A new haemocyanin in cuttlefish (Sepia officinalis) eggs: sequence analysis and relevance during ontogeny

Abstract

Background: Haemocyanin is the respiratory protein of most of the Mollusca. In cephalopods and gastropods at least two distinct isoforms are differentially expressed. However, their physiological purpose is unknown. For the common cuttlefish Sepia officinalis, three isoforms are known so far, whereas for only two of them the complete mRNA sequences are available. In this study, we sequenced the complete mRNA of the third haemocyanin isoform and measured the relative expression of all three isoforms during embryogenesis to reveal a potential ontogenetic relevance.

Results: The cDNA of isoform 3 clearly correlates to the known Sepia officinalis haemocyanin subunits consisting of eight functional units and an internal duplicated functional unit d. Our molecular phylogenetic analyses reveal the third isoform representing a potentially ancestral haemocyanin isoform, and the analyses of the expression of haemocyanin type 3 reveal that haemocyanin type 3 only can be observed within eggs and during early development. Isoforms 1 and 2 are absent at these stages. After hatching, isoform 3 is downregulated, and isoform 1 and 2 are upregulated.

Conclusions: Our study clearly shows an embryonic relevance of the third isoform, which will be further discussed in the light of the changes in the physiological function of haemocyanin during ontogeny. Taken together with the fact that it could also be the isoform closest related to the common ancestor of cuttlefish haemocyanin, the phylogeny of cuttlefish haemocyanin may be recapitulated during its ontogeny.

Figure 1

Haemocyanin expression during ontogeny. The mRNA expression of the different haemocyanin isoforms (SoH1, SoH2, SoH3) of Sepia officinalis in different tissues (yolk, embryo, gill, branchial gland) and developmental stages ranging from embryos (E1-E4) to hatchlings (H) and adults (A). The histogram shows the mRNA expression as normalised relative quantities with standard error. Thereby, asterisks indicate when the expression of the respective isoform differs significantly compared to the previous developmental stage on a significance level of 0.05 according to the Mann–Whitney-U test. Due to the low number of samples for E1 yolk and J gill, those stages are not taken into account for statistics. The lines illustrate the shift of the haemocyanin isoforms.

Figure 2

Molecular clock and exon-intron structure of cephalopod haemocyanin. Relaxed molecular clock with a mean evolutionary rate of 7.636 × 10^-4 calculated with BEAST v1.5.4 based on a maximum likelihood tree for the amino acid sequences of the complete haemocyanin molecules of Sepia officinalis (SoH1-SoH3), Enteroctopus dofleini (OdHA, OdHG), Nautilus pompilius (NpH) and Haliotis tuberculata (HtH1, HtH2). The split of gastropoda and cephalopoda about 550 ± 50 Mya was used as calibration point. Included are the exon-intron structures of the haemocyanin of N. pompilius, E. dofleini and S. officinalis. Grey bars indicate the 95% HPD (highest posterior density), i.e. the Bayesian confidence interval of the estimated age.

Figure 3

Alignment of the three cuttlefish haemocyanins (SoH1-SoH3). Highlighted are the characteristic six histidine residues forming the oxygen binding sites (green), one thioether bridge, three disulphide bridges and the potential glycosylation sites predicted with NetNGlyc1.0. The secondary structure of the FUg of E. dofleini (Cuff et al., 1998) is displayed on the bottom.

Figure 4

Maximum likelihood tree of the amino acid sequences of haemocyanin. The tree includes the haemocyanin sequences of Sepia officinalis isoform 1–3 (SoH1, SoH2, SoH3), Enteroctopus dofleini A-type (OdHA) and G-type (OdHG), Nautilus pompilius (NpH) and Haliotis tuberculata isoform 1 and 2 (HtH1, HtH2) calculated with PhyML using the LG + I + G model to illustrate the evolution of cephalopod haemocyanin. The tree topology of the Bayesian tree calculated with MrBayes using the WAG + I + G + F model is similar. Therefore, support for the tree nodes is given by the bootstrap values of 1,000 replicates and the consensus support (%).

Figure 5

Maximum likelihood tree of the amino acid sequences of FUd and FUd’. The tree includes the FUd and FUd’ sequences of Sepia officinalis (SoH1-SoH3), Enteroctopus dofleini (OdHA, OdHG) and Nautilus pompilius (NpH) and was calculated with PhyML using the WAG + I + G model to illustrate the duplication of FUd/d’. The tree topology of the Bayesian tree calculated with MrBayes using the WAG + G + F model is similar. Therefore, support for the tree nodes is given by bootstrap values of 1,000 replicates and the consensus support (%).