Cascading effects of prey identity on gene expression in a kleptoplastidic ciliate

Abstract

Kleptoplastidic, or chloroplast stealing, lineages transiently retain functional photosynthetic machinery from algal prey. This machinery, and its photosynthetic outputs, must be integrated into the host's metabolism, but the details of this integration are poorly understood. Here, we study this metabolic integration in the ciliate Mesodinium chamaeleon, a coastal marine species capable of retaining chloroplasts from at least six distinct genera of cryptophyte algae. To assess the effects of feeding history on ciliate physiology and gene expression, we acclimated M. chamaeleon to four different types of prey and contrasted well-fed and starved treatments. Consistent with previous physiological work on the ciliate, we found that starved ciliates had lower chlorophyll content, photosynthetic rates, and growth rates than their well-fed counterparts. However, ciliate gene expression mirrored prey phylogenetic relationships rather than physiological status, suggesting that, even as M. chamaeleon cells were starved of prey, their overarching regulatory systems remained tuned to the prey type to which they had been acclimated. Collectively, our results indicate a surprising degree of prey-specific host transcriptional adjustments, implying varied integration of prey metabolic potential into many aspects of ciliate physiology.

Keywords: Mesodinium chamaeleon; acquired metabolism; cryptophyte; photophysiology; transcriptomics.

FIGURE 1

Growth and photophysiology of M. chamaeleon. Bargraphs show results for starved (light bars) and fed (dark bars) M. chamaeleon, error bars represent ± 1 standard error. Points with error bars represent the same measurements for free‐living prey. (A) M. chamaeleon grew fastest when well‐fed, and prey with red plastid types produced the fastest growth rates (comparable to those of prey). Only R. salina‐acclimated M. chamaeleon maintained positive growth rates after 1 week of starvation. (B) Well‐fed cells had comparable photosynthetic efficiencies to those of their prey, but photosynthetic efficiency declined with starvation. (C) Only M. chamaeleon bearing red plastids achieved similar maximum photosynthetic rates (per chlorophyll‐a) to prey cells, but maximum photosynthetic rates decreased with starvation for all prey types except H. pacifica. (D) M. chamaeleon with red plastids also had elevated chlorophyll‐a per cell, but only when well‐fed. Generally, pigment content declined with starvation, except for cells acclimated to C. mesostigmatica. Photosynthetic rates scaled to cellular carbon (E) and cell (F) reflected pigment content, with the highest rates found in well‐fed, red‐plastid‐bearing M. chamaeleon, and declines with starvation across all prey types. Note that prey chlorophyll content (D) and per‐cell photosynthetic rates (F) are 10‐fold lower (left‐hand y‐axis) than M. Chamaeleon.

FIGURE 2

Relative proportion of Cryptophyte prey reads (top portion of bars) to Mesodinium chamaeleon reads (bottom portion of bars) for both well‐fed (darker color) and starved treatments (lighter color). Cryptophyte species fed to M. chamaeleon are on the x‐axis.

FIGURE 3

Pairwise comparisons of M. chameleon gene expression by prey species fed for both fed and starved cultures with M. chameleon cultures fed R. salina grouped by treatment as the control. The X‐axis for all graphs in the Log₂ fold differential expression compared to R. salina fed cultures. The Y‐axis is numbers of genes, color coded by species fed for the well‐fed treatment and gray for the starved treatment. First row: Histograms of all annotated differentially expressed genes (>2 fold with p‐value <0.01). Second row: Differentially expressed genes involved in carbohydrate metabolism. Third row: Differentially expressed genes involved in cofactor and vitamin metabolism. Fourth row: Differentially expressed genes involved in transcription. Fifth row: Differentially expressed genes involved in cellular signaling and processes.

FIGURE 4

Correspondence between cryptophyte phylogenetic and physiological divergence and variation in M. chamaeleon gene expression. Panel a. left: Variance stabilizing transformation of M. chamaeleon RNA expression. Each biological replicate is shown and color coded by cryptophyte species fed and treatment, solid lines indicate well‐fed, dashed lines starved. Right: Multigene maximum likelihood tree of the four cryptophyte prey species used in this study. Showing small genetic distance between the two red plastid bearing species (R. salina and S. major), a larger difference between the two green plastid bearing species (C. mesostigmatica and H. pacifica), and the largest difference between type of plastid (red and green). Panel b. left: Principal component analysis of cryptophyte physiology usingmultiple measures of physiology, including growth, photosynthetic efficiency, photosynthetic pigments and rate, and cellular carbon and nitrogen content. Cryptophytes cluster by species. Center: Across the same measures of physiology, well‐fed M. chamaeleon cells (filled symbols) were different from starved M. chamaeleon cells (hollow symbols). Further, cells with plastids from red cryptophytes (R. salina and S. major) performed differently from those with blue‐green plastids (from C. mesostigmatica and H. pacifica). Right: Principal component analysis of M. chamaeleon gene expression mirrors dendrogram in panel a. points are color coded by prey species fed (CM‐green, HP‐blue, RS‐coral, SM‐red) light shading indicates starved, solid well‐fed.