Research ArticleAPPLIED ECOLOGY

PhenoChip: A single-cell phenomic platform for high-throughput photophysiological analyses of microalgae

See allHide authors and affiliations

Science Advances  02 Sep 2020:
Vol. 6, no. 36, eabb2754
DOI: 10.1126/sciadv.abb2754
  • Fig. 1 PhenoChip, a microfluidic platform that integrates single-cell immobilization and delivery of physicochemical gradients with photophysiological imaging.

    (A) Schematic of the PhenoChip microfluidic device, composed of three principal design elements for cell immobilization, generation of temperature gradients, and generation of chemical gradients. Cells are immobilized individually within the main channel (width, 2.28 mm; height, 75 μm; length, 20.5 mm) in an arena of approximately 29,000 microwells. (B) In the main channel, the microwells (here, 20-μm diameter, 15-μm depth) are seeded with cells using gravitational deposition followed by slight compression of the flexible PDMS device. Loading can be confirmed by light microscopy (left, empty microwell; right, microwell containing a single microalgal cell). (C) For noninvasive, real-time photophysiological imaging of single cells, PhenoChip is coupled to a variable chlorophyll fluorescence imaging microscope (PAM). The image shows a PAM-derived, false-color image of quantum yields of immobilized single microalgal cells within a portion of the microwell array comprising ca. 1500 wells. (D) Photographic image of PhenoChip with colored solutes injected into both inlet ports (1) and the resulting gradient across the microwell arena (2). Temperature sink (3) and source (4) channels as well as the outlet (5) are shown with their respective transparent tubing or electrically conductive wires (black/red).

  • Fig. 2 PhenoChip enables the accurate application of thermal gradients.

    (A) Visual representation of temperature gradients within PhenoChip. Temperature gradients were visualized using temperature-sensitive liquid crystal (LC) sheets fixed to the glass microscope slide beneath the PhenoChip. Rows of microwells within PhenoChip (total n = 57) are indicated by horizontal lines. A gradient in color represents the gradient in temperature (green, low temperature; blue, high temperature) during the application (“‘ON,” left) and absence (“‘OFF,” right) of an electrical current. (B) CFD simulation of the temperature gradient across the 57 rows of microwells within PhenoChip, for five values of temperature (room temperature or “RT,” +1°, +2°, +3°, and +4°C). Color-coded points represent average temperatures ± SD (smaller than symbol size) as computed from the simulation at steady-state conditions. The dotted horizontal line indicates the middle of the channel, where the reference integer temperature differences were obtained. Sets of ten rows of microwells were consolidated into a total of six bins (indicated by gray horizontal bars) for a coarse characterization of temperature effects. Note that the intrinsic magnification of the PAM camera system only allows for 31 rows of microwells to be measured simultaneously (bins 1 to 3, highlighted in dark gray) and that all data presented are derived from measurements performed on rows 1 to 31, on a total of 1500 microwells. (C) CFD simulation of a single thermal cycle within PhenoChip. A thermal cycle is characterized by the initial rapid elevation of temperatures during the application of an electrical current [ON; see (A)] and the rapid decrease of temperature in the absence of an electrical current (OFF). The resulting temperature variation over time is shown for the discretized microwell bins defined in (B). The PAM measurement interval (90 s) is indicated by blue arrows along the x axis.

  • Fig. 3 PhenoChip enables the measurement of single-cell photophysiological responses to temperature, here demonstrated with two strains of Symbiodinium sp.

    (A and B) Maximum quantum yields (Fv/Fm) under repeated thermal cycling of incrementally increasing temperatures. Results are shown for Symbiodinium strains CCMP421 (A) and CCMP2467 (B). Curves represent average values of Fv/Fm for cells monitored at RT or within bins 1 to 3 during temperature experiments (as defined in Fig. 2B). In all experiments, cells were measured every 90 s using PAM in darkness. Cells in bins 1 to 3 were exposed to four temperature treatments (+1°, +2°, +3°, and +4°C), each consisting of five consecutive thermal cycles (sequentially numbered on colored bars), interspersed with five recovery periods (white bars). Cells were permitted a 28.5-min recovery period (Recovery) that started after the final cycle of the temperature treatment of +3°C, indicated by a blue arrow (Onset) and ended before the start of the next temperature treatment of +4°C, indicated by a black arrow (End).

  • Fig. 4 Single cells of Symbiodinium strains CCMP421 and CCMP2467 differ in their response to elevated temperatures.

    (A) For single cells of CCMP421, the average values of Fv/Fm under each elevated temperatures (+1° to +4°C) were plotted against the average value of Fv/Fm from before temperature exposure. Linear fits of the resulting plots are shown together with their respective R2 values (B) Same, for cells of CCMP2467. (C and D) Histograms of single-cell Fv/Fm values under progressively increasing temperatures in CCMP421 (C) and CCMP2467 (D). Fitting of normal distributions revealed progressively declining values of Fv/Fm in response to increasing temperatures. Insets: As in main plot, for histograms of cells with the highest Fv/Fm values before temperature exposure (top 10% of the population). Note the elevated Fv/Fm in this subpopulation compared to populations in the main panels. (E) Identification of temperature-resilient phenotypes by calculation of a total temperature exposure index, ϴ (see the main text). Single-cell values of Fv/Fm are plotted for each temperature bin, showing the mean (white box) and 1 and 2 SDs for both strains. Asterisks indicate significant differences between the two strains (nonparametric two-sample Kolmogorov-Smirnov tests, ***P < 0.001). Sample sizes (n) are indicated for CCMP2467/CCMP421. (F) Fv/Fm under repeated thermal cycling for selected cells of CCMP2467 from (E) with mean Fv/Fm values more than 2 SDs above (brown) or below (blue) the population mean (black).

  • Fig. 5 Symbiodinium strains CCMP421 and CCMP2467 differ in their photophysiological response to pH.

    Plots show the effective quantum yield, ΦII, at different levels of pH for single cells of Symbiodinium strains CCMP421 and CCMP2467 under saturating light conditions. Box and whisker plots of single-cell ΦII values display the mean (white box) and 1 and 2 SDs. Differences between the two strains at each pH level were assessed using nonparametric two-sample Kolmogorov-Smirnov tests [***P < 0.001, and NS (not significant)]. Sample sizes (n) are indicated above each plot for CCMP421/CCMP2467.

  • Fig. 6 Selecting and propagating single cells of Symbiodinium results in the transgenerational preservation of photosynthetic phenotypes.

    Single cells of Symbiodinium CCMP2467 were first assessed within PhenoChip for their maximum quantum yields, Fv/Fm, and based on this phenotypic assessment, specific photosynthetic phenotypes were ejected from PhenoChip using LCM. In this way, single cells with initial values of Fv/Fm close to the population average (n = 4 cells, blue triangles), and high initial values of Fv/Fm (n = 2 cells, red triangles) were retrieved from PhenoChip. These parent cells were propagated for 6 weeks to yield two distinct populations of cells (blue, average Fv/Fm initial population; red, high Fv/Fm initial population). The resulting populations differ in Fv/Fm (P = 5.2 × 10−8, nonparametric two-sample Kolmogorov-Smirnov test). Sample sizes (n) of single cells within propagated populations are indicated below normal distribution plots, which display the mean (blue or red line) of each population.

Supplementary Materials

  • Supplementary Materials

    PhenoChip: A single-cell phenomic platform for high-throughput photophysiological analyses of microalgae

    Lars Behrendt, M. Mehdi Salek, Erik L. Trampe, Vicente I. Fernandez, Kang Soo Lee, Michael Kühl, Roman Stocker

    Download Supplement

    The PDF file includes:

    • Figs. S1 to S10
    • Table S1
    • Legend for movie S1
    • Supplementary Materials

    Other Supplementary Material for this manuscript includes the following:

    Files in this Data Supplement:

Stay Connected to Science Advances

Navigate This Article