Neofunctionalization of Glycolytic Enzymes: An Evolutionary Route to Plant Parasitism in the Oomycete Phytophthora nicotianae

Abstract

Oomycetes, of the genus Phytophthora, comprise of some of the most devastating plant pathogens. Parasitism of Phytophthora results from evolution from an autotrophic ancestor and adaptation to a wide range of environments, involving metabolic adaptation. Sequence mining showed that Phytophthora spp. display an unusual repertoire of glycolytic enzymes, made of multigene families and enzyme replacements. To investigate the impact of these gene duplications on the biology of Phytophthora and, eventually, identify novel functions associated to gene expansion, we focused our study on the first glycolytic step on P. nicotianae, a broad host range pathogen. We reveal that this step is committed by a set of three glucokinase types that differ by their structure, enzymatic properties, and evolutionary histories. In addition, they are expressed differentially during the P. nicotianae life cycle, including plant infection. Last, we show that there is a strong association between the expression of a glucokinase member in planta and extent of plant infection. Together, these results suggest that metabolic adaptation is a component of the processes underlying evolution of parasitism in Phytophthora, which may possibly involve the neofunctionalization of metabolic enzymes.

Keywords: Phytophthora; adaptation; evolution; gene duplication; glucokinases; oomycetes; subfunctionalization; virulence.

Figure 1

Phylogenetic tree depicting relationships among glucokinases from oomycetes. The tree was generated using MEGA and the maximum likelihood method. Sequences from brown algae and diatoms are indicated in blue. Oomycete sequences from Saprolegniales are presented in beige. Numbers at the nodes indicate bootstrap replicate values. Abbreviations are Ach.: Achlya; Al.: Albugo; Aph.: Aphanomyces; Br.: Bremia; E.: Ectocarpus; F.: Fistulifera; Gl.: Globisporangium; Hy.: Hyaloperonospora; Na.: Nannochloropsis; P.: Phytophthora; Pe.: Peronospora; Ph.: Phaeodactylum; Pl.: Plasmopara; Py.: Pythium; Sa.: Saprolegnia; and Thr.: Thraustotheca.

Figure 2

Intraspecific structuration of GKI and GKII sequences from P. nicotianae, as depicted by phylogenetic analyses. (A) Phylogenetic relationships among protein GKI sequences were established by maximum likelihood, using the Tamura-3-parameter substitution model, incorporating an among-site rate variation, approximated by a discrete gamma distribution (five rates). Branches were supported by 100 bootstrap replicates. Branch and label colors reflect the origin of sequences: green: tobacco; red: tomato; yellow: citrus; and blue: ornamentals. Tobacco isolates from China are identified by a green frame. (B) Phylogenetic relationships among GKII sequences were established by maximum likelihood, using the Kimura-2-parameter substitution model with a gamma (five rates) distribution. Branch and label colors are similar to those used for Figure 2A.

Figure 3

Secondary and 3D structures of P. nicotianae GKs. (A) Sequence- and structure-based alignment of P. nicotianae glucokinases. The alignment was generated by MUSCLE [41] and ESPript 3 [54], based on the predicted 2D structure of PPTG_18934 (GKI) and PPTG_18927 (GKII). Helices (α, η), sheets (β), and turns (TT) in the potential structure of GKI and GKII are labeled. (B) Superposition of I-TASSER predicted 3D model structures of P. nicotianae GKs (in blue) over their best analog templates (in red). Superposed are cartoons of GKI and E. coli glucokinase (PDB: 1sz2A, left), GKII and Streptomyces glucokinase (PDB: 3vpzA, middle), and GKIII and N. fowleri glucokinase (PDB: 6da0, right). Represented are cartoons of predicted 3D models of GKs, after structural refinement and energy minimization with ModRefiner. Pairwise structure alignments were performed using FATCAT, and images were rendered using ChimeraX.

Figure 4

Unrooted maximum likelihood phylogeny of glucokinases. The tree was inferred with 57 protein sequences, retrieved from blastp searches, against the NR database at Genbank, using the four P. nicotianae GK proteins as queries. Sequences from oomycetes are presented in purple and other stramenopiles in blue. Bacteria (among which cyanobacteria) are represented in green, red algae in red, fungi in brown, and metazoan in pink. Less represented groups are described in the text. The tree was generated using MEGA and the maximum likelihood method. The numbers on branches indicate the results of the SH-like approximate likelihood ratio tests (aLRT). Branch support values larger than 0.7 were shown.

Figure 5

Expression patterns of GKs in pre-infection structures of P. nicotianae. (A) Values relative to each GK class were summed, normalized, and expressed, relative to the expression of the constitutive UBC gene [62]. (B) Expression levels of GKI, GKII, and GKIII in motile zoospores (Z), cysts (C), and germinating cysts (GC). Statistical analysis was performed in Graph Pad Prism v9.2 (GraphPad Software, San Diego, CA, USA), using one-way ANOVA and a Tukey’s multiple comparison tests. Statistical significance is denoted * p < 0.05, ** p < 0.01, **** p < 0.0001. ns: not significant.

Figure 6

Expression of GK genes during plant infection. Roots of tomato and tobacco plantlets were inoculated with P. nicotianae zoospores, and expression of GK genes was evaluated relative to the expression of the constitutive UBC gene. Presented are expression values from samples after 2- (A), 4- (B), and 6-days (C) post inoculation. (D). Values from each kinetic experiment were pooled. Statistical analyses were conducted using a two-way ANOVA Friedman test. Statistical analysis was performed in Graph Pad Prism v9.2 (GraphPad Software, San Diego, CA, USA), using a two-way ANOVA Friedman test. Statistical significance is denoted * p < 0.05, ** p < 0.01, **** p < 0.0001.

Figure 7

Relative expression of GK genes during leaf infection. Leaves from three tomato varieties were inoculated with zoospores of four P. nicotianae strains, and expression of GK genes was evaluated three-days post inoculation, relative to the expression of the constitutive UBC gene. (A) General view of expression values, as a function of tomato variety. (B) Representation of individual gene expression, for a better scale. SP: Saint Pierre; Ma: Marmande; MM: moneymaker. Statistical analysis was performed in Graph Pad Prism v9.2 (GraphPad Software, San Diego, CA, USA), using a two-way ANOVA Friedman test. Statistical significance is denoted ** p < 0.01.

Figure 8

Relative expression of GK genes during leaf infection. Data correspond to the experiment, depicted in Figure 7, but are presented as a function of each P. nicotianae strain. Statistical analysis was performed in Graph Pad Prism v9.2 (GraphPad Software, San Diego, CA, USA), using a Kruskall-Wallis test. Statistical significance is denoted * p < 0.05.

Figure 9

Expression of GKII is linked to P. nicotianae pathogenicity. (A) Comparison of the ability of the P. nicotianae strain to invade tomato leaves. Data refer to the same experiment than those analyzed in Figure 7 and Figure 8. Extent of infection was evaluated by the expression of a tomato gene, relative to the constitutive UBC gene (left), and measurement of invaded leaf surface, expressed in cm² (right). Statistical analysis was performed in Graph Pad Prism v9.2 (GraphPad Software, San Diego, CA, USA), using a Kruskall-Wallis test. Statistical significance is denoted * p < 0.05, *** p < 0.001, **** p < 0.0001. (B) Correlation between the level of GK expression, the level of tomato gene expression, and extent of leaf invasion. Correlation values are represented by blue (positive) and red (negative) circles. The size of the circles reflects the numerical values. Analysis was conducted using a Kendal’s tau test using Past [64].