Spatiotemporal monitoring of intracellular metabolic dynamics by resonance Raman microscopy with isotope labeling

Abstract

Cellular metabolites are valuable in a diverse range of applications. For example, the unicellular green alga Haematococcus lacustris produces as a secondary metabolite the carotenoid pigment astaxanthin (AXT), which is widely used in nutraceutical, cosmetic, and food industries due to its strong antioxidant activity. In order to enhance the productivity of H. lacustris, spatial and temporal understanding of its metabolic dynamics is essential. Here we show spatiotemporal monitoring of AXT production in H. lacustris cells by resonance Raman microscopy combined with stable isotope labeling. Specifically, we incorporated carbon dioxide (¹³CO₂) labeled with a stable isotope (¹³C) into H. lacustris cells through carbon fixation and traced its conversion to ¹³C-AXT using our resonance Raman microscope. We incubated H. lacustris cells under various conditions by switching, pulsing, and replacing ¹³CO₂ and ¹²CO₂. By measurement of these cells we determined the fixation time of ¹³C-carbon, visualized the intracellular localization of ¹³C- and ¹²C-AXTs, and revealed the dynamic consumption–production equilibrium of the accumulated AXT. This work is a valuable step in the development of effective screening criteria for high AXT-producing H. lacustris cells.

---

Introduction

The observation and analysis of intracellular dynamics is essential to the study of cell biology.^1–6 For example, organelles undergo dynamic morphological changes during autophagy¹ or cell death² by apoptosis or necrosis. Microtubules³ and nucleolus⁴ form a wide range of spatial configurations in cell-cycle transitions including cell division, differentiation, and morphogenesis. Mitochondria mount metabolic defenses against intracellular pathogens in response to infection.^5,6 Beyond basic science, cellular metabolites produced by animal cells and microbes are widely used in the pharmaceutical, nutraceutical, cosmetic, and food industries,^7–10 and increasing metabolite production efficiency by cell engineering remains a major goal. For instance, antibodies produced by Chinese Hamster Ovary (CHO) cells are used as therapeutic agents,⁷ and secondary metabolites produced by microalgae cells are used as biofuels^8,9 and nutritional supplements.¹⁰ Metabolic fluxes in a living cell are highly regulated and varied in response to environmental changes, bioactive substances, and transitions of the cell cycle. Accordingly, the spatiotemporal observation of intracellular metabolic dynamics, including production, accumulation, and transportation, is foundational to the understanding of cellular metabolism. It is also effective for developing screening criteria for applications such as those above as well as for regenerative medicine and cell therapy.^11–13

Fluorescence microscopy, with the help of fluorescent probes,^14,15 is widely used to identify the intracellular localization and distribution of metabolites with subcellular resolution. However, such fluorescent labeling inevitably comes with several fundamental drawbacks when observing intracellular metabolic dynamics. First, the hydrophobicity of small-molecule fluorescent probes¹⁶ and the bulkiness of fluorescent protein labels^17,18 often disturb natural biological functions and lead to cellular toxicity. Second, intracellular fluorescent staining with fluorescent dyes suffers from non-specific binding and misdistribution, leading to potential errors in quantitative live imaging.^19,20 Third, sequential labelling such as pulse-chase analysis using protein tags²¹ is not applicable to the study of small metabolites. For these reasons, fluorescence microscopy is not always suitable for monitoring intracellular metabolic dynamics.

On the other hand, Raman microscopy together with stable isotope (SI) labeling^22–37 provides significant advantages in monitoring intracellular metabolite dynamics. First, since SI probes can be incorporated and metabolized in cells similarly to the original substrates, their use does not lead to cellular toxicity, interference with biological functions, or non-specific binding. Second, the metabolic process from incorporation to production can be tracked by exposing cells to an SI-labeled substrate and detecting SI-labeled products. Third, by analysis of the peaks of Raman spectra, Raman mapping quantifies the ratio of SI-labeled products to unlabeled products at spatial positions of interest. Though Raman microscopy with SI-labeled substrates has revealed the cellular anabolic processes of biopolymers including proteins,^24–31 DNAs,^26–28 and lipid droplets,^32–35 the spatiotemporal dynamics of secondary metabolites has not been monitored to date despite the usefulness of these metabolites in a wide range of applications.^38–41

In this paper, we show spatiotemporal monitoring of intracellular metabolic dynamics, for screening the production of secondary metabolites, using resonance Raman microscopy with an SI-labeled substrate exposed to various conditions (Fig. 1). Specifically, we investigated the biosynthesis process of a commercially valuable carotenoid, astaxanthin (AXT), produced by the unicellular photosynthetic alga Haematococcus lacustris,⁴² a species known to accumulate AXT under induced stress, usually accompanied by a drastic morphological transformation from green motile cells to red non-motile cyst cells.⁴³ While fluorescence from AXT is weak,⁴⁴ resonance Raman microscopy can provide an enhanced Raman signal from AXT^45–48 and can be used to selectively identify AXT in single H. lacustris cells from their primary cellular constituents including proteins, nucleotides, and lipids. We incorporated carbon dioxide (¹³CO₂) labeled by an SI (¹³C) into H. lacustris cells through carbon fixation and traced its conversion to ¹³C-AXT by resonance Raman microscopy with the SI. We incubated H. lacustris cells under various conditions by switching, pulsing, and replacing ¹³CO₂ and ¹²CO₂. By measurement of these cells we determined the fixation time of ¹³C-carbon, visualized the intracellular localization of ¹³C- and ¹²C-AXTs, and revealed the dynamic consumption–production equilibrium of the accumulated AXT. We believe this work represents a valuable step in the development of effective screening criteria for high AXT-producing H. lacustris cells.

Fig. 1 Schematic of our approach. We monitored the production of AXT in a single H. lacustris cell by resonance Raman microscopy with ¹³CO₂ and ¹²CO₂ incubation.

Results and discussion

Resonance Raman imaging of H. lacustris cells accumulating ¹³C-AXT

First, we performed resonance Raman imaging of H. lacustris cells incubated with ¹³CO₂ long enough for ¹³CO₂ to convert to ¹³C-AXT. We induced intracellular production of AXT for 5 days under nitrogen deficiency and strong light irradiation and measured the resonance Raman spectra of the cells. We obtained typical resonance Raman spectra of carotenoids including ν₁–ν₄ peaks from the resultant red cyst cells (Fig. 2a, red).⁴³ The ν₁–ν₄ peaks are assigned to vibrational modes as follows: ν₁, C=C stretching; ν₂, C–C stretching/C–H in-plane bending; ν₃, CH₃ rocking; ν₄, C–H out-of-plane wagging.^45,49 Spectra consist mainly of peaks corresponding to AXT, as it is the major ingredient (>90%) of carotenoid production by H. lacustris cells.⁵⁰ The AXT spectra of cells incubated with ¹³CO₂ are shifted to lower wavenumbers compared to those of cells incubated with ¹²CO₂ (Fig. 2a). This red shift is attributed to the substitution of ¹²C by the heavier isotope ¹³C, resulting in a lower frequency of molecular vibration. In particular, large shifts were observed in ν₁ and ν₂ peaks, which are assigned to vibrational modes of carbon–carbon bonds. The largest shift (32 cm⁻¹) was observed in the ν₁ peak, which shifted from 1518 cm⁻¹ (¹²C: green) to 1486 cm⁻¹ (¹³C: red). These results indicate that ¹³CO₂ was incorporated into H. lacustris cells through carbon fixation and that ¹³C-carbon was converted to AXT through the non-mevalonate pathway and subsequent carotenogenesis.⁵¹

Fig. 2 Resonance Raman spectroscopy of mixed H. lacustris cells that accumulate ¹³C- or ¹²C-AXT. (a) Resonance Raman spectra of intracellular AXT. Red: cells incubated with ¹³CO₂. Green: cells incubated with ¹²CO₂. (b) Bright-field image of the mixed cells. (c) Resonance Raman image of the mixed cells. ¹³C- and ¹²C-AXT are color-coded in red (¹³C) and green (¹²C), respectively. Scale bars: 40 μm. Color bars: intensities of Raman spectra of ¹³C- (red) and ¹²C-AXT (green).

We then demonstrated discrimination of H. lacustris cells accumulating ¹³C- or ¹²C-AXT by resonance Raman imaging based on the spectral shift. We mixed in equal number H. lacustris cells that accumulated ¹³C- and those that accumulated ¹²C-AXT, and then analyzed the cell mixture by Raman mapping. Raman images were generated by classical least squares modeling of the ¹³C- and ¹²C-AXT spectra obtained during the long-term induction. At each image pixel position, a linear combination of factors was calculated to estimate the measured spectrum. ¹³C- and ¹²C-AXT in the images were color-coded as red and green, respectively. As a result, the two types of cells were clearly discriminated (Fig. 2c). This result shows the advantage of Raman imaging over bright field microscopy (Fig. 2b) for detecting targets labeled with stable isotopes.

Resonance Raman spectroscopy of H. lacustris cells incubated with a mixture of ¹³CO₂ and ¹²CO₂

Second, we investigated the resonance Raman spectral shift of AXT in H. lacustris cells incubated with ¹³CO₂ and ¹²CO₂ mixed in various ratios. We induced production of AXT in the cells under various ¹³CO₂ ratios (¹³CO₂: 0, 25, 50, 75, and 100 vol%) for 24 h (Fig. 3a). The Raman peaks of AXT (ν₁ and ν₂) gradually shifted toward lower wavenumber positions as the ¹³CO₂ ratio increased (Fig. 3b). When plotting the ratio of ¹³CO₂ versus the wavenumber of each peak (ν₁, ν_2a, and ν_2b), all showed a linear relationship (Fig. 3c). The slope (m) of the ν₁ peak, attributed to C=C stretching, has a larger absolute value (ν₁: m = −0.33) than that of the ν₂ peaks, which include a contribution from C–C stretching (ν_2a: m = −0.25, ν_2b: m = −0.22). These results are congruent with observed resonance Raman spectral shifts of carotenoids (mainly β-carotene) in cyanobacterial cells incubated with a ¹³C source.²² H. lacustris cells fix ¹³CO₂ and ¹²CO₂ gases with equal probability, since stable isotopes have the same biochemical properties.⁵² Consequently, the carbon ratio of ¹³C and ¹²C in an AXT molecule produced in a cell is presumably the same as the ratio of ¹³CO₂ and ¹²CO₂ in the culturing atmosphere. On the other hand, the ratio of the ¹³C carbon–carbon bond wavenumber (σ_¹³C) to the ¹²C carbon–carbon bond wavenumber (σ_¹²C) corresponds to the square root of the ratio of the reduced masses, given that the wavenumber is defined as: σ = √k/2πc√μ, where k is the spring constant of the corresponding chemical bond and μ is the reduced mass.³⁰ In fact, all of the ratios of resonance Raman peaks (σ_¹³C-100%/σ_¹²C-100%) are constant (ν₁, 0.979; ν_2a, 0.983; ν_2b, 0.978; Fig. 3b and c) and approximately consistent with the theoretical value (σ_¹³C/σ_¹²C = 0.960). These results show that the resonance Raman spectra of AXT in H. lacustris cells incubated with a mixture of ¹³CO₂ and ¹²CO₂ shift to lower wavenumbers in a linear manner corresponding to the ¹³CO₂ ratio in the culturing atmosphere.

Fig. 3 Resonance Raman spectroscopy of H. lacustris cells incubated with mixture of ¹³CO₂ and ¹²CO₂. (a) Incubation timeline of H. lacustris cells with ¹³CO₂ and ¹²CO₂ gases mixed at various ratios. (b) Resonance Raman spectra of H. lacustris cells incubated at various ¹³CO₂ ratios. (c) Raman shifts of the peaks (ν_2a, ν_2b, ν₁) against the ratio of ¹³CO₂ in the incubation atmosphere.

Resonance Raman imaging of H. lacustris cells incubated with atmosphere switching from ¹³CO₂ to ¹²CO₂

Third, we switched the incubation atmosphere from ¹³CO₂ to ¹²CO₂ during AXT induction, and then investigated the ratios of ¹³C- and ¹²C-AXT in H. lacustris cells by resonance Raman imaging. We varied the time to switch from ¹³CO₂ to ¹²CO₂ during the 24 h incubation period to investigate cells incubated with ¹³CO₂ for 12 h, 15 h, 18 h, 21 h, and 24 h (Fig. 4a). In most cases, resonance Raman spectra of the H. lacustris cells were a superposition of the ¹³C-AXT spectrum (peaks: ν₂, 1128 cm⁻¹ and 1173 cm⁻¹; ν₁, 1484 cm⁻¹; in red) and the ¹²C-AXT spectrum (peaks: ν₂, 1156 cm⁻¹ and 1192 cm⁻¹; ν₁, 1514 cm⁻¹; in green) with varying intensities of the two spectra (representative four spectra are shown in Fig. 4b). This is in contrast to the gradual red shift observed in spectra of cells incubated with a mixture of ¹³CO₂ and ¹²CO₂ (Fig. 3). The results of this atmosphere switching experiment indicate that cells contain both 100% ¹³C-AXT and 100% ¹²C-AXT in various proportions, with no or a negligible amount of AXT molecules composed of both ¹³C- and ¹²C-carbons. This can be attributed to the rapid conversion rate from carbon source to AXT molecule, due to the small molecular size of AXT. In other words, after switching from ¹³CO₂ to ¹²CO₂, the ¹³C-carbon source was consumed rapidly for AXT production, and then the carbon source quickly switched to ¹²C-carbon, resulting in productions of 100% ¹³C-AXT and 100% ¹²C-AXT. In resonance Raman images generated by the classical least squares modeling of the ¹³C- and ¹²C-AXT spectra, H. lacustris cells are represented as a gradation of red (¹³C-AXT) and green (¹²C-AXT) in an RGB color model (e.g., Cell 1, yellow-green; Cell 2, yellow; Cell 3, orange; Cell 4, red-orange; Fig. 4b), since they include both 100% ¹³C-AXT and 100% ¹²C-AXT. On the other hand, AXT inside of a cell was distributed non-uniformly, suggesting that the AXT-accumulation process was still in a transition period to maturation after AXT-induction at 24 h (Fig. 4b). Notably, localization of ¹³C-AXT (red) and ¹²C-AXT (green) was observed inside some cells (Fig. S1†), revealing the progression of carbon fixation and conversion to AXT.

Fig. 4 Resonance Raman imaging of H. lacustris cells incubated with atmosphere switching from ¹³CO₂ to ¹²CO₂. (a) Incubation timeline of H. lacustris cells with incubation switching from ¹³CO₂ to¹²CO₂ at various timings. (b) Resonance Raman images and spectra of H. lacustris cells that accumulate both ¹³C- and ¹²C-AXT. ¹³C- and ¹²C-AXT are color-coded in red (¹³C) and green (¹²C), respectively. Cell 1, 12 h; Cell 2, 15 h; Cell 3, 15 h; Cell 4, 18 h. Scale bars: 20 μm. (c) Low-magnification resonance Raman images of H. lacustris cells. Red, ¹³C-AXT; green, ¹²C-AXT. Scale bars: 100 μm. Color bars: intensities of Raman spectra of ¹³C- (red) and ¹²C-AXT (green). (d) Box-and-whisker plot of ¹³C-AXT ratios in the resonance Raman spectra of H. lacustris cells (n = 30).

We then observed the H. lacustris cells with low-magnification resonance Raman images and investigated the comprehensive changes of ¹³C-AXT in the cells (Fig. 4c). As the incubation time with ¹³CO₂ increased, green color (¹²C-AXT) decreased and red color (¹³C-AXT) increased, as observed in resonance Raman images of cells incubated with ¹³CO₂ for 12 h, 15 h, 18 h, and 21 h (Fig. 4c). This gradual change in color indicates that H. lacustris cells have CO₂ fixation activity during the 12 h to 21 h period and then complete the conversion to AXT by the end of the incubation (24 h). On the other hand, comparing the images at 21 h and 24 h, no significant color change is observed, showing only red color (¹³C-AXT) without green color (¹²C-AXT). This result indicates that the conversion process from ¹²CO₂ fixation to ¹²C-AXT production was incomplete during this period (21–24 h). In order to quantitatively support this, we measured resonance Raman spectra of 30 cells (at the center of the cells) randomly selected from individual images and calculated the ¹³C-AXT ratios by spectral separation using a multivariate curve resolution–alternating least squares (MCR-ALS) algorithm (Fig. 4d). From these spectra, we confirmed that ¹³C-AXT accumulated until 21 h and became saturated afterwards. We then investigated the activity of CO₂ fixation and AXT production after 24 h. After 24 h of ¹³CO₂, the incubation atmosphere was switched from ¹³CO₂ to ¹²CO₂ at 6 hour intervals, and the total AXT-induction time was 48 h (Fig. S2a†). In these cases, accumulation rate of ¹³C-AXT decreased (Fig. S2b and c†), which indicates that the conversion activity from 24 h to 48 h was lower than that of the 12 h to 24 h period. As demonstrated above, the active conversion period from CO₂ fixation to AXT production can be probed by switching from ¹³CO₂ to ¹²CO₂ at various times and comparing the difference of ¹³C- or ¹²C-AXT accumulation by resonance Raman imaging.

Resonance Raman imaging of ¹³C-AXT in H. lacustris cells labeled by pulse incubation with ¹³CO₂

Fourth, we labeled AXT in H. lacustris cells by pulse incubation, where the duration of incubation with ¹³CO₂ was shorter, and detected ¹³C-AXT by resonance Raman imaging to specify the carbon fixation time. In the aforementioned incubations with the switching of ¹³CO₂ to ¹²CO₂, whether ¹³CO₂ was fixed at earlier or later times could not be determined, since the ¹³C-AXT indicated by the red color accumulated during the ¹³CO₂ incubation. In this experiment, we incubated the cells with ¹³CO₂ for a constant “pulse” of time (9 h), with the remaining incubation under ¹²CO₂ (15 h), for a total AXT induction time of 24 h. The ¹³CO₂ pulse incubations were shifted by 3 h in the timeline (Fig. 5a) to probe the carbon fixation time. We calculated the ¹³C-AXT ratios of 30 cells randomly selected for each ¹³CO₂ pulse timing (Fig. 5b). The plot shows that ¹³CO₂ fixation and the resultant ¹³C-AXT production are significantly higher in the late ¹³C-induction phases ((iv) 9–18 h, (v) 12–21 h, (vi) 15–24 h), while both are lower in the early ¹³C-induction phases ((i) 0–9 h, (ii) 3–12 h, (iii) 6–15 h). Notably, there is a large gap between conditions (iii) and (iv), indicating that CO₂ is actively fixed and converted into AXT after 15 h. Among the late pulse incubations, conditions (v) and (vi) show higher ¹³C-AXT ratios than condition (iv) (t-test; p < 0.05), whereas no significant difference was observed (t-test; p = 0.20) between (v) and (vi). These results indicate that the activity of carbon-fixation and conversion is high during the 15 to 21 h period of incubation.

Fig. 5 Resonance Raman imaging of ¹³C-AXT in H. lacustris cells labeled by pulse incubation with ¹³CO₂. (a) Incubation timeline of H. lacustris cells with ¹³CO₂ incubation pulsed (9 h) at various timings in 24 h. (b) Box-and-whisker plot of ¹³C-AXT ratios in the resonance Raman spectra of H. lacustris cells (n = 30). ¹²C shows a control culture with ¹²CO₂ incubation. (c) Low-magnification resonance Raman images of H. lacustris cells. The ratio of ¹³C-AXT is represented in pseudo-color. Scale bars: 100 μm.

We then compared the ¹³C-AXT ratios of H. lacustris cells in the same resonance Raman image frame. We represent only ¹³C-AXT in pseudo color (Fig. 5c) to highlight ¹³C-carbon fixed during the pulse incubation, rather than representing both ¹³C-AXT (red) and ¹²C-AXT (green) in the RGB color model (Fig. S3a†). Cells that actively fixed ¹³CO₂ and converted it to ¹³C-AXT are clearly detected in the image of condition (v) 12–21 h, while such cells were scarcely observed in the image of condition (i) 0–9 h or the image of the control condition, which was incubation under only ¹²CO₂. These results demonstrate that this method of pulse incubation can be useful for screening active H. lacustris cells which show a high efficiency of carbon-fixation and AXT-conversion.

Resonance Raman imaging of H. lacustris cells after replacing ¹³C-AXT by ¹²C-AXT by ¹²CO₂ incubation

Lastly, we accumulated ¹³C-AXT in H. lacustris cells and then replaced it with ¹²C-AXT by incubating the cells under ¹²CO₂. We induced ¹³C-AXT production in H. lacustris cells for 24 h by incubation under ¹³CO₂ and then replaced the atmosphere with ¹²CO₂ for different durations (3 h increments) before observation (Fig. 6a). Calculation of the ¹³C-AXT ratios of 30 cells randomly selected for each condition showed a large dispersion of ¹³C-AXT ratios, which reflects the difference in ¹²C-AXT production activity among the cells (Fig. 6b). Interestingly, we observed that ¹³C-AXT in some cells was completely replaced with ¹²C-AXT after 12 h (Fig. 6b and c), suggesting that AXT is in dynamic equilibrium between consumption and production. This information on carbon flow cannot be obtained by merely detecting non-labeled AXT and thus demonstrates the advantage of combining resonance Raman imaging with ¹³C-labeling. Furthermore, we observed heterogeneity of ¹³C-AXT (red) and ¹²C-AXT (green) in a single cell after incubation with ¹²CO₂, but the AXTs became homogenized as the time of incubation with ¹²CO₂ was extended (Fig. 7). Specifically, ¹²C-AXT infiltrated from the edge of the cell towards its center (Fig. 7a). In fact, the resonance Raman spectra around the cell periphery confirmed that the ratio of ¹²C-AXT was high at the edge and low at the center (Fig. 7b). This observation can be attributed to the transportation of newly produced ¹²C-AXT precursors to lipid droplets around the center.^53,54 In an intermediate cell during encystment, chloroplasts, which produce an AXT precursor, are located around the cell periphery, while oil droplets, which serve as a “sink” for AXT, are located around the nucleus at the center.^53,54 The newly produced 100% ¹²C-AXT precursors are delivered to the oil droplets from the periphery, converted to ¹²C-AXT by an enzyme inside, and mixed with pre-existing 100% ¹³C-AXT. Consequently, a gradient of ¹²C-AXT and ¹³C-AXT forms, causing heterogeneity in a single cell. These results demonstrate that the distribution of AXTs produced at different times can be visualized in a single cell by resonance Raman imaging combined with ¹³C- and ¹²C-labeling.

Fig. 6 Resonance Raman imaging of replacing ¹³C-AXT by ¹²C-AXT in H. lacustris cells by incubating with ¹²CO₂. (a) Incubation timeline of H. lacustris cells with ¹²CO₂ incubation following 24 hours of ¹³CO₂ incubation. (b) Box-and-whisker plot of ¹³C-AXT ratios in the resonance Raman spectra of H. lacustris cells (n = 30). (c) Low-magnification resonance Raman images of H. lacustris cells. ¹³C- and ¹²C-AXT are color-coded in red (¹³C) and green (¹²C), respectively. Scale bars: 100 μm. Color bars: intensities of Raman spectra of ¹³C- (red) and ¹²C-AXT (green).

Fig. 7 Resonance Raman imaging of the heterogeneity of ¹³C-AXT and ¹²C-AXT in a single H. lacustris cell. (a) Resonance Raman images of H. lacustris cells incubated with ¹²CO₂ after incubation with ¹³CO₂. The incubation conditions used here are identical to those in Fig. 6(b and c). Magnified resonance Raman images (b) and spectra (c) of H. lacustris cells incubated with ¹²CO₂ for 3 h after their accumulation of ¹³C-AXT. The spectra correspond to the spots in the images of (b). ¹³C- and ¹²C-AXT are color-coded in red (¹³C) and green (¹²C), respectively. Scale bars: 20 μm. Color bars: intensities of Raman spectra of ¹³C- (red) and ¹²C-AXT (green).

Conclusions

In this paper, we demonstrated spatiotemporal monitoring of AXT production in H. lacustris cells, incubated under varying ¹³CO₂ and ¹²CO₂ atmospheric conditions, by resonance Raman imaging. First, we induced AXT production in H. lacustris by incubating the cells with ¹³CO₂ for enough time to produce ¹³C-AXT, the resonance Raman spectrum of which shows red shifts compared to that of ¹²C-AXT. This result shows that ¹³CO₂ was incorporated into H. lacustris cells through carbon fixation and that ¹³C-carbon was converted to AXT through cellular metabolism. Based on the spectral difference between ¹³C- and ¹²C-AXT, we clearly discriminated H. lacustris cells accumulating ¹³C- or ¹²C-AXT by resonance Raman imaging. Second, we investigated the resonance Raman spectral shift of AXT in H. lacustris cells when incubated with a mixture of ¹³CO₂ and ¹²CO₂ gases in various ratios. We observed that the resonance Raman spectra shifted to positions at lower wavenumbers in a linear manner corresponding to the ¹³CO₂ ratio of the culturing atmosphere. Third, we switched the incubation atmosphere from ¹³CO₂ to ¹²CO₂ at various times and then investigated the ratios of ¹³C- and ¹²C-AXT in the cells by resonance Raman imaging. The results indicated that the cells contained both 100% ¹³C-AXT and 100% ¹²C-AXT in various proportions. By comparing via resonance Raman imaging the differences in ¹³C- and ¹²C-AXT accumulation, the active conversion period from CO₂ fixation to AXT production could be probed. Fourth, we labeled AXT with ¹³C by incubating cells under ¹³CO₂ for short pulses of time, to specify the active time of carbon fixation and conversion. Lastly, we accumulated ¹³C-AXT in H. lacustris cells in advance and subsequently replaced it with ¹²C-AXT by incubating the cells under ¹²CO₂ for different durations. We observed a high variation between cells in the rate at which ¹³C-AXT decreased, reflecting individual differences in ¹²C-AXT production activity among cells. In some cells, ¹³C-AXT was completely replaced with ¹²C-AXT, suggesting that AXT is in dynamic equilibrium between consumption and production. Furthermore, localizations of ¹³C-AXT and ¹²C-AXT in a single cell were observed by magnified resonance Raman imaging.

Our method provides spatiotemporal information for single cells, including the conversion process from substrate to metabolite, the distribution process of the metabolite, and the turnover of the metabolite in dynamic equilibrium. By virtue of non-cytotoxic labelling and non-invasive imaging, long-term monitoring of metabolic dynamics in living cells can be achieved. Furthermore, by integrating our method with technologies for single cell isolation,^55,56 the intracellular metabolism of an identified cell can be monitored. When combined with state-of-the-art Raman spectroscopy techniques that have higher spatiotemporal resolution, such as stimulated Raman scattering (SRS)^57–60 or coherent anti-Stokes Raman scattering (CARS),^61–63 the process of metabolite distribution can be monitored in a single cell with further detail. Leveraging such information for biotechnological production of target metabolites could provide significant value in terms of productivity. Furthermore, using Raman-activated cell sorting (RACS)^64–66 will enable screening for cells with high metabolite productivity from a genetically diverse cell population.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was supported primarily by the ImPACT Program of the CSTI (Cabinet Office, Government of Japan) and partly by the JSPS Core-to-Core Program and White Rock Foundation. The authors thank Prof. Tomohisa Hasunuma at Kobe University for discussion.

References

1. N. Mizushima, T. Yoshimori and B. Levine, Cell, 2010, 140, 313–326.
2. O. Kepp, L. Galluzzi, M. Lipinski, J. Yuan and G. Kroemer, Nat. Rev. Drug Discovery, 2011, 10, 221–237.
3. J. Zenker, M. D. White, R. M. Templin, R. G. Parton, O. Thorn-Seshold, S. Bissiere and N. Plachta, Science, 2017, 357, 925–928.
4. A. Sakaue-Sawano, H. Kurokawa, T. Morimura, A. Hanyu, H. Hama, H. Osawa, S. Kashiwagi, K. Fukami, T. Miyata, H. Miyoshi, T. Imamura, M. Ogawa, H. Masai and A. Miyawaki, Cell, 2008, 132, 487–498.
5. L. Pernas, C. Bean, J. C. Boothroyd and L. Scorrano, Cell Metab., 2018, 27, 886–897.e4.
6. T. Yoshizumi, T. Ichinohe, O. Sasaki, H. Otera, S. Kawabata, K. Mihara and T. Koshiba, Nat. Commun., 2014, 5, 4713.
7. J. Y. Kim, Y. G. Kim and G. M. Lee, Appl. Microbiol. Biotechnol., 2012, 93, 917–930.
8. M. S. Cooper, W. R. Hardin, T. W. Petersen and R. A. Cattolico, J. Biosci. Bioeng., 2010, 109, 198–201.
9. K. Yamada, H. Suzuki, T. Takeuchi, Y. Kazama, S. Mitra, T. Abe, K. Goda, K. Suzuki and O. Iwata, Sci. Rep., 2016, 6, 26327.
10. M. A. Borowitzka, J. Appl. Phycol., 2013, 25, 743–756.
11. J. Nielsen and J. D. Keasling, Cell, 2016, 164, 1185–1197.
12. J. C. Caicedo, S. Cooper, F. Heigwer, S. Warchal, P. Qiu, C. Molnar, A. S. Vasilevich, J. D. Barry, H. S. Bansal, O. Kraus, M. Wawer, L. Paavolainen, M. D. Herrmann, M. Rohban, J. Hung, H. Hennig, J. Concannon, I. Smith, P. A. Clemons, S. Singh, P. Rees, P. Horvath, R. G. Linington and A. E. Carpenter, Nat. Methods, 2017, 14, 849–863.
13. K. D. Piatkevich, E. E. Jung, C. Straub, C. Linghu, D. Park, H. J. Suk, D. R. Hochbaum, D. Goodwin, E. Pnevmatikakis, N. Pak, T. Kawashima, C. T. Yang, J. L. Rhoades, O. Shemesh, S. Asano, Y. G. Yoon, L. Freifeld, J. L. Saulnier, C. Riegler, F. Engert, T. Hughes, M. Drobizhev, B. Szabo, M. B. Ahrens, S. W. Flavell, B. L. Sabatini and E. S. Boyden, Nat. Chem. Biol., 2018, 14, 352–360.
14. J. W. Lichtman and J. A. Conchello, Nat. Methods, 2005, 2, 910–919.
15. Z. Liu, L. D. Lavis and E. Betzig, Mol. Cell, 2015, 58, 644–659.
16. L. Wang, M. S. Frei, A. Salim and K. Johnsson, J. Am. Chem. Soc., 2019, 141, 2770–2781.
17. H.-S. Liu, M.-S. Jan, C.-K. Chou, P.-H. Chen and N.-J. Ke, Biochem. Biophys. Res. Commun., 1999, 260, 712–717.
18. M. R. Depaoli, H. Bischof, E. Eroglu, S. Burgstaller, J. Ramadani-Muja, T. Rauter, M. Schinagl, M. Waldeck-Weiermair, J. C. Hay, W. F. Graier and R. Malli, Pharmacol. Ther., 2019, 202, 98–119.
19. M. C. Wang, W. Min, C. W. Freudiger, G. Ruvkun and X. S. Xie, Nat. Methods, 2011, 8, 135–138.
20. S. Yamashiro, D. Taniguchi, S. Tanaka, T. Kiuchi, D. Vavylonis and N. Watanabe, Biophys. J., 2019, 116, 142–150.
21. A. Gautier, A. Juillerat, C. Heinis, I. R. Correa Jr, M. Kindermann, F. Beaufils and K. Johnsson, Chem. Biol., 2008, 15, 128–136.
22. M. Li, D. P Canniffe, P. J. Jackson, P. A. Davison, S. FitzGerald, M. J. Dickman, J. G. Burgess, C. N. Hunter and W. E. Huang, ISME J., 2012, 6, 875–885.
23. D. McIlvenna, W. E. Huang, P. Davison, A. Glidle, J. Coopera and H. Yin, Lab Chip, 2016, 16, 1420–1429.
24. H. N. Noothalapati Venkata and S. Shigeto, Chem. Biol., 2012, 19, 1373–1380.
25. Y. Shen, F. Xu, L. Wei, F. Hu and W. Min, Angew. Chem., Int. Ed., 2014, 53, 5596–5599.
26. Y. Wang, Y. Ji, E. S. Wharfe, R. S. Meadows, P. March, R. Goodacre, J. Xu and W. E. Huang, Anal. Chem., 2013, 85, 10697–10701.
27. M. Li, W. E. Huang, C. M. Gibson, P. W. Fowler and A. Jousset, Anal. Chem., 2013, 85, 1642–1649.
28. P. Kubryk, J. S. Kölschbach, S. Marozava, T. Lueders, R. U. Meckenstock, R. Niessner and N. P. Ivleva, Anal. Chem., 2015, 87, 6622–6630.
29. H.-J. van Manen, A. Lenferink and C. Otto, Anal. Chem., 2008, 80, 9576–9582.
30. L. Wei, Y. Yu, Y. Shen, M. C. Wang and W. Min, Proc. Natl. Acad. Sci. U. S. A., 2013, 110, 11226–11231.
31. L. Wei, Y. Shen, F. Xu, F. Hu, J. K. Harrington, K. L. Targoff and W. Min, ACS Chem. Biol., 2015, 10, 901–908.
32. H.-J. van Manen, Y. M. Kraan, D. Roos and C. Otto, Proc. Natl. Acad. Sci. U. S. A., 2005, 102, 10159–10164.
33. D. Zhang, M. N. Slipchenko and J.-X. Cheng, J. Phys. Chem. Lett., 2011, 2, 1248–1253.
34. J. Li and J.-X. Cheng, Sci. Rep., 2014, 4, 6807.
35. F. Hu, L. Wei, C. Zheng, Y. Shen and W. Min, Analyst, 2014, 139, 2312–2317.
36. D. Berry, E. Mader, T. K. Lee, D. Woebken, Y. Wang, D. Zhu, M. Palatinszky, A. Schintlmeister, M. C. Schmid, B. T. Hanson, N. Shterzer, I. Mizrahi, I. Rauch, T. Decker, T. Bocklitz, J. Popp, C. M. Gibson, P. W. Fowler, W. E. Huang and M. Wagner, Proc. Natl. Acad. Sci. U. S. A., 2014, 112, 194–203.
37. Y. Yonamine, Y. Suzuki, T. Ito, Y. Miura, K. Goda, Y. Ozeki and Y. Hoshino, ChemBioChem, 2017, 18, 2063–2068.
38. M. A. Borowitzka, J. Appl. Phycol., 1995, 7, 3–15.
39. M. Donia and M. T. Hamann, Lancet Infect. Dis., 2003, 3, 338–348.
40. B. Singh, T. K. Bhat and B. Singh, J. Agric. Food Chem., 2003, 51, 5579–5597.
41. A. L. Demain, Appl. Microbiol. Biotechnol., 1999, 52, 455–463.
42. M. Guerin, M. E. Huntley and M. Olaizola, Trends Biotechnol., 2003, 21, 210–216.
43. S. Boussiba, Physiol. Plant., 2000, 108, 111–117.
44. K. Ukibe, T. Katsuragi, Y. Tani and H. Takagi, FEMS Microbiol. Lett., 2008, 286, 241–248.
45. B. Robert, Photosynth. Res., 2009, 101, 147–155.
46. A. Kaczor, K. Turnau and M. Baranska, Analyst, 2011, 136, 1109–1112.
47. A. M. Collins, H. D. Jones, D. Han, Q. Hu, T. E. Beechem and J. A. Timlin, PLoS One, 2011, 6, e24302.
48. S. R. Fagerer, T. Schmid, A. J. Ibanez, M. Pabst, R. Steinhoff, K. Jefimovs, P. L. Urban and R. Zenobi, Analyst, 2013, 138, 6732–6736.
49. S. Sato and M. Tasumi, J. Raman Spectrosc., 1983, 14, 310–321.
50. L. Recht, N. Topfer, A. Batushansky, N. Sikron, Y. Gibon, A. Fait, Z. Nikoloski, S. Boussiba and A. Zarka, J. Biol. Chem., 2014, 289, 30387–30403.
51. K. Grunewald, J. Hirschberg and C. Hagen, J. Biol. Chem., 2001, 276, 6023–6029.
52. B. Koletzko, T. Sauerwald and H. Demmelmair, Eur. J. Pediatr., 1997, 156, S12–S17.
53. U. Pick, A. Zarka, S. Boussiba and L. Davidi, Planta, 2019, 249, 31–47.
54. M. Wayama, S. Ota, H. Matsuura, N. Nango, A. Hirata and S. Kawano, PLoS One, 2013, 8, e53618.
55. A. E. Vasdekis and G. Stephanopoulos, Metab. Eng., 2015, 27, 115–135.
56. N. Ota, Y. Yonamine, T. Asai, Y. Yalikun, T. Ito, Y. Ozeki, Y. Hoshino and Y. Tanaka, Anal. Chem., 2019, 91, 9631–9639.
57. B. G. Saar, C. W. Freudiger, J. Reichman, C. M. Stanley, G. R. Holtom and X. S. Xie, Science, 2010, 330, 1368–1370.
58. Y. Ozeki, W. Umemura, Y. Otsuka, S. Satoh, H. Hashimoto, K. Sumimura, N. Nishizawa, K. Fukui and K. Itoh, Nat. Photonics, 2012, 6, 845–851.
59. Y. Wakisaka, Y. Suzuki, O. Iwata, A. Nakashima, T. Ito, M. Hirose, R. Domon, M. Sugawara, N. Tsumura, H. Watarai, T. Shimobaba, K. Suzuki, K. Goda and Y. Ozeki, Nat. Microbiol., 2016, 1, 16124.
60. Y. Suzuki, K. Kobayashi, Y. Wakisaka, D. Deng, S. Tanaka, C.-J. Huang, C. Lei, C.-W. Sun, H. Liu, Y. Fujiwaki, S. Lee, A. Isozaki, Y. Kasai, T. Hayakawa, S. Sakuma, F. Arai, K. Koizumi, H. Tezuka, M. Inaba, K. Hiraki, T. Ito, M. Hase, S. Matsusaka, K. Shiba, K. Suga, M. Nishikawa, M. Jona, Y. Yatomi, Y. Yalikun, Y. Tanaka, T. Sugimura, N. Nitta, K. Goda and Y. Ozeki, Proc. Natl. Acad. Sci. U. S. A., 2019, 116, 15842–15848.
61. C. L. Evans, E. O. Potma, M. Pouris'haag, D. Cote, C. P. Lin and X. S. Xie, Proc. Natl. Acad. Sci. U. S. A., 2005, 102, 16807–16812.
62. K. Hashimoto, M. Takahashi, T. Ideguchi and K. Goda, Sci. Rep., 2016, 6, 21036.
63. M. Tamamitsu, Y. Sakaki, T. Nakamura, G. K. Podagatlapalli, T. Ideguchi and K. Goda, Vib. Spectrosc., 2016, 91, 163–169.
64. D. A. Watson, L. O. Brown, D. F. Gaskill, M. Naivar, S. W. Graves, S. K. Doorn and J. P. Nolan, Cytometry, Part A, 2008, 73, 119–128.
65. Q. Zhang, P. Zhang, H. Gou, C. Mou, W. E. Huang, M. Yang, J. Xu and B. Ma, Analyst, 2015, 140, 6163–6174.
66. K. Hiramatsu, T. Ideguchi, Y. Yonamine, S.-W. Lee, Y. Luo, K. Hashimoto, T. Ito, M. Hase, J.-W. Park, Y. Kasai, S. Sakuma, T. Hayakawa, F. Arai, Y. Hoshino and K. Goda, Sci. Adv., 2019, 5, eaau0241.

---

Footnote

† Electronic supplementary information (ESI) available: Experimental section and supporting figures. See DOI: 10.1039/d0ra02803g

---