Increased biomass yield of Lactococcus lactis by reduced overconsumption of amino acids and increased catalytic activities of enzymes

Abstract

Steady state cultivation and multidimensional data analysis (metabolic fluxes, absolute proteome, and transcriptome) are used to identify parameters that control the increase in biomass yield of Lactococcus lactis from 0.10 to 0.12 C-mol C-mol(-1) with an increase in specific growth rate by 5 times from 0.1 to 0.5 h(-1). Reorganization of amino acid consumption was expressed by the inactivation of the arginine deiminase pathway at a specific growth rate of 0.35 h(-1) followed by reduced over-consumption of pyruvate directed amino acids (asparagine, serine, threonine, alanine and cysteine) until almost all consumed amino acids were used only for protein synthesis at maximal specific growth rate. This balanced growth was characterized by a high glycolytic flux carrying up to 87% of the carbon flow and only amino acids that relate to nucleotide synthesis (glutamine, serine and asparagine) were consumed in higher amounts than required for cellular protein synthesis. Changes in the proteome were minor (mainly increase in the translation apparatus). Instead, the apparent catalytic activities of enzymes and ribosomes increased by 3.5 times (0.1 vs 0.5 h(-1)). The apparent catalytic activities of glycolytic enzymes and ribosomal proteins were seen to follow this regulation pattern while those of enzymes involved in nucleotide metabolism increased more than the specific growth rate (over 5.5 times). Nucleotide synthesis formed the most abundant biomonomer synthetic pathway in the cells with an expenditure of 6% from the total ATP required for biosynthesis. Due to the increase in apparent catalytic activity, ribosome translation was more efficient at higher growth rates as evidenced by a decrease of protein to mRNA ratios. All these effects resulted in a 30% decrease of calculated ATP spilling (0.1 vs 0.5 h(-1)). Our results show that bioprocesses can be made more efficient (using a balanced metabolism) by varying the growth conditions.

Figure 1. Distribution of proteins.

All proteins quantified at all specific growth rates has been divided into different groups by function. Proteins in the network covers enzymes that are involved in the reactions used for metabolic flux analysis and listed in the File S1, Table S1. Lists of all proteins can be seen in the File S2, Table S8.

Figure 2. Distribution of protein concentrations.

Concentrations (copies fl⁻¹) of proteins at specific growth rate 0.2 h⁻¹ are shown. Lower picture shows top 40 proteins fl-1 at the same specific growth rate.

Figure 3. Changes of protein abundances and cost of ATP.

Abundances are present in copies fl⁻¹ and cost of ATP in ATP fl⁻¹. Ten the most abundant pathways/cellular processes with increase of specific growth rate are illustrated. Distribution has been made according to the classification by Bolotin et al . All data and distribution according to BioCyc database can be seen in the File S2, Tables S9 and S10.

Figure 4. Simplified scheme of carbon flux rates.

Fluxes are shown in C-mmol (gdw*h)⁻¹ from A-stat experiments of Lactococcus lactis. Blue line represents average values of three independent experiments and red lines represent upper and lower values of standard deviations. Originally input values were experimentally measured at 20 time points and the other points were extrapolated between the measured values to calculate metabolic fluxes at interval of 0.01 h⁻¹. Violet boxes are substrates, orange boxes are products and blue boxes intracellular metabolites. Diamonds illustrates proteins and pathways involved in the given conversion of metabolites. ace - acetate, lact - lactate, etOH - ethanol, Glc - glucose, Orn - ornithine, Glx - glutamate + glutamine, Asx - aspartate + asparagine, PPP - peptose phosphate pathway, Pyr - pyrimidine synthesis, Pur - purine synthesis, I_AA_Glu_Group - the sum of consumption of arginine, proline, glutamine and glutamate, I_AA_Ala_Group - the sum of consumption of alanine, asparagine, aspartate, cysteine, glycine, serine and threonine, I_AA_His_Group - the sum of consumption of histidine, isoleucine, leucine, lysine, methionine, phenylalanine, tryptophan, tyrosine and valine, X_prod_Pyr - unmeasured products to balance carbon in the calculations, X_prod_AA3 - unmeasured products from histidine, isoleucine, leucine, lysine, methionine, phenylalanine, tryptophan, tyrosine and valine to balance carbon in the calculations and X_prod_Glu - unmeasured products from arginine, proline, glutamine and glutamate to balance carbon in the calculations.

Figure 5. Regulation patterns between mRNA-s and proteins.

Covariation coefficients between specific growth rate and mRNA, protein and pm (protein to mRNA ratio) ordered from top to down. Genes are sorted according to the covariance between specific growth rate and pm in decreasing order from left to right. Negative covariance between specific growth rate and pm is indicating that genes are regulated post-transcriptionally as the translation rate is decreasing compared to transcription rate i.e. there is less protein per mRNA with the increase of specific growth rate. Here are presented 50 most pronounced post-transcriptionally regulated genes. Red gene labels indicate that negative values of covariance between pm and specific growth rate are statistically significant (see the details of statistical hypothesis testing and P-value calculations in Materials and Methods). All covariation coefficients can be seen in Figure S2.

Figure 6. Multidimensional analysis of flux rates in A-stat cultures of Lactococcus lactis.

Analysis is based on protein abundance (prot, copies fl⁻¹) and apparent catalytic activities (k_cat, s⁻¹). In the three dimensional figures the average line and all three 2D projections are shown. The average values of proteins, specific fluxes and apparent catalytic activities are calculated based on all proteins present in the given pathway. Symbols are the same as described in the legend of Figure 4. More detailed picture can be seen in the Figure S3 and interactive figures including individual proteins of different pathways are available through the supplementary link.

Figure 7. Two dimensional heat map of apparent k_cat values grouped.

Grouping has been done according to the metabolic functions from Bolotin et al . The small figure next to the heat map illustrates the colors of two dimensions corresponding to the correlation between apparent k_cat values of two genes (1st dimension, red or blue) or the correlation between the difference of relative apparent k_cat value and relative increase in specific growth rate of two genes (2nd dimension, green or yellow). Color in the crossing of two genes explains whether the k_cat values change in the same direction as the specific growth rate increases (red, green) or not (blue, yellow), and whether the changes in relative k_cat values are higher than the changes in relative growth rate (red, blue) or not (green, yellow).