Part:BBa_K4665005

SazCA-INPN Membrane Display Module

Usage and Biology

Biomineralization is the process by which living organisms synthesise minerals (Dhami et al., 2013). Microbial calcium carbonate production can proceed through two main metabolic pathways, using urease or carbonic anhydrase (CA) as the catalysts of the reaction (Chaparro-Acuña et al., 2019). However, synthesis through urea hydrolysis produces toxic byproducts which is not observed in the CA catalyzed pathway.

SazCA, derived from the thermophilic bacterium Sulfurihydrogenibium azorense, is the fastest known carbonic anhydrase to date, with an approximate kcat/KM value of 3.5 × 108 M⁻¹ s⁻¹ (De Simone et al, 2015; De Luca et al., 2013). SazCA facilitates the hydration of carbon dioxide to bicarbonate and protons, creating alkaline conditions that aid the formation of calcium carbonate crystals on the extracellular matrix (EPS) of bacterial cells (Fig. 1) (Anbu, et al., 2016).

To enhance enzymatic efficiency, this composite part expresses the SazCA enzyme as a fusion protein on the cell surface of E. coli. This approach bypasses cellular limitations and directly exposes the enzyme to extracellular CO2, increasing calcium carbonate production on limestone surfaces. This component is based on the findings of Zhu et al. (2022), wherein a membrane fusion protein was designed to showcase SazCA on the surface of E. coli cells. This is achieved by linking the E. coli codon-optimized SazCA enzyme (BBa_K4665120) to the integral membrane protein INPN (BBa_K4665001) using a flexible GGGGS linker (BBa_K4665175).

This composite part consists of three basic parts:

1) Ice nucleation protein N-terminal (INPN): This is the N-terminal of ice nucleation protein which will be embedded into the E. coli cell membrane. The sequence coding for the INPN is preceded by a pelB leader sequence as its expression promotes the secretion of the protein via the Sec pathway whilst avoiding hydrolysis by cytoplasmic proteases that might lower the quantity of proteins on the cell’s surface (Mergulhao et al., 2005). By attaching the pelB signal peptide in front of the INP protein, the fusion protein will be directed towards the bacterial periplasm where it will be anchored in the cell membrane (Singh et al., 2013). The INPN sequence is followed by two front-end sub-repeat sequences important for the stability of the fusion protein (Zhu et al., 2022).

2) GGGGS linker: The GGGGS flexible linker is composed of a sequence of 4 glycine repeats followed by a serine amino acid. This flexible linker is used to connect the N-terminal of the INP to the carbonic anhydrase which creates an elongated fusion mode that allows for optimal carbonic anhydrase stability (Hartmann et al., 2022; Zhu et al., 2022).

3) SazCA: This sequence codes for the carbonic anhydrase derived from Sulfurihydrogenibium azorense (SazCA). This sequence has been codon optimised for E. coli. The SazCA coding sequence is followed by a His-tag which facilitates the purification and detection of the fusion protein.

Figure taken from Zhu et al., (2021).

Zhu et al. (2022) were able to show that the surface display of the INP-SazCA fusion protein significantly elevates the enzyme’s stability, optimising whole-cell activity at 25°C and pH 9, retaining minimal metal inhibition.

Characterisation

Expression:

After successful transformation into BL21 DE3 E. coli, the expression of our recombinant protein was tested. The construct is preceded by the T7 promoter, therefore expression of recombinant protein can be induced through addition of Isopropyl β-D-1-thiogalactopyranoside (IPTG). Liquid overnight cultures of the transformed E. coli were induced with IPTG and samples were taken at different incubation times. Membrane protein extraction was performed by incubation with NPI-10 buffer followed by sonication. SDS PAGE and Western Blot was performed with antibodies that bind to the 6X-His tag attached at the end of the fusion protein. Previous iterations of the Western Blot revealed the problem of a leaky T7 promoter in our recombinant plasmid, leading to protein expression in absence of IPTG. Previous research has shown that addition of glucose to BL21 DE3 bacterial culture can reduce the background expression of the uninduced gene (Pan & Malcolm, 2000). In our Western Blot we tested various controls, the transformed BL21 DE3 cultures induced with IPTG at different time intervals and the transformed BL21 DE3 cultures that were grown overnight in the presence of glucose. Prior to IPTG induction, the glucose-containing medium was removed from the 0.5% and 1% glucose samples and they were resuspended in normal LB medium to prevent reduced expression of SazCA during IPTG induction.

Gel 1:
Protein ladder
1. Not induced GG3B
2. Negative control BL21 DE3
3. Not induced 0.5% glucose
4. Not induced 1% glucose
5. IPTG Induced GG3B 1hr
6. Induced GG3B 0.5% 1hr
7. Induced GG3B 1% hr

Gel 2:
Protein ladder -
8. IPTG Induced GG3B 2hr
9. Induced GG3B 0.5% 2hr
10. Induced GG3B 1% 2hr

The absence of a band in the BL21 DE3 control confirms that the observed bands in uninduced samples are due to a leaky T7 promoter and not due to the presence of native histidine-rich proteins expressed in BL21 DE3 E. coli . The results show that the addition of 0.5% or 1% glucose in the growth medium leads to reduced expression of the recombinant protein in the uninduced samples. However, despite the removal of the glucose-containing medium, the samples that contained the glucose showed reduced expression during IPTG induction. Our recombinant protein was most abundant in the transformed BL21 DE3 culture that had been induced with IPTG for 2 hours.

Protein Quantification

Cells from liquid culture (LB + Kanamycin + 0.5 mM ZnSO4 + 0.5 mM IPTG) were sonicated to lyse the cell membrane. Prior to the procedure, cells were centrifuged and shocked at -80℃ for 1h. Pellets were resuspended in 3x volume of NPI-10 buffer (50mM NaH₂PO₄, 300mM NaCl, 10mM Imidazole) and left on ice for 30 minutes. Samples were sonicated on ice using a 40 kHz pulse applied for 30s, followed by a 2 min pause. This sonication step was repeated 10 times. Given the fact that the construct is a protein embedded protein, a cell fractionation step should be followed by sonication. Unfortunately, this step was not performed due to lack of proper available equipment (ultracentrifuge). Protein extraction was performed using affinity chromatography with HisPur™ Ni-NTA Resin (ThermoFisher), optimising yield through the addition of Triton-100X to the protein buffers to reduce nonspecific interactions with untagged proteins. Extracted proteins were preliminarily quantified using a NanoDrop microvolume UV-Vis spectrophotometer set at 280 nm. A Bradford assay was carried out in order to determine protein concentration for the SazCA-INPN mineralisation module. A low working range (1-25µg/mL) microplate protocol was followed according to NanoDrop values. Working volumes of standard/sample was 150µL, with an equal volume of Bradford Protein Assay Reagent (ThermoFisher). Absorbance was measured at 595nm using a microplate reader. Average protein concentration was determined to be 68.34 µg/mL (0.5 mM ZnSO₄ + 0.5 mM IPTG).

In Vitro Mineralization:

To test the ability of engineered BL21(DE3) E. coli strain to precipitate CaCO₃, we performed an in vitro mineralisation assay, adapting Zhu, et al. 's technique. Bacteria were cultured overnight in 30mL of LB +Kanamycin medium and 0.5 mM ZnSO₄ at 25℃. IPTG induction was performed 3 hours prior to experimentation. The assay was run on 8 mL Tris-HCl buffer 8.3 and 50mL of saturated CO₂ aqueous solution at 0℃. 3 mL of cell pellet were introduced into the solution, and the reaction was allowed to proceed on ice for an hour. At this point, the bacteria should have been able to produce bicarbonate ions. Cells were removed from the solution by centrifugation (15 min x 5000g). 25mL of a 0.3M solution of CaCl₂ was added to the remaining supernatant as a calcium source. The reaction was left to run at 25℃ for 12h. Samples were filtered using vacuum filtration and dried at 50℃ to evaporate the solvent. Solid mass was weighed and recorded as “Wet Weight”. Upon preliminary analysis of FT-IR data, it was concluded that the mineral sample contained a large amount of water, elucidated by the stretching O-H peak at 3400 cm.1. Hence, the sample was dried further in liquid nitrogen for 48 hours, final weight was recorded at 2.3 g (yield=306.17%). Precipitated dry crystals were analysed using ATR-IR.

Enzymatic activity of SazCA:

To measure the activity of the SazCA construct, a colorimetric Wilbur Anderson assay was adapted from Kim & Jo, (2022). The assay measures the ability of carbonic anhydrase to hydrate CO₂. Protons released during the hydration reaction cause a decrease in the pH of the solution. Such displacement of H⁺ can be recorded as a function of time taken for pH to shift from 8.5 to 6.5.

The colorimetric approach taken for the assay indirectly measured the change of pH by recording the colour change of phenol red upon the addition of SazCA. A reaction buffer of 20mM Tris pH. 8.3 (pKa=8.1) and 100µM phenol red (pKa=7.9) was used. Phenol red was chosen as the pH indicator as it shifts colours from yellow to pink over a pH range of 6.3 to 8.3. Total reaction volume was 1mL. 10µL of SazCA-BL21(DE3) culture were added into a cuvette for each reaction.  Varying amounts of saturated CO2 aqueous solution (0.279M) were added, volume filled to 1mL with corresponding amounts of reaction buffer. Data collection was performed by UV-Vis spectrophotometry, measuring absorbance change at 570 nm using the kinetics function of the spectrophotometer, recording every 0.1s for 30 minutes.  All reactions were performed ice-cold. The blank reaction consisted of 600µL reaction buffer and 400µL CO₂ solution. SazCA samples compared to WT BL21. Abs values were obtained as a colorimetric reference for the reaction buffer adjusted for pH at 8.3, 7.5, and 6.5.

Wilbur-Anderson Units (WAU) were calculated according to the following formula:

WAU=(t0-t)/t, where t0 is the time (s) taken for the control to undergo a colour shift, and t is the time taken for SazCA samples to undergo a colour shift.

FTIR

Precipitated dry crystals were analysed using ATR-IR. IR results were analysed along a biological control, consisting of a reaction using unmodified BL21 bacteria, as well as a blank chemical reaction between the saturated aqueous CO₂ solution and CaCl₂, without any biological component. Samples were compared to three standards: commercial CaCO₃ in the calcite phase, a sample of limestone sourced from the Maastricht quarry (in vaterite phase), and a reference spectrum for bacterially precipitated CaCO₃ retrieved from the National Institute of Standards and Technology,(n.d.). IR analysis indicates that the engineered bacteria successfully precipitated calcium carbonate, as opposed to samples from the bacterial control and blank reaction. The IR spectrum for the BL21 sample and the blank reaction show similar peak and shift patterns at around 3500-3400 cm^-1 indicative of H bonded oxygen. Peaks at 3000-2800 cm^-1, 2280 cm^-1 and 1600 cm^-1 indicate water presence on both samples. Peaks at 1400-1000 cm^-1 are indicative of ethanol contamination. Peaks between 900-700 cm^-1 do not correspond to carbonate vibrations, corroborated by the lack of a strong peak at 1400 cm^-1.

The IR spectrum from the SazCA bacteria (SIMD CaCO₃) shows bands corresponding to the vibrations of carbonate (CO₃^2–):  The sharp peak around 1400-1500 cm^-1 can be attributed to the symmetric stretching vibration of the carbonate. Vibrational bands at 872.251 and 712.251 cm^-1 align with the carbonate out-of-plane bending (v2 mode) and in-plane bending (ν4 mode) vibrations of calcite (Zhou et al., 2004; Levi et al., 1998). Furthermore, the absence of peaks at 744 and 1086 cm^-1 indicates that the compound is not vaterite (Ivanova et al., 2023); and the absence of peaks at 1080, 854, and 700 cm^-1 characteristic of aragonite ( Levi et al., 1998), indicate that the CaCO₃ mineralised by the bacteria is closer to the crystalline phase of calcite.

However, the IR spectrum for the bacterial sample shows peaks corresponding to biocontamination. Split peaks 3400 and 1600 cm^-1 and noise peaks at 2100m^-1 suggest the presence of organic materials in the sample, although they are aligned to the peaks from the precipitated CaCO₃ sample from NIST. Hence, there is reason to believe that these peaks are correlated with the method of precipitation, and are therefore irrelevant in this setting. X-Ray crystallography should still be performed to address the slightly shifted values for calcite, and confirm the phase state of the compound.

ThermoGravimetric Analysis

Further characterisation of the SazCA-BL21 precipitated CaCO₃ was performed through ThermoGravimetric Analysis (TGA). Degradation temperatures for calcium carbonate range between 800-850℃. Typical weight loss initiates within the temperature range of 650-750°C. This range is associated with the thermal process of CO₂ removal from CaCO3 (Silva et al., 2017). Experimentally, the sample shows a weight decrease (Δ3.239%)  between 66.7-119.41°C corresponding to the dehydration of physically absorbed water by the sample. CO₂ removal can be observed between 654.48 and 748.43°C. The adjacent segment elucidates the degradation of CaCO₃, observed by the increased steepness of the TGA and the high value of the weight derivative at 877.51°C, which indicates the loss of a large proportion of the sample. The slightly shifted value of degradation for the sample may be reasonably attributed to the composition being potentially calcite. This alteration in degradation temperature aligns with the principle that more stable structures exhibit greater resistance to heat.

To evaluate the functionality of the fusion protein while entrapped in hydrobeads, TGA was used to screen for the presence of calcium carbonate. Alginate beads were dried at 80°C prior to TGA analysis to remove moisture. The temperature was increased from 26°C to 900°C at a speed of 10°C/min, under nitrogen flow (Wang et al., 2015). Calcium carbonate has a characteristic sharp weight loss percentage at between 650-800°C (Oniyama,& Wahlbeck, 1995), while the degradation of alginate hydrogels can be observed between 200-400°C (Wang et al., 2015). The sample showed continuous degradation until 187.23°C, indicating the dehydration of the sample. A steep decline in weight percentage between 187.23°C - 287.39 °C (Δ32.994%), indicates the degradation of the alginate bead, corresponding to 47.1% initial weight. The derivative shows another relevant decrease between 533.85°C-676.59°C (Δ20.7951%), which could suggest the degradation of a non-stable phase of CaCO₃ (Liu et al., 2021). However, the lack of a sharp drop at ∼800 °C indicates the absence of calcite. Differences exist between the bacteria entrapped beads and the control. Control samples show no sharp decrease after 247.18°C, and the derivative indicates a continuous weight loss. Further characterisation with X-Ray Crystallography can elucidate the phase of CaCO₃ formed in the beads, and spectroscopic analysis of the reaction medium is required to test if CaCO₃ permeates through the beads.

Sequence and Features

Assembly Compatibility:

• 10
  INCOMPATIBLE WITH RFC[10]

  Illegal PstI site found at 470
  Illegal PstI site found at 592
  
• 12
  INCOMPATIBLE WITH RFC[12]

  Illegal PstI site found at 470
  Illegal PstI site found at 592
  
• 21
  COMPATIBLE WITH RFC[21]
• 23
  INCOMPATIBLE WITH RFC[23]

  Illegal PstI site found at 470
  Illegal PstI site found at 592
  
• 25
  INCOMPATIBLE WITH RFC[25]

  Illegal PstI site found at 470
  Illegal PstI site found at 592
  Illegal NgoMIV site found at 54
  Illegal AgeI site found at 555
  
• 1000
  COMPATIBLE WITH RFC[1000]

References

Anbu, P. et al. (March 1, 2016). Formations of calcium carbonate minerals by bacteria and its multiple applications. Springerplus 5(250). https://doi.org/10.1186/s40064-016-1869-2

Chaparro-Acuña, S.P., et al. (June, 2018). Soil bacteria that precipitate calcium carbonate: mechanism and applications of the process. Acta Agronómica 67(2). https://doi.org/10.15446/acag.v67n2.66109

De Luca, V. et al. (March 15, 2013). An α-carbonic anhydrase from the thermophilic bacterium Sulphurihydrogenibium azorense is the fastest enzyme known for the CO2 hydration reaction. Bioorganic & Medicinal Chemistry Letters, 21(6): 1465.1469. https://doi.org/10.1016/j.bmc.2012.09.047

De Simone, G., et al. (May 1, 2015). Crystal structure of the most catalytically effective carbonic anhydrase enzyme known, SazCA from the thermophilic bacterium Sulfurihydrogenibium azorense. Bioorganic & Medicinal Chemistry Letters, 1;25(9): 2002-2006. https://doi.org/10.1016/j.bmcl.2015.02.068

Dhami, N.K., et al. ( May 2013). Biomineralization of calcium carbonate polymorphs by the bacterial strains isolated from calcareous sites. Journal of Microbiology and Biotechnology, 23(5): 707-714. https://doi.org/10.4014/jmb.1212.11087

Hartmann, S., et.al. (January 22, 2022). Structure and protein-protein interactions of Ice Nucleation Proteins drive their activity. BioRxiv. https://doi.org/10.1101/2022.01.21.477219

Kim, J. H., & Jo, B. H. (2022). A Colorimetric CO2 Hydration Assay for Facile, Accurate, and Precise Determination of Carbonic Anhydrase Activity. Catalysts, 12(11), 1391. MDPI AG. http://dx.doi.org/10.3390/catal12111391

Pan, S. H., & Malcolm, B. A. (2000). Reduced background expression and improved plasmid stability with pET vectors in BL21 (DE3). BioTechniques, 29(6), 1234–1238. https://doi.org/10.2144/00296st03

Singh, P., et al.. (2013). Effect of signal peptide on stability and folding of Escherichia coli thioredoxin. PloS one, 8(5), e63442. https://doi.org/10.1371/journal.pone.0063442

Zhu, Y., et.al (December 6, 2021). Surface display of carbonic anhydrase on Escherichia coli for CO2 capture and mineralisation. Synthetic and Systems biotechnology, 7(1): 460-473. https://doi.org/10.1016%2Fj.synbio.2021.11.008