Ancient cytokines, the role of astakines as hematopoietic growth factors

Abstract

Hematopoiesis is the process by which hemocytes mature and subsequently enter the circulation. Vertebrate prokineticins (PKs) are known to take part in this process, as are the invertebrate prokineticin domain proteins, astakines. In Pacifastacus leniusculus, astakine 1 is essential for the release of new hemocytes into the open circulatory system of these animals. In addition to astakine 1, we have now cloned a homologue of astakine 1 with an insert of 13 amino acids, named as astakine 2. Both crustacean astakines lack the N-terminal AVIT motif, which is present in vertebrate PKs, and hence receptor binding differs from that of vertebrate PKs. We have found astakine-like sequences in 19 different invertebrate species, and the sequences show that some motifs are conserved among invertebrate groups. Previously we showed that astakine 1 is directly involved in hematopoiesis, and now we show that astakine 1 and astakine 2 have different roles in hemocyte lineage differentiation. Astakine 1 can stimulate proliferation of hematopoietic tissue (Hpt) cells (precursor of hemocytes) as well as specifically induce differentiation of Hpt cells along the semigranular cell lineage, whereas astakine 2 plays a role in granular cell differentiation. Moreover, we discuss the impact of the putative structures of different astakines in comparison with the vertebrate prokineticins.

FIGURE 1.

Two different astakines are present in P. leniusculus. A new astakine in crayfish (pl-astakine 2) is more similar to the previously cloned astakine (pm-astakine 2) from shrimp; both of them have an insert of 13 amino acid residues (boxed) when compared with crayfish astakine (pl-astakine 1). The conserved GX₂RYSX(P/R)XC motif and LXYP motif are underlined.

FIGURE 2.

Astakine 1 and astakine 2 share a similar tissue expression pattern. Semiquantitative RT-PCR was used for this expression analysis. Lane 1, muscle; lane 2, hepatopancreas; lane 3, Hpt; lane 4, hemocytes; lane 5, heart; lane 6, intestine; lane 7, nerve cord; lane 8, testis.

FIGURE 3.

Astakine 1 stimulates proliferation and differentiation into the semigranular cell lineage of Hpt cells both in vitro and in vivo. A, Hpt cell cultures were supplemented with r-astakine 1 or r-astakine 2 every second day. After 1 week, RT-PCR was conducted to monitor the expression of PCNA (proliferation marker of Hpt cells), KPI (gene marker for semigranular cells), and SOD (gene marker for granular cells). The addition of astakine 1 could stimulate transcription of PCNA and KPI significantly. The effect of r-astakine 1 and r-astakine 2 is detectable from ∼3 nm (4.5 ng/well) and is optimal at a concentration of ∼65 nm (∼100 ng/well), whereas a toxic effect to the cells is obtained at ≥120 nm. B, r-astakine 1 mainly stimulates the synthesis of semigranular cells in vivo. Relative hemocyte counts for total hemocytes, semigranular cells, or granular cells were calculated as hemocyte number 24 h postinjection/initial hemocyte number. r-astakine 1 injection induced a significant increase in the number of total hemocytes, mainly due to an increase of semigranular cells. The columns represent the mean of three separate experiments; the error bars indicate S.D. values.

FIGURE 4.

Astakine 2 promotes maturation of granular cells in vivo. A, r-astakine 2 injection slightly stimulated the total number of hemocytes in vivo. Relative hemocyte counts were calculated as the total number of hemocytes after injection of r-astakine 1 or r-astakine 2/initial total number of hemocytes. Crayfish saline (CFS) was used for control injection. B, r-astakine 2 injection stimulated the maturation of granular cells in vivo. The relative ratio of granular cells was calculated as the ratio of granular cells after 24 h post-treatment/initial ratio of granular cells. r-astakine 2 can increase the ratio of granular cells in vivo by stimulating the maturation of granular cells, whereas the mutated r-astakine 2^Mut showed no such activity. The columns represent the mean of six separate experiments, and error bars represent S.D. values. *, t test, p < 0.01.

FIGURE 5.

Astakine 1 and astakine 2 move differently in SDS-PAGE. A, recombinant astakine 1 shows an oligomer and a monomer, but recombinant astakine 2 only runs as a monomer. Lane 1, r-astakine 1 (Trx-S tag-astakine 1) in reducing conditions (molecular mass is about 30 kDa); lane 2, r-askakine1 (Trx-S tag-astakine 1) in non-reducing conditions (molecular mass is about 370 kDa, which was measured by Gel filtration); lane 3, monomer of r-astakine 1 (Trx-S tag-astakine 1) in reducing conditions (purified by gel filtration); lane 4, monomer of r-astakine 1 (Trx-S tag-astakine 1) in non-reducing conditions (purified by gel filtration); lane 5, r-astakine 2 (Trx-S tag-astakine 2) in reducing conditions; lane 6, r-astakine 2 (Trx-S tag-astakine 2) in non-reducing conditions. B, oligomer formation of r-astakine 1 is independent of the fusion tag. The oligomer form of r-astakine 1 was treated with the protease thrombin to remove the fusion tag Trx. Lane 1, r-astakine 1 without Trx tag in reducing conditions; lane 2, r-astakine 1 without Trx tag in non-reducing conditions; lane 3, free Trx in reducing conditions; lane 4, free Trx in non-reducing conditions. C, r-astakine 1 expressed in the insect cell-baculovirus expression system similarly forms oligomers. Lysates of Sf9 cells infected with recombinant baculovirus containing the ORF of astakine 1 were used for Western blot using astakine 1 antibody. D, in order to separate the oligomer and monomer form of recombinant fusion astakine 1, HiPrep 16/60 Sephacryl S-200 HR column (GE Healthcare) gel filtration was performed. Apoferritin (443 kDa), alcohol dehydrogenase (150 kDa), and BSA (66 kDa) were used as standard proteins to calibrate the column.

FIGURE 6.

Homology modeling of crayfish P. leniusculus astakine 1 and surface charge prediction of astakine 1 and astakine 2, compared with the template of black mamba intestinal toxin (MIT1). A, model of MIT1 (Protein Data Bank entry 1IMT) showing the predicted five cysteine bridges. B, homology model of astakine 1, predicted cysteine bridges, and free cysteines are shown. C, surface charge calculation of 1IMT. The N terminus is indicated. D, surface charge prediction of P. leniusculus astakine 1. The N terminus is indicated. E, surface charge prediction of P. leniusculus astakine 2. The N terminus is indicated, and the conserved tyrosine-proline residues in the astakine 2 loop are indicated by an arrow.

FIGURE 7.

Mutation of two free cysteines in astakine 1 to glycines does not affect its properties. A, purified recombinant astakine 1 and mutant of astakine 1 were separated by 12.5% SDS-PAGE. Lane 1, r-astakine 1 (Trx-S tag-astakine 1) in reducing conditions; lane 2, r-astakine 1 (Trx-S tag-astakine 1) in non-reducing conditions; lane 3, mutant of r-astakine 1 (Trx-S tag-astakine 1^Mut) in reducing conditions; lane 4, mutant of r-astakine 1 (Trx-S tag-astakine 1^Mut) in non-reducing conditions. B, effects of r-astakine 1, and mutation of r-astakine 1 and r-astakine 2 on extracellular ATP synthesis by Hpt cells. 3 × 10⁵ Hpt cells were seeded to tissue culture dishes, and each dish of Hpt cells was first incubated with reaction buffer, and the formation of extracellular ATP was assayed and used as the individual control value for each dish. Then the cells were washed with Hepes buffer, incubated with 0.2 μm recombinant protein for 60 min, and subsequently incubated with reaction buffer. The formation of ATP was calculated, and the remaining ATP formation was calculated by comparison with its individual control value. The columns represent the mean of three separate experiments, and error bars represent S.D. values.

FIGURE 8.

Phylogenetic analysis of invertebrate astakine and vertebrate PKs. Phylogenetic analysis of the astakine sequences in Table 1 was performed using ClustalW2 (see the EMBL-EBI Web site), and the phylogenetic unrooted tree was constructed using the Phylip 3.69 package (27).

FIGURE 9.

Different roles of astakines as hematopoietic growth factors. The hypothetical model shows five types of Hpt cells and two main lineages of hemocytes: SGC and GC (adapted from Refs. and 26). Astakine 1 and astakine 2 have different roles in hemocyte lineage differentiation. Astakine 1 can stimulate proliferation of Hpt cells as well as specifically induce differentiation of Hpt cells along the semigranular cell lineage, whereas astakine 2 plays a role in granular cell differentiation.