A βPix–Pak2a signaling pathway regulates cerebral vascular stability in zebrafish

Supporting Information

Files in this Data Supplement:

Fig. 5. Blood from widespread hemorrhages in bbh^m292 mutants restricts to brain ventricles with time. Isolectin staining highlights individual blood cells within blood vessels and the heart in wild-type fish at 55 hpf (A), 72 hpf (C and E), or 100 hpf (G and I). In bbh^m292 mutants, blood pools in widespread locations of the brain at 55 hpf (B), and then gradually localizes to the brain ventricles at 77 hpf (D and F), and 100 hpf (H and J). Very little blood remains in normal circulation in the embryo as seen by the lack of blood in the heart.

Fig. 6. Deletion of an exon in bbh^m292 genetic mutants. Sequencing of bbh^m292 mutants reveals the complete deletion of exon 14 (marked in green). A cryptic TGA stop codon which terminates translation when exon14 is deleted is marked in red.

Fig. 7. Splicing defects in morpholino injected embryos. (A) GEF7mphof1 and GEF7 mphor1 were used to amplify bPix around the exon-intron boundary targeted by the bPix^exon6 morpholino. Red arrows denote the wild-type band in 0.2 ng and 8 ng bPix^exon6 MO injected embryos. Black arrows denote mis-spliced new products seen only in the morphants. (B) Primers exon19-mo-f1 and exon19-mo-r1 were used to amplify bPix around the exon-intron boundary targeted by the bPix^exon19 morpholino. 0.85 ng bPix^exon19 MO was injected. (C) Primers Pak2-mo-f and Pak2-mo-r were used to amplify bPix around the exon-intron boundary targeted by the Pak2a^e5i6 morpholino. 2 ng of Pak2a^e5i6 MO was injected. (D) Primers bPix-B^exon1-f and bPix-B^exon2-r amplified the splice junction targeted by bPix-B^exon1-MO. 2 ng of morpholino was injected. In all gels, M is the marker, and Ctrl or TL are uninjected siblings.(A-C) Embryos were harvested at 48-60 hpf. (D) Embryos were harvested at 24 hpf.

Fig. 8. Zebrafish Paks have strong neural expression during development. (A-C) At 24 hpf, Pak1 is highly enriched in the central nervous system including the brain, neural tube and ventral lateral hindbrain. (C) It is also expressed in the caudal notochord. At 42 hpf, Pak2b is enriched in the brain (D), especially in the brain ventricle, retinal cells, and ventral neural tube (H), blood vessels of the trunk (E) and the neural tube epithelium (F). (F) In contrast, at 48 hpf, Pak2a is expressed in the otic vesicle and mesenchymal cells in the head. (G) At 72 hpf Pak6 is highly expressed in the central nervous system. Tel: telencephalon; Die, diencephalon; Teg, tegmentum; Ce, cerebellum; nt, neural tube; nc, notochord; cns, central nervous system; Vz, ventricular zone; ov, otic vesicle; Me, mesenchymal cells; r, retinal cells; Ve, ventricle

Fig. 9. Mosaic expression of bactin-bPixA-myc and lmo2-bPixA-myc. At 54 hpf, the expression of bactin-PixA-myc and lmo2-PixA-myc was detected by anti-myc staining. (A-C) bactin-PixA-myc shows a non-tissue restricted, but mosaic expression pattern in the whole fish (A), head (B) and trunk (C). (D and E) lmo2-PixA-myc shows expression largely restricted to blood vessels of both the head (D) and tail (E). MceV, mid cerebral vein; ISV, intersegmental vessel; A, aorta; V, vein.

Fig. 10. bbh^m292 mutants do not have a defect in thrombin generation. Thrombin activity was tested with an in vitro fibrin-forming assay as described by Jagadeeswaran¹. Three experiments of triplicate samples combining six 48 hpf mutant or wild-type embryos were performed on in vitro extracts after incubation with human fibrinogen according to the method of¹. The optical density of the fibrin clot was measured at 280 nM after resuspension in 8M urea. Control samples had no embryos. The error bars represent standard deviation. 1. Jagadeeswaran, P. & Liu, Y. C. Developmental expression of thrombin in zebrafish embryos: a novel model to study hemostasis. Blood Cells Mol Dis 23, 147-56 (1997).

Fig. 11. Vascular permeability and tight junction formation are normal in bbh^m292. (A and B) Injection of Alexafluor 555-labeled BSA into the circulation of wild type or bbh mutants showed most dye is retained by vessels, but there is also a substantial leakage of dye from blood vessels into tissue, indicating that the blood-brain barrier has not yet formed at 52 hpf in wild type or mutants. At 52 hpf ZO-1 staining is indistinguishable between wild types and bbh mutants in the large vessels of the head. Staining is distributed around the vessel (C and D) or in brain ventricles where the staining is localized to the apical surface near the lumen (E and F), suggesting there are no defects in the formation of tight junctions. P, primordial hind brain channel; l, ventricular lumen; n, neuroepithelium

Table 2. Overexpression of a constitutively active pak2a^{T395E/R189G/P190A} RNA rescues bbh^m292 genetic mutants

Injection of this construct results in a 17.5% rescue with a P value by the paired Student's t test of 0.004.

Table 3. Overexpression of bpix-A RNA rescues bbh^m292 genetic mutants

Injection of this construct results in a 17.7% rescue with a P value by the paired Student's t test of 0.024.

Table 4. Overexpression of a constitutively active pak2a^T395E RNA fails to rescue bbh^m292 genetic mutants

The differences in Hemorrhage rates between control and rescues were not statistically significant when pak2a^T395E RNA was injected.

Table 5. Overexpression of flk-1:bpix-A or flk-1:bpix-B DNA in endothelial cells enhances hemorrhage

200 pg of morpholino was injected into single-cell embryos in the presence or absence of 12 pg of plasmid DNA. Expression of flk-1:bpix-A results in a 41.3% exacerbation of the hemorrhage phenotype (P value of 0.005 by the paired Student's t test). Expression of flk-1:bpix-B results in a 15.2% exacerbation of hemorrhage (p-value of 0.017). All injections were into the wild-type line, TL.

Table 6. Overexpression of flk-1:bpix-A or flk-1:bpix-B DNA in endothelial cells of wild type embryos promotes hemorrhage

Note the TL and TL + DNA samples are not paired. Overexpression of flk-1:?bpix-A plasmid DNA in wild type TL fish results in a 300% increase in hemorrhage over flk-1: GFP alone, while overexpression of flk-1:bpix-B in wild type fish leads to a 63.2% increase in hemorrhage.

Table 7. Overexpression of lmo2:bpix-A in endothelial cells via the Tol2 transposon enhances hemorrhage

200 pg of morpholino was injected into single-cell embryos in the presence or absence of 47.5 pg of Tol2:lmo2:bpix-A DNA and 47.5 pg of tol-2 transposase RNA. Expression of lmo2:bpix-A results in a 23.0 % exacerbation of the hemorrhage phenotype (P value of 0.027 by the paired Student's t test). All injections were into the wild-type line, TL.

Table 8. Overexpression of bactin:bPixA via the Tol2 transposon rescues bPix morphants

400 pg of morpholino was injected into single-cell embryos in the presence or absence of 23.8 pg of Tol2:bactin:bpix-A DNA and 47.5 pg of tol-2 transposase RNA. Expression of bactin:bpix-A in all cells of the embryo results in a 64 % rescue of the hemorrhage phenotype. All injections were into the wild-type line, TL.

Table 9. Primer sequences

Table 10. Morpholino sequences

All morpholinos target splice junctions unless denoted by "ATG," in which case they target the translational start site.

*This morpholino gave strong toxicity and resulted in early embryonic death.

SI Methods

Linkage and Genotyping. Single-embryo DNAs were prepared from homozygous mutants and wild-type siblings in a final volume of 50 ml as described in ref. 1. In a bulked segregation approach to determine linkage, PCR was performed on 0.125 ml of genomic DNA pooled from 24 mutants or wild-types, with 180 SSR primers from the zebrafish genetic map (2). Polymorhpisms were analyzed on either 6% Urea-polyacrylamide gels, or on 3% Metaphor agarose (Cambrex).

Initial linkage was found to Z21548 on chromosome 1, and there were no recombinants in the 1590 meioses. The two closest markers tested on either side of Z21548 were Z48722 with 15 recombinants in 1,594 meioses and Z9704, with 19 recombinants in 1,594 meioses. We designed two new polymorphic markers to define the left and right sides of the interval- ctg13117-2 with 5 recombinants in 1,508 meioses and ctg 13117-3 with 2 recombinants in 1,578 meioses.

Identifying Mutations in bpix in bbh^m292 Genetic Mutants. Total RNA was prepared from 25-100 55 hpf embryos from an m292 homozygous mutant or homozygous wild-type cross using the RNeasy kit. PCR products were cleaned using the Qiagen PCR purification kit and eluted in water for direct sequencing. Mutations were identified based on differences compared to the wild type m292 cDNA. We identified exon14 as missing from mutant bpix cDNA, and proceeded to examine the mutation in genomic DNA. Genomic DNA from homozygous mutant or homozygous wild type embryos were used for PCR, and the mutation at the splice site identified with the primers GEF7ex12-f and GEF7ex12-r. Five mutant and five wild-type sequences were examined.

bbh^fn40 Quantitative PCR Methods. At 3 dpf, embryos from bbh^fn40a heterozygous or bbh^fn40a mutant x bbh^fn40a heterozygous crosses were collected and phenotyped. Five biological replicates were prepared for mutant and wild type using pools of 12 embryos. Total RNA was prepared with the Qiagen RNA Easy kit and 1 mg of RNA was used to make cDNA using the iScript RT kit (BioRad). Quantitative PCR was performed on the ABI 7500 (Applied Biosystems) using the standard Taqman reagent mixture (Applied Biosystems). bPix (Arhgef7) mRNA levels were compared to endogenous gapdh levels. Each of five biological replicate samples were performed in triplicate.

arhgef7 primers: Forward 5'-CTCGGTGTGTTCCCGTCAAT-3'

Reverse 5'-ACAGTTGAGCGAGTCGAAAGAC-3'

Reporter 5'-TCCGCAGAGAAGACG-3'

gapdh Primers Forward 5'-GATAACTTTGTCATCGTTGAAGGTCTT-3'

Reverse 5'-CGGTCTTCTGTGTTGCTGTGA-3'

Reporter 5'-TGAGCACTGTTCATGC -3'

Morpholino Injections. Morpholinos were diluted in 200 mM potassium chloride in concentrations ranging from 50 mM-1 mM and microinjected into1-4 cell stage zebrafish embryos. Volumes were calculated by the use of a graticle.

5' and 3' RACE. Total RNA was isolated from »1,600 52-55 hpf wild type embryos using the Qiagen RNeasy midi-prep kit. To isolate poly(A)⁺ mRNA, total RNA was purified with the Qiagen Oligotex mRNA purification kit. 1 mg of poly(A)⁺ mRNA was then used in the Marathon cDNA kit from BD CLONTECH to make an adaptor-ligated cDNA-library. 5' and 3' RACE products were generated as recommended in the kit instructions using BD Advantage 2 Polymerase mix using internal primers GEF7race-1f and 1r. These products were then A-tailed Amersham Pharmacia Taq polymerase before cloning into pCRII-TOPO. Resultant white colonies were fingerprinted by restriction digest with BamHI, HindIII or NotI for the different possible inserts. Several of each representative pattern were then sequenced. The consensus sequences from RACE were derived from between 2 to 6 different independent clones for each sequence.

In situ Hybridization Probes. GEF7-B8 is a 0.8 kb BamHI fragment of bpix, cloned into pBS SK+. This probe should highlight expression of all splice variants. An exon specific probe for exon 1 of bpix-A was synthesized by PCR using the primers bPix-A^exon1-f and bPix-A^exon1-r-T7. An exon specific probe for exon 1 of bpix-B was synthesized by PCR using the primers bPix-B^exon1-f and bPix-B^exon1-r-T7 An exon specific probe for exon19 was made using the primers GEFaltex-f2 and GEFaltex-T7 (171 bp).

Zebrafish pak sequences were identified from GenBank and the Zebrafish Information Network. pak 1 (CF269264) pak2a (BU670714), pak2b (AL927138), and pak6 (CD753164) clones were obtained from Open Biosystems or the Zebrafish International Resource Centre (Oregon, USA).

DNA and RNA Constructs. Full-length bpix was cloned using primers bPix-A^EcoRI or bPix-A^EcoRI with bPix-A^HA, and cloned into the Eco-Not sites of pflk-Topo (3). To synthesize RNA, the gene was subcloned into pCRII, and transcribed in vitro using Mmessage Machine (Ambion) from T7. Tol2 transposon (4) bactin:bPix-A and lmo2:bPix-A were cloned using the Tol2kit series of plasmids (Kristen Kwan, Clemens Grabher, and Chi-Bin Chien, personal communication). bPix-A was cloned into pDONR221 using PixA-f-attb1 and PixA-r-attb2 primers, while promotors were cloned into pDONRp4p1r, and a myc tag added from Tol2kit p3E-MTpA. All three fragments were cloned into the vector pDestTol2pA2, by Multisite Gateway cloning (Invitrogen, CA).

1. Shimoda, N. et al. (1999) Genomics 58:219-32.

2. Michelmore R, Paran I, Kesseli RV. (1991) Proc Natl Acad Sci USA 88:9828-32.

3. Jin SW, Beis D, Mitchell T, Chen JN, Stainier DY (2005) Development (Cambridge, UK) 132:5199-209.

4. Kawakami K, Shima A, Kawakami N (2000) Proc Natl Acad Sci USA 97:11403-8.