Chromochloris zofingiensis strain SAG 211–14 was obtained from the Culture Collection of Algae at Göttingen University (SAG) (Friedl and Lorenz, 2012). The liquid cultures were cultivated in 125 mL Erlenmeyer flasks and maintained on a shaker (100–150 rpm) under diurnal illumination (100 μmol photons m⁻² s⁻¹ with 16:8 light and dark cycle) at 25°C. The cultures were grown in a modified SDM (Jenkins et al., 2017) supplemented with Wolfe’s vitamins and Wolfe’s mineral solutions (ATCC, Manassas, VA, USA). The SDM was chosen to test the substrate use and secretion patterns of C. zofingiensis for common soil metabolites. All sugars and sugar alcohols in the original recipe were excluded in the modified SDM. The total carbon content of the medium was 0.40 mM from the SDM metabolites and 1.87 mM from the vitamin supplements, too low to be a meaningful source of energy nor contribute to cell growth. The detailed composition of the modified SDM is described in Supplementary Table S1.

A pre-culture of C. zofingiensis was grown in SDM as described above for 7 days to reach a cell density of 5 × 10⁶ cells mL⁻¹, measured using the Multisizer 3 Coulter Counter (Beckman Coulter, Brea, CA, USA). For the first phase of the experiment (transition to heterotrophy; +glucose phase), the treatment cultures were prepared in triplicate by resuspending pre-culture cells into SDM supplemented with 40 mM glucose (+glucose culture; n = 3). Control cultures were prepared in triplicate by also resuspending pre-culture cells in SDM but without glucose (continuous −glucose control; n = 3). Samples were collected at 0, 24, 48, 72, and 96 h after the beginning of the experiment. The second phase of the experiment (transition back to photoautotrophy; −glucose phase) started after 96 h of the first phase. Each +glucose culture was aliquoted into two sets and cells were collected by centrifugation. Then, cells from one aliquot were resuspended into fresh modified SDM to induce transition back to photoautotrophy (−glucose culture; n = 3), and cells from another aliquot were resuspended into modified SDM with 40 mM glucose (continuous +glucose control; n = 3). Samples were collected at 120, 144, 168, and 192 h after the start of the experiment (or 24, 48, 72, and 96 h after the start of the second phase). Uninoculated, but incubated sterile medium controls were included in triplicate for all phases of the experiment. Throughout the experiment, all cultures were grown under the 16:8 light:dark cycle as described above. Samples were collected at the same time every day during the light cycle. The experimental design is visualized in Figure 1.

Chromochloris zofingiensis was grown in SDM as described above until the culture reached a cell density of 5 × 10⁶ cells mL⁻¹. Then the culture was divided into forty-two 50 mL cultures (14 treatments x 3 replicates). For each treatment, each individual sugar was added to a final concentration of 240 mM carbon: 48 mM for each pentose (ribose and arabinose), 40 mM for each hexose (glucose, fructose, mannose, galactose, and rhamnose), 20 mM for each disaccharide (lactose, trehalose, maltose, cellobiose, and sucrose), and 13.3 mM for trisaccharide (raffinose). Maximum photosynthetic efficiency was measured as described above every 24 h after the sugar addition for 3 days.

Maximum photosynthetic efficiency was measured with the FMS2 system (Hansatech Instruments, Pentney, UK). After cells were dark acclimated while shaking for 30 min, 3.5 × 10⁶ cells were collected onto a glass fiber filter which was then placed into the instrument’s leaf clip. The maximum efficiency of PSII, (F_{m –} F₀)/F_m = F_v/F_m, was measured using a 0.5 s saturating pulse (at >2,000 μmol photons m⁻² s⁻¹). Cell number, cell mean diameter, and cell volume of the liquid culture were measured with a Multisizer 3 Coulter Counter (Beckman Coulter, Brea, CA, USA).

Intracellular and extracellular metabolites were extracted for LCMS analysis. Cultures were spun down at 12,000 g for 5 min. The supernatant was transferred to a new tube and then both cell pellets and growth medium supernatants were lyophilized. Media controls were included for supernatants. Empty tubes were included as extraction blanks. The lyophilized cell pellets were bead milled, using 3.2 mm stainless steel beads three times, for 5 s at a time using a BioSpec Mini-beadbeater-96 (Bartlesville, OK, USA). Just prior to analysis, dried materials were resuspended in 150 μL of LCMS grade methanol containing internal standards (Supplementary Table S2). The samples were vortexed and centrifuged for 5 min at 5,000 g to remove insoluble material. Supernatants were filtered through 0.22 μm PVDF Millipore Ultrafree centrifugal filter tubes (5 min at 5,000 g). Filtrates were transferred to amber glass vials with caps for LCMS analysis.

LC–MS/MS was performed on intracellular and extracellular extracts. Polar metabolites were separated with hydrophilic liquid interaction chromatography using and Agilent InfinityLab Poroshell 120 HILIC-Z column (2.1 × 150 mm. 2.7 μm) (Agilent, Hayward, CA, USA), with MS and MS/MS data collected using a Q Exactive Hybrid Quadrupole-Orbitrap Mass Spectrometer (ThermoFisher Scientific, Waltham, MA, USA). LC–MS/MS parameters are available in Supplementary Table S2. Samples consisted of four biological replicates each and four extraction controls, with sample injection order randomized and an injection blank run between each sample; internal and external standards were included for quality control purposes.

LC–MS/MS raw data were analyzed using MetAtlas (Yao et al., 2015). Briefly, samples were annotated by comparing mass-to-charge ratios (m/z), MS/MS spectra, and retention times (RT) with an in-house library of authentic reference standards. The metabolic pathways of the targeted metabolites were studied using the Kyoto Encyclopedia of Genes and Genomes (KEGG) (Kanehisa, 2019; Kanehisa et al., 2023; Kanehisa and Goto, 2000).

To calculate the fold changes of metabolites, the peak intensities of metabolites from treatment cultures (in trophic transition) were divided by those of their respective controls (replete media or continuous culture controls) and were log₁₀ transformed. For all calculations, the peak intensities of the replicates were averaged for each treatment or control. For endometabolomics, the peak intensities of the cell pellets were normalized by their respective average culture volumes. All statistics and calculations were conducted with R (version 4.3.1) (R Core Team, 2023). The significance of fold changes between treatment samples and their respective controls were tested by non-parametric Kruskal–Wallis test, commonly used for metabolomics data that are often sporadic and skewed, at ɑ = 0.05. Figures were created in the R environment using the package ggplot2 (Wickham, 2016).

The trophic transition of microalgae involves morphological and physiological changes that may also involve significant shifts in substrate use or exudation patterns (Roth et al., 2019a; Zhang H. et al., 2021; Zhao et al., 2014). Since C. zofingiensis is a terrestrial alga isolated from soil, we examined its substrate use in a soil defined media (SDM) which is composed of many commonly detected soil metabolites. Overall, lysine, arginine, adenine, guanine, and hypoxanthine in the media showed statistically significant decreases (p-value <0.05) between about 10-fold and 700-fold during both +glucose and − glucose phases (Figure 2A). This pattern was also observed during the continuous −glucose and + glucose controls (Figure 2B). Lysine during continuous −glucose control was an exception as this pattern was not as prominent as it was for the other metabolites (Figure 2B). These results indicated that C. zofingiensis strongly prefers these five exogenous metabolites regardless of trophic modes, except for lysine during continuous −glucose control. Supplementary Figure S1 visualizes the fold changes of the complete list of SDM metabolites.

We observed significant release of some metabolites into the media (Figures 2A,B). The extracellular level of orotic acid showed between 10- to 200-fold increases throughout the incubation for all treatment and control cultures (Figures 2A,B). The timeseries of orotic acid fold changes between individual cultures and the replete media control (Supplementary Figure S3) showed that its levels increased until the fourth day after the beginning of the experiment and plateaued with some fluctuations until the end of the incubation. This pattern was more consistent and stable with cultures in +glucose conditions, compared to −glucose conditions (Supplementary Figure S3). These results demonstrated that C. zofingiensis is likely secreting orotic acid up to a certain extracellular level and tends to maintain that level. Orotic acid is an intermediate of polyamine and pyrimidine synthesis and has been reported to be accumulated and secreted by fungi (Chiara et al., 2023; Nezu and Shimokawa, 2004; Silva et al., 2022; Swietalski et al., 2022). The endometabolite profile (Figure 2C) of C. zofingiensis did not show intracellular accumulation of orotic acid unlike these past reports. However, this species does have genes encoding dihydroorotase dehydrogenase, which catalyzes synthesis of dihydroorotate from glutamate metabolism into orotic acid (Roth et al., 2017). Nevertheless, this is the first report of orotic acid accumulation in the media by microalgae to our knowledge. Orotic acid has high pharmaceutical and industrial values as it is often used as a carrier in drug formulation and is an important substrate for pyrimidine production (Swietalski et al., 2022). Indeed, the reproducibility of our results should be tested, and further investigation is needed to understand why C. zofingiensis released orotic acid, and how this process is regulated. Yet, significant release of orotic acid into the media even under −glucose condition hints at potential biotechnological applications of this alga for orotic acid production for pharmaceutical applications and industrial pyrimidine production.

In the −glucose control culture spent media, sugars and sugar alcohols showed more than 800-fold decreases, suggesting strong uptake most likely to be spent as carbon sources or be stored (Figure 2B). Contrarily, strong exudation patterns were observed for these metabolites in the −glucose culture after transition from +glucose condition (Figure 2A). Sugar alcohols, both C5 and C6, showed a similar pattern as hexose (Figures 2A,B). Arabitol, one of C5 sugar alcohols, exudation has been reported in other organisms (Bowen and Northen, 2010), and both arabitol and mannitol, C6 sugar alcohol, are byproducts of glucose catabolism as observed in fungi (Adamczyk et al., 2023; Bernard et al., 1981; Diamantopoulou and Papanikolaou, 2023).

Overall, the profile of SDM endometabolites in C. zofingiensis in trophic transition showed an opposite pattern between sugar and sugar alcohols versus other metabolites. The intracellular levels of sugars and sugar alcohols showed significant (p-value <0.05) increases during +glucose condition compared to continuous −glucose control; they decreased after transition to −glucose condition when compared to continuous +glucose control (Figure 2C). Meanwhile, other metabolites showed an opposite and weaker trend where their intracellular levels decreased during +glucose phase, while increasing after transition to −glucose phase. These results suggested that internal levels of sugars and sugar alcohols tend to be greater during the transition to heterotrophy, while other metabolites tend to be lesser. This result is consistent with the exudation patterns of the sugars and sugar alcohols during +glucose phase as seen in exometabolomics results (Figures 2A,B) and supports our speculation that they are secreted as overflow metabolites from the catabolism of glucose. The endometabolite trend was more consistent for serine, glutamic acid, and aspartic acid compared to other metabolites. Indeed, these are important amino acids that are associated with some steps of the TCA cycle (Chen and Wang, 2021). Furthermore, serine is an important source of one-carbon units used in various metabolisms like nucleotide synthesis (Reid et al., 2018). Our endometabolomics results suggest that the consumption of metabolites is increased in general during the transition to heterotrophy, leading to relatively lower intensities of non-sugar metabolites than those of photoautotrophic cells. This is potentially to support higher cellular activities due to the abundant carbon sources.

Finally, we tested the effectiveness of different sugars to induce heterotrophy in the light for C. zofingiensis and confirmed glucose repressed photosynthesis the most. We compared the photosynthetic efficiency of the algae grown photoautotrophically versus grown on thirteen sugars individually. Compared to constant photoautotrophic control, all sugars except lactose decreased photosynthetic efficiency (F_v/F_m) to some level (Figure 4). Expectedly, glucose and fructose had the most consistent and dramatic effect on the F_v/F_m, reducing it from 0.62 down to 0.14 by day 3. While other hexoses, namely mannose, galactose, and rhamnose, showed comparable decrease in F_v/F_m up to day 2, the maximum photosynthetic efficiency started to recover rapidly by day 3, suggesting that they may have exhausted the sugar. Pentose, disaccharides, and trisaccharides all showed less impact on F_v/F_m, with maximum reduction down to only around 0.4. While maltose, sucrose, and arabinose showed continuous decrease by day 3, F_v/F_m either stabilized or recovered for all other pentoses, disaccharide, and trisaccharide. Overall, we found that glucose most dramatically and consistently induce the transition to heterotrophy in C. zofingiensis.