Materials and methods

Exome analysis of case-parent trios and single exomes in individuals with cardiovascular laterality defects

DNA isolation and exome sequencing of affected case-parent trios (HET15 and HET17) was carried out consenting to the IRB ethical approval and authorized by the local ethics committee of the medical faculty of the University of Bonn (Lfd Nr. 141/15), as previously published for other laterality defects associated disease genes by Breuer et al. in 202211. Since family members of the index case might present with clinically unapparent features of the laterality defect spectrum, all family members included received a thorough clinical exam and ultrasound studies by two paediatric cardiologists in order to assess the affection status of all study participants11. Furthermore, all patients included were assessed for extracardiac manifestations. In the patients described here, no extracardiac manifestations were found. Single exome data was retrieved from an earlier study by Sifrim et al. 201628. Filtering of exome data was performed as published earlier by Breuer et al. in 202211.

Zebrafish husbandry and Embryo preparation

Under provided husbandry §11-license of the Zebrafish Core Facility in Bonn, zf are maintained according to national law and to recommendations by Westerfield29. The adult zf were kept in a 28°C temperature 3.5-L tank in a water-recirculating rack-system on an adequate 14-h light and 10-h dark cycle. Zfl of wt AB/TL, Tg(kdrl:eGFP;gata); s844Tg (ZDB-ALT-051214–6) and Tg(kdrl:eGFP;gata ds:red) fam222bb hu10539; fam222ba hu10755; s843Tg (ZDB-FISH-161130–1, s843Tg) TALEN derived strain were obtained by natural spawning and raised in an incubator at 28°C in 30% Danieau’s buffer longest until independent ingestion on day five post fertilisation. For imaging and WISH-analysis performed at ≤ 5 days post fertilization pigmentation was suppressed by adding 0.003% phenylthiourea (PTU) to the Danieau solution.

Fam222ba/bb dd-KO

TALEN- derived Tg(kdrl:eGFP;gata ds:red) fam222bb hu10539; fam222ba hu10755; s843Tg (ZDB-FISH-161130–1) strain obtained from B.C. Kirchmaier, Goethe University Frankfurt, and firstly described by S. Spiegler, University of Greifswald, were spawned for our experiments. TALE nuclease injection and non-homologous end-joining led to a 10-base pair deletion in the fam222ba gene (fam222ba hu10755) 5 ‘_-_TTGTCCATGG-3 ‘ and in the fam222bb gene (fam222bb hu10573) 5 ‘-CGGACGGGCC-3 ‘, respectively, resulting in a frameshift with a premature stop in the protein coding sequence. These deletions were confirmed by resequencing during the time of experiments.

Sequencing and Genotyping of fam222ba/bb dd-KO line

For sequencing and genotyping of the received Tg(kdrl:eGFP;gata ds:red) fam222bb hu10539; fam222ba hu10755; s843Tg (ZDB-FISH-161130–1) TALEN derived strain, PCR was performed on skin-swabbing30 derived genomic DNA with HOT FIREPol® Blend Master Mix Ready to Load (Solis Biodyne, Cat. Num. 04–25-02,025). For sequencing, the PCR-product was sent to Eurofins Genomics, Bonn. Primer sequences are listed in supplemental material, primer.

Whole-mount zf in situ hybridization

For preparation of sense and antisense probes of fam222ba, fam222bb, fam222aa and myl7, cDNA plasmids were generated by RT-PCR amplification and then cloned into the SK(-) pBluescript® vector. As a PCR template, cDNA of 56 hpf old wt AB/TL zfl was used. For probe preparation of fam222ba, fam222bb, fam222aa and myl7 digoxigenin-labelling was performed using the DIG RNA Labeling Kit (Cat. No. 11 175 025 910) from Roche. Double staining with an additional fluorescein-labelled adprhl1 hard specific probe was performed using the same Kit but instead of DIG RNA Labeling Mix (Cat. No. 11 277 073 910) Fluorescein RNA Labeling Mix (Cat. No. 11 685 619 910) from Roche was used. High resolution in situ hybridization on whole-mount zf embryos was established according to Thisse, C.& Thisse, B.31. Double-staining was performed according to Moens, C.32 https://research.fredhutch.org/content/dam/stripe/moens/other/RNA_in_situ_protocol.pdf.

Prior to experimental procedure, the zfl for WISH analysis were treated with 1-phenyl 2-thiourea (PTU) Danieau solution in order to prevent natural pigmentation. Images were taken using the Nikon Eclipse Ni-U Microscope with DS-Fi2 Camera and the NIS-Elements BR 5.20.00 64-bit software. For general expression analysis of the zf ‘ fam222b paralogues and fam222aa, zfl were positioned laterally (Fig. 2A). For imaging analysis of myl7 cardiac looping (Fig. 2B) the zfl tails were cut off, zfl were placed on the stump in a drop of glycerol and imaged from frontal towards cranial.

Heart dissection of adult zf

For heart preparation, approximately two-year-old zf were euthanized by hypothermic ice shock and fixed in a lateral position left side up on a dissection Styrofoam plate. The skin was cut from the pelvic fin along the belly of the zf to the operculum. After a 90-degree cut towards the upper edge of the zf eye, the skin and subcutaneous fat was unfolded to expose the heart. The dissected heart was then washed in PBS-buffer and imaged using the Zeiss Stemi 508 ZOOM Stereomicroscope. All area measurements were performed using Fiji ImageJ version 1.53c, Java 1.8.0_172, win64-bit (National Institutes of Health, USA, Wayne Rasband) measuring software.

Haematoxylin–eosin (HE) staining

The dissected hearts were fixed in 4% paraformaldehyde in PBS-buffer; pH 7.4 and then dehydrated. After this, they were embedded in paraffin_,_ cut in ~ 12 µm sections and stained with HE. For HE-staining an intern protocol was followed.

Masson’s trichrome stain

The dissected hearts were fixed in 4% paraformaldehyde in PBS-buffer; pH 7.4 and then dehydrated. After this, they were embedded in paraffin_,_ cut in ~ 12 µm sections and stained using Masson’s trichrome stain. Staining was performed according to a modified in-house protocol for zf.

Transmission electron microscopy (TEM)

For TEM of the adult zf heart, samples underwent fixation with a solution based on cacodylate buffer containing 1.5% Glutaraldehyde, 1% Paraformaldehyde and 3% Sucrose. The samples were then rinsed in phosphate buffer and subsequently post-fixed for 1 h at room temperature in aqueous 2% osmium tetroxide in phosphate buffer. A graded ethanol series for dehydration was followed by a transition to propylene oxide and an embedment in Spurr epoxid resin. Slices of 1 µm and 50 nm thickness were cut with a diamond knife on a Reichert-Jung Ultracut E Microtome 70 17 74 and mounted on formvar coated single slot grids. Contrasting with aqueous solution of 0.5% uranyl acetate and 3% lead citrate was realised prior to examination in transmission electron microscope (Jeol JEM-1400 Plus).

Structure modelling

Complex structures of human FAM222B (UniProt accession code Q8WU58; 562 aa) bound to various proposed interaction partners were modelled with AlphaFold323, either as heterodimers or hetero-tetramers (dimer of dimers) using full-length sequences. Best results were achieved for the 1:1 stoichiometric complex with the human Ser/Thr-kinase NLK (UniProt Q9UBE8), where R300 of FAM222B was proposed to mediate intensive interactions with NLK. Addition of ATP and two magnesium ions led to the calculation of a kinase active model, where Ser296 of FAM222B is facing the catalytic centre. Protein graphics were generated using the PyMOL Molecular Graphics System (Version 2.5.5 Schrödinger, LLC).

Mutagenesis and mRNA Synthesis of human FAM222B

For overexpression and mutagenesis experiments, a cDNA IMAGE clone 4,040,385 (Art. Num. IRAUp969A0655D) of human FAM222B cloned into pOTB7 Vector was obtained from Source BioScience.

Mutagenesis of human variants into wt FAM222B was performed using an Agilent Technologies QuikChange Lightning Site-Directed Mutagenesis Kit (Cat. Num. 210,519–5; for Variant 1/2) and a New England Biolabs Q5® Site-Directed Mutagenesis Kit (Cat. Num. E0552S; for Variants 3–7). Kit associated protocols were used.

After linearization of respective cDNA plasmid, mRNA transcription and poly(A) tailing were performed using the mMessage mMACHINE® Kit (SP6; Cat. Num. AM1340) and the Poly(A) Tailing Kit (Cat. Num. AM1350) by Invitrogen (Thermofisher). The kit-attached protocols were followed.

Microinjections of mRNA

In one to two-cell Tg(kdrl:eGFP;gata) transgenic wt embryos, overexpression of human FAM222B mRNA was performed by injecting 0.3 ng (2 nl at 150 ng/ul) of the wt mRNA into the embryos ‘ yolk sac. We prioritized the FAM222B variant 1/2 (rs753661690, c.899G > A, p.Arg300His) because of its independent appearance in two non-related families and its characteristics in protein structure modelling. Poly(A) mRNA of rs753661690 was injected in two different concentrations: 75 ng/µl (0.15 ng mRNA) and 150 ng/µl (0.3 ng mRNA). As a control, un-injected and embryos sham injected with a phenol red, water mix were compared. For pressure injection through mechanical-manipulator positioned glass-capillaries the “Milli-Pulse Pressure Injector, Model MPPI-3” (Applied Scientific Instrumentation, Inc. 29,391 W. Enid Rd. Eugene, OR 97,402–9533, United States) was used.

In vivo fluorescent zfl imaging

Phenotype evaluation after mRNA injection was performed at 48 hpf using the Zeiss AXIO Zoom.V16 microscope equipped with ZEN 2.6 pro software. For statistical analysis, we performed a dead zf quantification at 24 and 48 hpf as well as a phenotype observation on remaining zfl at 48 hpf. The cardiovascular system was analysed using the green fluorescence reporter-signal of the Tg(kdrl:eGFP;gata) transgenic strain.

Statistical analysis

Statistical analysis was performed using GraphPad Prism version 6.01 for Windows. The alpha level was set to α = 0.05. Differences with a p-value of < 0.05* are mentioned as statistically significant and p-value < 0.01** as very significant within the main text. Data distribution, statistical test forms and exact p-values of each test are given in the according figure legend. For binary, nominally scaled data two-sided Chi2-test without Yates’ correction (expected frequency > 5) or two-sided Fisher’s-exact-test (expected frequencies < 5) were used, for ratio scaled data unpaired two-tailed t-test was used. Normality of ratio scaled data was verified for all groups using D’Agostino-Pearson (n > 25) or Shapiro–Wilk-test (n < 25). Nominally scaled data error bars were created according to individual SEₚ.

Regulatory and guideline compliance statement

We do confirm that all methods were performed in accordance with the relevant guidelines and national regulations.