| First Author | Griffin CS | Year | 1995 |
| Journal | Mouse Genome | Volume | 93 |
| Issue | 3 | Pages | 868-70 |
| Mgi Jnum | J:28909 | Mgi Id | MGI:76447 |
| Citation | Griffin CS, et al. (1995) In Situ hybridisation assigns cDNAs from a normalised human foetal brain library to specific mouse tissues. Mouse Genome 93(3):868-70 |
| abstractText | Full text of Mouse Genome contribution: IN SITU HYBRIDISATION ASSIGNS cDNAs FROM A NORMALISED HUMAN FOETAL BRAIN LIBRARY TO SPECIFIC MOUSE TISSUES. Carol S. Griffin1, David J. Evans2, Jacqueline Attan3, D. Ross Sibson4, Michael P. Starkey5, Tony Vickers4. 1formerly St. Mary's Hospital Medical School, now MRC Radiobiology Unit, Chilton, Didcot, Oxon, OX11 ORD; 2Department of Histopathology, St. Mary's Hospital Medical School, Imperial College of Science, Technology and Medicine, Norfolk Place, London, W2 1PG; 3formerly HGMP RC, now Muscle Cell Biology Group, Clinical Sciences Centre, Royal Postgraduate Medical School, Du Cane Road, London, W12 ONN; 4formerly HGMP RC, now Clatterbridge Cancer Research Trust, J.K. Douglas Laboratories, Clatterbridge Hospital, Bebington, Wirral, L63 4JY; 5UK Human Genome Mapping Project Resource Centre, Hinxton Hall, Hinxton, Cambridgeshire, CB10 1RQ. Introduction Within the human genome mapping project, we are engaged in a systematic program aimed at producing a catalogue of human genes. This is rapidly achieved by partial sequencing of randomly selected cDNAs (1, 2). Comparison of the sequences to the public databases (3) highlights those cDNAs which have similarities to previously known sequences, from which the function of the new sequences can be inferred. In common with other investigators, we find that approximately 60% of new sequences are similar to known sequences and this is in agreement with the number of anciently conserved sequences which are considered to exist (4). In our study, cDNAs were randomly selected as probes to examine patterns of transcription thereby further characterising the genes identified by partial sequencing. The resultant information on the distribution of mRNAs was anticipated to assist in identifying those whose function is restricted to one or a few specialised cell types, in contrast to those having a more widespread function. This was particularly desirable when partial sequencing had failed to suggest a function for the genes identified. Synteny, and the conservation of coding sequences between apparently homologous genes in man and mouse is well established (5, 6), and therefore mouse tissue was selected as the source of target mRNAs in this investigation. This allowed not only rapid fixation of the tissues by perfusion techniques, but also the use of sections of whole late gestation embryos so that tissues are shown on a single slide, in the original anatomical relationship. We show that directly labelled human cDNAs can be used to hybridise specifically to mRNA in a wide variety of mouse tissue. This is convenient and illustrates the feasibility of provisionally elucidating the expression patterns of human cDNAs by in situ hybridisation to mouse tissues. The localisation of mRNAs by their hybridisation in situ to labelled cDNAs, and the analysis of many tissues simultaneously, enables characterisation at a rate similar to the identification of new genes by partial sequencing. Materials and Methods. Probe preparation. cDNAs for in situ hybridisation studies were randomly selected from a non-directional, oligo dT-primed "normalised" human foetal brain library (2), from which approximately 40% of the cDNAs sequenced show no significant similarity to known sequences of either the Genbank (nucleic acids), or Swissprot (proteins) databases. cDNA fragments sized over 300 base pairs and cloned in pBluescript II KS+ were amplified by PCR employing a biotinylated primer in order to generate forward (F) or reverse (R) biotinylated strands. The F and R strands were isolated using streptavidin-coated magnetic beads, and provided a template for the production of labelled ss DNA fragments. cDNA fragments were labelled with [alpha-35S]dCTP, or digoxygenin-11-dUTP (DIG), using random hexamer-primed DNA labelling systems (NEN, Boehringer-Mannheim). T4 DNA polymerase was substituted for the Klenow fragment of DNA polymerase I in order to ensure that only labelled molecules complementary to the template strand were synthesised. Sequencing. Double stranded cDNAs were sequenced by Taq polymerase dideoxy cycle sequencing (7,8), employing a fluorescent primer and a Pharmacia automated laser fluorescent sequencer. CDNA sequences were compared to the nucleic acid and protein sequences held within the Genbank and Swissprot databases using the BLAST database searching program (3). Mouse tissue preparation. Embryos (16 days gestation) were fixed for 4-16h in 4% paraformaldehyde prior to paraffin embedding. Adult tissue was removed from mice perfused with 4% paraformaldehyde, and embedded in paraffin. Serial 5um sections were cut. In situ hybridization. Sections were pretreated as described previously (9) with a few minor modifications. Sections hybridised with 35S-labelled and DIG-labelled probes were pre-incubated in 20mg/ml proteinase K at 20 degrees C for 7.5min and 30min respectively. Sections were hybridised at 37 degrees C for 16h in 50% formamide, 6 x SSC, l x Denhardts, 10% dextran sulphate, 100ug/ml salmon sperm DNA(400ug/ml for DIG-labelled probes), 100ug/ml tRNA, 10mM dithiothreitol (35S-labelled probes only), l x 10(6)cpm 35S-labelled ssDNA/2ug/ml DIG labelled ss cDNA. Sections were washed in 2 x SSC for 2 x 10min at 20 degrees C, followed by 1 x SSC for 4 x 15min at 56 degrees C (S35-labelled probes) or 1 x SSC 4 x 15min (DIG-labelled probes). Non-specific hybridisation was eliminated by treatment with Sl nuclease (20mg/ml for 30min at 42 degrees C). RNA:DNA hybrids were detected by autoradiography or by gold conjugated anti-DIG antibody (diluted 1:30 in blocking buffer), with silver enhancement (10-40min at 20 degrees C), and sections were counterstained in Haematoxylin. Results and Discussion. Random primer labelled cDNA probes of 50-300 bases were used for ISH. Table 1 illustrates the patterns of transcription indicated by exclusively one strand of each of seventeen cDNAs analysed. CDNAs are listed with the name of the database match and the degree of similarity. Knowledge of the cDNA strand hybridising to mRNA is useful for identifying the coding strand, which cannot necessarily be determined by partial sequencing. The only probe (A3F) which failed to detect any mRNA transpired to represent the polylinker of pBluescript, derived from a clone lacking a cDNA insert. In contrast, the ubiquitous elongation factor cDNA B3R, hybridised to each tissue screened. cDNA A10R, 67% similar to lens membrane protein, hybridised to mRNA in kidney and liver. CDNA B11 showed 97% similarity to heterogeneous RNP, and was detected in brain, gut, kidney, liver, and cartilage, but not in heart and lung. The remaining thirteen cDNAs employed as probes showed relatively low, or indeed no significant similarity to known human sequences. The cDNAs used showed no apparent differences in the distribution of signal amongst the various tissues selected, whether the tissue originated from an adult mouse, or a 16 day gestation embryo. The results were not quantified. Although the cDNAs originated from brain, they did not always produce detectable signals in this tissue, yet they could detect mRNA in other tissues. This may reflect the lower abundance of the corresponding mRNAs in brain compared to the tissues where they were detected, or the absence of the particular cells from which the cDNA originated in the sections probed. Legend. Figure 1 shows signal detected on a 16day gestation mouse embryo developing bone (A) and adult mouse brain (C) after hybridization with probe B8. No signal was detected with the complementary strand (B and D respectively).Bar=25um. Localisation of mRNAs in mouse embryonic and adult tissue demonstrates that good resolution can be obtained with both 35S and DIG-labelled ss cDNA probes. Non-isotopic labelled probes have the advantage of speed, probe stability, and improved signal localisation compared with radiolabelled probes, and the use of colloidal gold conjugated anti-DIG antibody amplified by silver precipitation facilitates quantitation together with increased sensitivity. We have shown that cDNAs from a "normalised" human foetal brain library will hybridise not only to specific areas of brain tissue, but also to other mouse tissues. Worldwide human cDNA sequencing programmes have identified a large number of cDNAs which on the basis of sequence similarity analysis, do not match known genes. The analysis of transcription patterns of "unknown" human cDNAs via in situ hybridisation onto mouse tissue sections is a possible approach to adding biological information to essentially cryptic DNA sequences. Using ISH, 10 cDNAs can be localised in one week by one individual, which is approximately the rate at which new genes can be identified using an automated sequencer. Mass production of sections and probes could result in gene identification by ISH to its mRNA at considerably higher rates. Table 1. Localisation of human cDNAs onto mouse tissues. DATABASE COMPARISON. cDNA probe: A2F; Best Genbank match: repetitive sequence; Spp: D; Length: 72; %ID: 65; TISSUE: Brain: ++; Gut: ++; Lung: +; Kidney: ++; Liver: ++; Heart muscle: ++; Cartilage (embryo): ++. cDNA probe: A3F; Best Genbank match: pBluescript II KS+; Length: 122; %ID: 99; TISSUE: Brain: -; Gut: -; Lung: -; Kidney: -; Liver: -; Heart muscle: -; Cartilage (embryo): -. cDNA probe: A4F; Best Genbank match: H2AvD gene for histone H2A; Spp: D; Length: 57; %ID: 66; TISSUE: Brain: +; Gut: +; Lung: -; Kidney: -; Liver: ++; Heart muscle: +; Cartilage (embryo): ++. cDNA probe: A5R; Best Genbank match: Non-receptor tyrosine phosphatase; Spp: H; Length: 30; %ID: 86; TISSUE: Brain: +; Gut: +; Lung: +; Kidney: +; Liver: ++; Heart muscle: -; Cartilage (embryo): +. cDNA probe: A8F; Best Genbank match: fus gene for elongation factor; Spp: S; Length: 36; %ID: 86; TISSUE: Brain: ++; Gut: -; Lung: -; Kidney: +; Liver: +; Heart muscle: -; Cartilage (embryo): -. cDNA probe: A9F; Best Genbank match: fus gene for elongation factor; Spp: S; Length: 36; %ID: 86; TISSUE: Brain: ++; Gut: ++; Lung: -; Kidney: -; Liver: -; Heart muscle: -; Cartilage (embryo): n/d. cDNA probe: A10R; Best Genbank match: Lens membrane protein; Spp: H; Length: 233; %ID: 67; TISSUE: Brain: -; Gut: -; Lung: -; Kidney: +; Liver: ++; Heart muscle: -; Cartilage (embryo): -. cDNA probe: A11F; Best Genbank match: Coagulation factor XI gene, exon 4; Spp: H; Length: 61; %ID: 67; TISSUE: Brain: +; Gut: +; Lung: -; Kidney: ++; Liver: +++; Heart muscle: -; Cartilage (embryo): -. cDNA probe: A12F; Best Genbank match: Ghox-7 homeobox-containing gene; Spp: C; Length: 87; %ID: 63; TISSUE: Brain: -; Gut: -; Lung: +; Kidney: +; Liver: ++; Heart muscle: +; Cartilage (embryo): n/d. cDNA probe: B1F; Best Genbank match: Wegener's granulomatosis autoantigen proteinase gene; Spp: H; Length: 134; %ID: 63; TISSUE: Brain: +; Gut: ++; Lung: -; Kidney: ++; Liver: +++; Heart muscle: +; Cartilage (embryo): n/d. cDNA probe: B2R; Best Genbank match: HepG2 3'-directed MboI cDNA; Spp: H; Length: 364; %ID: 93; TISSUE: Brain: ++; Gut: -; Lung: -; Kidney: -; Liver: -; Heart muscle: -; Cartilage (embryo): -. cDNA probe: B3R; Best Genbank match: Elongation factor 1-alpha; Spp: H; Length: 169; %ID: 88; TISSUE: Brain: +; Gut: +; Lung: +; Kidney: +; Liver: ++; Heart muscle: ++; Cartilage (embryo): n/d. cDNA probe: B5F; Best Genbank match: pst-cathepsin gene; Spp: M; Length: 41; %ID: 75; TISSUE: Brain: +; Gut: ++; Lung: -; Kidney: -; Liver: +++; Heart muscle: -; Cartilage (embryo): n/d. cDNA probe: B8R; Best Genbank match: HGMP putative partial cDNA; Spp: H; Length: 55; %ID: 94; TISSUE: Brain: ++; Gut: +; Lung: +; Kidney: -; Liver: ++; Heart muscle: ++; Cartilage (embryo): ++. cDNA probe: B9F; Best Genbank match: actin 2; Spp: B; Length: 67; %ID: 58; TISSUE: Brain: +; Gut: +; Lung: -; Kidney: -; Liver: -; Heart muscle: -; Cartilage (embryo): ++. cDNA probe: B10R; Best Genbank match: tonin gene; Spp: R; Length: 28; %ID: 89; TISSUE: Brain: ++; Gut: +; Lung: -; Kidney: +; Liver: +; Heart muscle: +; Cartilage (embryo): +. cDNA probe: B11F; Best Genbank match: heterogeneous RNP; Spp: H; Length: 197; %ID: 97; TISSUE: Brain: +; Gut: +; Lung: -; Kidney: +; Liver: ++; Heart muscle: -; Cartilage (embryo): +. The extent of database similarity is denoted by length (base pairs) and percentage "identity" (%ID). Species (Spp.) names are abbreviated by: B, bovine; C, chicken; D, Drosophila; H, Human; M, Slime Mould; R, Rat; S, Sulfolobus. cDNAs A2-B2 were labelled with 35S, and cDNAs B3-B11 were labelled with DIG. Relative signal intensities are represented by - to +++. References (1) Adams, M.D., Kelley, J.M., Gocayne, J.D., Dubnick, M., Polymeropoulos, M.H., Xiao, H., Merril, C.R., Wu, A., Olde, B., Moreno, R.F., Kerlavage, A.R., McCombie, W.R.,Venter, J.C. (1991) Science 252: 1651-1656. (2) Attan, J., Dearlove, A., Howells, D.D., Kelly, M., Parsons, J., Starkey, M.P., Sibson, D.R. (1993) Genome Mapping and Sequencing, Cold Spring Harbor, New York, p204. (3) Altschul, S.F., Gish, W., Miller, W., Myers, E.W., Lipman, D.J. (1990) J. Mol. Biol. 215: 403-410. (4) Green, P., Lipman, D., Hillier, L., Waterston, R., States, D., Claverie, J-M. (1993) Science 259: 1711-1716. (5) Copeland, N.G., Jenkins, N.A., Gilbert, D.J., Eppig, J.T., Maltais, L.J., Miller, J.C., Dietrich, W,F., Weaver, A., Lincoln, S.E., Steen, R.G., Stein, L.D., Nadeau, J.H., Lander, E.S. (1993) Science 262: 57-66. (6) Searle, A.: Humouse database. HGMP RC, Hinxton, Cambridge. (7) Craxton, M. (1991) In Methods: A companion to methods in enzymology 3: 20-. (8) Lee, J.S. (1991) DNA 10: 67-73. (9) Shiels, A., Griffin, C.S. (1993) Curr. Eye. Res.: 12: 913-921. |
