactose-Induction-of-the-lac-operon-in-Escherichia-BORRALHO-Chang/ca20cdac1d995b724637589b8b94b513aaf21b0d.

[5] Brenner, S. X., & Miller, J. H. (2001). Encyclopedia of genetics. San Diego: Academic Press.

[6] Hidalgo, C., Reyes, J., & Goldschmidt, R. (1977, February). Induction and general properties of beta-galactosidase and beta-galactoside permease in Pseudomonas BAL-31. Retrieved from https://www.ncbi.nlm.nih.gov/pmc/articles/PMC235017/.

[7] Juers, D. H., Matthews, B. W., & Huber, R. E. (2012, December). LacZ β-galactosidase: structure and function of an enzyme of historical and molecular biological importance. Retrieved from https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3575911/.

[8] Smale, S. T. (2010, May). Beta-galactosidase assay. Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/20439410.

[9] Stephenson, F. H. (2016). Calculations in molecular biology and biotechnology: a guide to mathematics in the laboratory (3rd ed.). San Diego, California.: Academic Press.