High-Throughput Metabolic Profiling for Model Refinements of Microalgae

The phenotype micro-array technology is an effective high-throughput method that functionally determines cellular metabolic activities in response to a wide array of entry metabolites. The utilization in this technology is measured in the form of cellular respiration determined by the amount of color development produced by the NADH reduction of a tetrazolium-based redox dye. In this work, we will introduce the use of phenotype micro array assays in the context of micro-algae using model species Chlamydomonas reinhardtii.

The goal of this study is to establish a reliable method for characterizing metabolic phenotyping of micro-algae that can be used to expand existing algal metabolic network models or guide the reconstruction of new models. Chlamydomonas reinhardtii strain CC-503 can be obtained from the Chlamydomonas resource center at the University of Minnesota USA. Grow the cells in fresh Tris-acetate phosphate fat media to mid-log phase.

Check the cells under the microscope to make sure that they are in good shape and without any contamination. Spin down the culture at 2000 G-force for 10 minutes. Discard the supernatant without disturbing the pellet.

Prepare fresh tap media containing 0.1%tetrazolium violet dye D"Add different antibiotics including Timentin, Ampicillin, and Kanamycin to the media to inhibit bacterial growth. Fat media and this step should be omitted from nutrients, depending on each plate type. Re-suspend the pellet and the fresh tap media to a final concentration of 1 million cells per milliliter.

Use chemical compound array assay plates, such as carbon sources, nitrogen sources, phosphorous, and sulfur sources plates, and the peptide nitrogen sources. Inoculate 100 microliter adequate of self-containing media into each well of the essay plates. Make sure to duplicate the assays.

It should be noted that at this stage, cells should be tested with gram-negative staining before and after the assay to monitor bacterial contamination. Insert the chemical compound area assay plates into the microplate reader system. Incubate all the plates at 30 degrees for up to seven days, and program the microplate reader system to read the dye color change every 15 minutes.

As most microplate readers do not provide a source of continuous light during incubation, the algae should be able to carry out heterotrophic respiration. Export the raw kinetic data from the microplate reader as CSV files, which will subsequently be used as input to the phenotype microarray software package and our software. To carry out the phenotype microarray data analysis, use the OmniLog phenotype micro Erik OPM software package that runs within the R software environment.

In R studio, the graphical user interface for R, install the OPM package and its dependencies using the commands detailed in the manuscript. Navigate to the directory that contains the CSV files of the kinetic data and import the data using the read OPM function. The kinetic data can be aggregated and discretized using pref parameter estimation.

Using the function XY plot allows the respiration or growth measurements to be mapped as a function of time for the assay 96-well plates. The data can be visualized as a heat map using the function level plot to allow for a quick comparative overview of the kinetic data. The next step is to identify the reactions and genes associated with the new metabolites.

In case of presence of an existing model for the algae, the data analysis from the phenotype microarray system can be used to refine this model. Here we present the pipeline we used for genome-scale metabolic network refinement for the Chlamydomonas model using phenotype microarray data. When a new compound test positive for utilization, the compound's relevant reaction profiles are defined using metabolic knowledge basis, providing the associated enzyme commission numbers.

The first step is to search Kegg and MetaCyc to identify enzyme commission numbers, ECs, for reactions using metabolized found from the chemical compound arrays. Next, we use the identified EC numbers as a search basis in multiple available algal annotation resources, such as joint genome Institute, JGI, Phytozome, and peer reviewed publications. The reactions and metabolites are added to the COBRA-based Chlamydomonas reinhardtii metabolic network model, iRC1080, to expand and refine the model.

If genetic evidence is not found in support of the EC number, a profile based search such as NCBI Position-Specific Iterative, NCBI PSI-BLAST, can be performed to identify candidate genes associated with the reaction. The results are then manually evaluated. Those passing this QC step with E values less than 0.05 for relevance to the search EC numbers, are then added to the metric model.

The last step of this protocol is refining the model, evaluating and comparing the models. Use the latest COBRA toolbox version 3.0 and the MATLAB platform to carry out the steps for model refinement. After installing the Cobra toolbox, you can download the iRC1080 model.

Then in MATLAB, the first thing to do is to navigate to the folder containing the reference model, iRC1080. Add the identified reactions with their associated genes to the metabolic model, such as iRC1080, using the Cobra toolbox functions. Add reaction and change gene association.

Navigate to the directory that contains the iRC1080 model and execute the commands to load the model. Rename it and add a new reaction and its associated gene. In some cases when the metabolite is not produced intracellularly, but is taken from the medium, transport reactions for the new metabolites need to be added to the model using the add/exchange reaction function to input or output the metabolite into the extracellular medium.

Test the behavior of the new result and model, for example, iBD1106, by carrying out flux balance analysis, FBA, using the function optimize CB model under light and dark conditions for the maximization of biomass as the object function. Amongst others, the FBA solution outputs three vectors the reaction fluxes, the metabolite shadow prices, and the reaction reduced costs. Here, an example is provided where the iRC1080 model is compared with its refined version, iBD1106, by obtaining the shadow prices that represent the sensitivity of the biomass objective function to changes and metabolites accounted for in each model.

Here we show respiration XY plots and level plots of the carbon sources and nitrogen sources assay plates for blank and CC-503 in teal color and purple, respectively. Within each well, respiration curves represent dye conversion by reduction as function of time. The highlighted peak in panel A represents the detection of acetate as the only carbon source from this plate, which is consistent with Chlamydomonas literature.

The comparison of the number of metabolites identified using three different methods, iRC1080, which is the well-curated Chlamydomonas metabolic model, the gas chromatography time of flight GC-TOF, and the phenotype microarray essay shows that only six metabolite overlap between the three sets, while 149 were common between iRC1080 and the phenotype microarray assay set. This shows that while each technology has a strength towards the metabolic profiling research, the phenotype microarray assay set can be a significant source of new metabolic information. The obtained information was used to expand and refine the metabolic network model iRC1080.

Here, we compare the content of iRC1080 model and the expanded model iBD1106, including the number of reactions, number of metabolites and number of genes. We show that our model refining added over 254 reactions to the new resulted network. These reactions are classified into amino acids, dipeptides, tripeptides and transport reactions.

The newly identified metabolites were used for metabolic network expansion and refinement of the existing Chlamydomonas reinhardtii metabolic model. The phenotype microarray assays can be used to characterize metabolic phenotypes of existing and newly isolated strains. In addition, the protocol that we used in the context of micro-algae can guide the refinement of metabolic models of other species.