A novel nanoluciferase transgenic reporter to measure proteinuria in zebrafish

The zebrafish is an important animal system for modelling human diseases. This includes kidney dysfunction as the embryonic kidney (pronephros) shares considerable molecular and morphological homology with the human nephron. A key clinical indicator of kidney disease is proteinuria, but a high-throughput readout of proteinuria in the zebrafish is lacking. We used the Tol2 transposon system to generate a transgenic zebrafish line that uses the fabp10a liver-specific promoter to over-express a nanoluciferase molecule fused with the D3 domain of Receptor-Associated-Protein (which we term NL-D3). Using a luminometer, we quantified proteinuria in NL-D3 zebrafish larvae by measuring the intensity of luminescence in the embryo medium. In the healthy state, NL-D3 is not excreted, but when embryos were treated with chemicals that affected either proximal tubular reabsorption (cisplatin, gentamicin) or glomerular filtration (angiotensin II, Hanks Balanced Salt Solution, Bovine Serum Albumin), NL-D3 is detected in fish medium. Similarly, depletion of several gene products associated with kidney disease (nphs1, nphs2, lrp2a, ocrl, col4a3, and col4a4) also induced NL-D3 proteinuria. Treating col4a4 depleted zebrafish larvae (a model of Alport syndrome) with captopril reduced proteinuria in this system. Our findings confirm the use of the NL-D3 transgenic zebrafish as a robust and quantifiable proteinuria reporter. Given the feasibility of high-throughput assays in zebrafish, this novel reporter will permit screening for drugs that ameliorate proteinuria, thereby prioritising candidates for further translational studies.


Zebrafish husbandry
Zebrafish were maintained and staged according to established protocols 13 and in accordance with the project licenses of Martin Lowe (70/9091) and Rachel Lennon (P1AE9A736) under the current guidelines of the UK Animals Act 1986. Embryos were collected from group-wise matings of wild-type AB Notts, or y-crystallin:mcherry/fabp10a:NL-D3 fish.

Molecular Cloning
The cDNA for rat RAP was a gift from Prof. Zebrafish transgenesis 5 µl of 80 ng/µl DNA of the transgenesis construct were mixed with 5 µl of 50 ng/µl tol2 transposase mRNA and 1 µl of 10X phenol red. All solutions were prepared fresh immediately before injection. Each single cell stage embryo was oriented to allow access into the cell and 1 nl of this solution was injected directly into the cell using a microinjector. Successful transgenesis was confirmed in F0 embryos by selecting for mCherry expression in the lens at day 5, and founder fish were then grown to adulthood.

Preparation of recombinant NL-D3
His-tagged NL-D3 was expressed in E. coli BL21(De3) Codon Plus cells by induction with 1 mM IPTG for 16 hours at 18˚C. The protein was purified using Ni-NTA agarose using standard methods, snap frozen in liquid nitrogen, and stored at -80˚C until use. for assessment of luciferase activity.
Embryos were then placed in an injection chamber containing MS-222 and injected into the common cardinal vein with a microinjection needle. Embryos were then transferred to fresh E3 embryo medium to recover and grown to desired stage. For incubations, embryos were transferred with a transfer pipette to fresh E3 embryo medium containing the drug at the desired concentration. Control embryos were treated with the equivalent concentration of vehicle substance (DMSO).
For gentamicin and cisplatin treatments, previous methods described by Chen et al, (2020) were used. In brief, gentamicin (Sigma, #G1264) was re-suspended in a 0.9% NaCl solution to a concentration of 10 ng/µl. This was diluted to 6 ng/µl for injection, which was performed at 48 hpf into the common cardinal vein (either 0.5 nl or 1 nl injection volume). For cisplatin treatments, the drug was re-suspended directly into E3 embryo medium containing 0.01% methylene blue. Concentrations of 0.5 mM and 1.5 mM were exposed to embryos at 48 hpf for a period of four hours. Embryos were then washed three times in E3 medium and left to grow to the desired stage for proteinuria analysis.

Morpholino oligonucleotide treatments
For lrp2a

CRISPR-Cas9 knockdowns
To genetically knockdown expression of gene products, we adopted the CRISPR-Cas9 approach described by 19  Multi-mode microplate reader and luminescence was measured by reading each well for 1 second. Data was collected in relative luminescence units (RLU) using SoftMax  Pro software.
Using RLU values of each serial dilution minus the blank, a standard curve of known concentrations of recombinant NL-D3 relative to RLU was established. Conversion of RLUs to NL-D3 amount is not fundamental to infer changes in proteinuria in the reporter, but the empirical standard curve data can be provided upon request, with which users can interpolate their RLU readings in GraphPad Prism and generate NL-D3 ng/ml readings.

Zebrafish proteinuria reporter assay
The pipeline of the assay for proteinuria in NL-D3 zebrafish is shown in the schematics accompanying Figures 2B and 6B. Embryos were grown to 4 dpf, then three embryos per well were placed in one well of a 96-well dish ( Figure 2B). E3 embryo medium was removed and replaced with 200 µl fresh E3. 24 hours later 50 µl of E3 medium was removed from each well and placed in the corresponding well of a fresh opaque 96-well plate. 50 µl of substrate from the NanoGlo® Luciferase Assay System (Promega #N1110) was then added to each well.
Plates were then briefly spun down at 700 rpm for 1 minute and then immediately assayed for luminescence on a Flexstation 3 multi-mode microplate reader. Softmax Pro 5.4 software (Molecular Devices) was used to detect luminescence. Endpoint analysis was selected with a 1000 millisecond integration of luminescence (RLUs). Opaque 96-well option was selected in 'Assay Plate', as well as Costar 96 opaque 3mL in 'Compound Source'.

RT-PCR
For RT-PCR, cDNA was diluted to 1 ng/µl and mixed with 7.8 µM of primers and added in equal volume to 2X Power SYBR TM Green PCR Master Mix (Thermofisher #4367659). The primers used are described below; The PCR reaction was run on a Bio-Rad CFX96 Touch Real-Time PCR machine. All analyses of the data used the ∆∆Ct method of quantification. RT-PCRs were run in triplicate with -RT and -cDNA controls.

CRISPR mutation analysis
Embryos injected with gRNAs were grown to 24 hpf, dechorionated, then ten randomly selected embryos were placed in a 1.5 ml capped tube. All excess E3 embryo medium was removed and replaced with 50 l of 50 M NaOH. Embryos were then placed at 95˚C in a heating block for 15 minutes. After a brief spin, 5 l of Tris-HCl pH 8 was added to neutralize the samples. The genomic DNA preps were then homogenized by pipetting before 1 l was used in a PCR to amplify the region around the gRNA cut sight using the primers shown in the primer list below. For TIDE analysis, the PCR product was purified, ran on a gel to check size, and then diluted to 5 ng/l before being sequenced.

In situ hybridisation
Whole-mount in situ hybridisation on zebrafish embryos was performed as previously described 21 . Digoxigenin-labelled anti-sense riboprobes were made using T3 RNA polymerase transcription kits (Roche Diagnostics). Probe templates for zebrafish col4a3, col4a4 and col4a5 were generated by PCR amplification from cDNA taken from 4 dpf zebrafish embryos. The primers used were as follows: A T3 anchor sequence (GGATCCAATTAACCCTCACTAAAGGG) on the 5' end of the reverse primer was used to enable T3-mediated RNA synthesis from the purified PCR product.
Following colorimetric assay, the embryos were treated with 100% Methanol for 10 minutes

Wax sectioning and staining
For liver histological analysis, 3-month-old adult zebrafish were humanely killed by schedule 1 method and livers dissected. Isolated tissue was placed in 4% PFA overnight at 4˚C. Tissue infiltration was performed in a tissue processor and then embedded and mounted with paraffin wax. 5 m sections were cut on a Leica RM2255 microtome. Sections were adhered to glass slides and dewaxed and stained with hematoxylin and eosin. Stained slides were coverslipped and imaged on a 3D Histec Pannoramic250 slide scanner.

Transmission electron microscopy
Samples were prepared according to protocols described previously 22 . Images were taken on T12 Biotwin transmission electron microscope. Distances were measured in Fiji/ImageJ, measurements for foot process width were taken and normalized to the length of the glomerular basement membrane. The total number of GBM width measurements for each sample was: control crispant, n=36; col4a3 crispant, n=32; col4a4 crispant, n=37; col4a5 crispant, n=30. For foot process/ GBM length measurements, five TEM images per treatment were used and the total GBM length (nm) was measured. The number of foot processes along this GBM length was counted, then this number was divided by the GBM length. The mean ± SEM was calculated, and a student's t-test was performed using GraphPad Prism version 8.4.3 for 120 Windows, GraphPad Software, San Diego, California 121 USA, www.graphpad.com

Statistical analysis
The mean ±SD was calculated using GraphPad Prism version 9 for Windows, GraphPad Software, San Diego, California USA, www.graphpad.com. Statistical significance scores were measured in GraphPad Prism version 9 using unpaired parametric Student's t test. P values in 95% confidence limits were characterised as significant, R squared scores from this statistical analysis was also noted to determine the size of the difference between the two compared datasets.