Smith et al. 10.1073/pnas.0508952102. |
Fig. 3. Gene Ontology (GO) pie charts of differentially expressed genes between donor cells and NT embryos. (A) Genes up-regulated n = 795 in NT embryos that are categorized by GO term "biological process" level 3. (C) Genes up-regulated in NT embryos that are categorized by "molecular function" level 2. (B and D) Genes up-regulated (n = 751) in donor cells representing "biological process" level 3 and "molecular function" level 2 categories of interest.
Supporting Text
Microarray Design and Annotation.
Development of the 7,872 cattle cDNA microarray was done at University of Illinois, Urbana-Champaign, and is described in detail by Everts et al. (1). Briefly, the 7,872 cDNA array consisted of the original 3800 cDNA microarray (2) and was supplemented with sequences selected from normalized and subtracted placenta and spleen cDNA libraries. The cDNA inserts were unidirectionally cloned and were sequenced from the 5' end, trimmed, filtered, and clustered. A unique high-quality insert set was selected and amplified by PCR by using M13-forward (GTTTTCCCAGTCACGACGTTG) and M13-reverse (TGAGCGGATAACAATTTCACACAG) oligonucleotide primers. After purification, PCR products along with positive, negative, and exogenous spiking controls were redissolved in 3 × SSC and 1.5 M betaine and spotted onto Telechem Superamine slides (Arrayit, TeleChem, Sunnyvale, CA) in duplicate by using the GeneMachines OmniGrid 100 arrayer (Genomic Solutions, Ann Arbor, MI). After spotting, the slides were kept in a dust free environment overnight and then baked for 1.5 h at 80°C.For annotation of the sequences, blastn similarity searches were performed against the human UniGene (build 166), mouse UniGene (build 135), and the human genome (build 34). blastn against the human UniGene, resulted in 6,089 hits with expectation value E <e-5. Another 810 hits were added after blastn and tblastx against the mouse and human DNA sequence databases. In summary, 6,899 (87.4%) cattle sequences could be annotated by using the human and mouse databases, and 5,325 of these have unique UniGene identifiers or positions on the human genome. Additionally, there are 973 sequences that represent putative novel cattle genes, highly divergent homologs or genes with long divergent 3' UTRs. In total, the 7,872 cattle microarray potentially contains 6,298 unique sequences.
Generation of Nuclear Transfer (NT), in Vitro Fertilization (IVF), and Artificial Insemination (AI) Embryos.
IVF and nuclear transfer was based on Kubota et al. (3, 4). In this study, cultured skin fibroblast cells from an adult Holstein cow were used for nuclear transfer. The cells were cultured to passage 8 and 9 (DMEM + 10% FBS) and then serum-starved (0.5% FBS) for 3 days after reaching confluency. Before nuclear transfer, the donor cells were trypsinized, washed by centrifugation, and resuspended in 5% FBS + DMEM. Abattoir oocytes were used in this study for the production of IVF and NT embryos. After 20-22 h of successful in vitro maturation, the oocytes were enucleated. This was confirmed by Hoechst 33342 staining. Using our standard procedure, donor cells with a diameter of 10-15 mm were transferred to the perivitelline space of the recipient cytoplast. Fusion was induced with two pulses of direct current of 2.5 kV/cm for 10 msec each, with an Electorcell Manipulator 200 (BTX, San Diego), and was confirmed by microscopic examination. Oocyte activation was simultaneously induced with the electrical pulses. Further activation of fused embryos was achieved by culture with cyclohexamide (CHX, 10 mg/ml; Sigma), cytochalasin D (CD, 2.5 mg/ml; Sigma), in M199 + 7.5% FBS for 1 h and then CHX (10 mg/ml; Sigma) in M199 + 7.5% FBS for an additional 4 h. For the first 48 h, the NT embryos were cultured in CR1aa medium supplemented with 6 mg/ml BSA at 39°C in a humidified atmosphere of 5% CO2, 5% O2, and 90% N2. Cleaved embryos were further cultured for 5 days in CR1aa medium with 10% FBS and cumulus cell coculture. Blastocyst development was recorded on day 7 and IETS grade 1 (C1) embryos were vitrified. Fifteen NT embryos (35% blastocyst rate) from two different NT batches generated from the same fibroblast cell line were used for microarray analysis. For in vitro fertilization (3), frozen-thawed sperm were washed twice by centrifugation in modified Brackett-Oliphant (m-BO) medium supplemented with 10 mM caffeine and 4 mg/ml BSA. The sperm were resuspended in fertilization medium that contained m-BO medium with 10 mg/ml heparin and 4 mg/ml BSA. Final sperm concentration was adjusted to 1 × 106/ml, and 20 in vitro matured oocytes were added to 100 ml of fertilization-medium droplets. The sperm and oocytes were incubated for 6-8 h, and then the oocytes were washed and cultured in CR1aa medium with 6 mg/ml BSA for 48 h. The cleaved embryos were freed of their cumulus cells by pipetting and then cultured to the blastocyst stage (day 7) in CR1aa with 10% FBS and cumulus cell coculture. Fifteen C1-quality blastocysts (37% blastocyst rate) were vitrified and thawed according to a modified previously described vitrification protocol (5). Briefly, blastocysts were placed in Dulbecco’s phosphate-buffered saline (DPBS) with 6% BSA for 3 min and then equilibrated in 25% of VS3a solution (1.6 M glycerol with 6% BSA in DPBS) for 20 min at room temperature. Then the blastocysts were rinsed for 30 sec in 65% VS3a (4.2 M glycerol/6% BSA in DPBS) and transferred into a column of 100% VS3a in 0.25-ml straw. After heat sealing and incubation at room temperature for 1 min, the straw was held in liquid nitrogen (LN2) vapor (-150°C to -180°C) for 3 min and then plunged directly into liquid N2. For warming of the vitrified embryos, the frozen straw was held for 10 sec in the air, placed in a 22°C water bath for 10 sec. and then shaken vigorously to mix the contents. The recovered warmed blastocysts were transferred into 0.5 M sucrose and then 0.25 M sucrose for 2 min each. Then the blastocysts were washed in PBS for 2 min three times and transferred individually to microcentrifuge tubes, snap-frozen on dry ice, and placed in a -80°C freezer to await RNA isolation, linear amplification, and microarray analysis. The vitrified IVF and NT embryo survival rate was essentially 95–100% based on our previous experience (unpublished data). Additionally, standard cryopreserved AI embryos from two donor cows/sires, 11 and 3 embryos, respectively, were used as in vivo controls at day 7 after superovulation (standard Folltropin-V regime, Bioniche, Belleville, ON, Canada) and artificial insemination. Only Holstein embryos were used.Linear Amplification and Validation.
The linear amplification procedure was done as described by Baugh et al. (6) for 2 ng of total RNA. Briefly, RNA was isolated from individual day 7 blastocysts using TRIzol reagent (Invitrogen) and linear acrylamide coprecipitant (Ambion, Austin, TX). RT was carried out in a 2-ml reaction with 20 ng of T7dT21 primer [GCATTAGCGGCCGCGAAATTAATACGACTCACTATAGGGAGA(T)21V (V = A, C, G or T)], SuperScript III 200 units/ml (Invitrogen) for the first round at 42°C for 1 h. The second and third rounds were done in a 10-ml RT reaction with 0.5 mg of random primers (mostly hexamers). Second-strand synthesis (SSS) was carried out in 15 ml for the first round and 75 ml with the addition of 100 ng of T7dT21 primer for the second and third rounds at 16°C for 2 h. Two units of Escherichia coli T4 polymerase (New England Biolabs) for 15 min at 16°C was used to blunt end the double-stranded cDNA after the first round, and 10 units was used for the second and third rounds. The cDNA was purified with phenol chloroform/IAA (Ambion) in phase-lock tubes (Brinkmann) and then put through a Micro Bio P-6 purification column (BioRad). Precipitation was done with 0.5 volumes of 7.5 M ammonium acetate, 5 mg of linear acrylamide (for second and third rounds), and 2.5 volumes of 100% ethanol. The in vitro transcription (IVT) reaction was done overnight (»15 h) at 42°C [Epicentre Ampliscribe kit, RNasin and HC T7 RNA Polymerase (Promega)] in 10 ml for the first round and 20 ml for the subsequent rounds. After each round, amplified RNA (aRNA) was purified by using Qiagen’s (Valencia, CA) RNeasy Kit, and 40 units of RNasin was added. After the completion of the third round, the aRNA was quantified by using OD260 absorbance measurement. From a single blastocyst, we were able to generate on average 51 mg of aRNA after three rounds of amplification. Amplified RNA was stored at -80°C in 4 mg of aliquots until use on the microarray.Adult skin fibroblast cells from the same donor animal were thawed (frozen at passage 5, DMEM + 10%/DMSO + 20% FBS), cultured to passage 7 (DMEM + 10% FBS), serum-starved for 3 days (0.5% FBS), washed, and distributed in 1,500 cell aliquots. For RNA isolation, 1 ml of TRIzol was added to the 1500 cell pellet. After homogenization, the TRIzol was then separated into 10 100-ml aliquots (»150 cells each). RNA isolation and amplification were carried out exactly as performed for the embryos.
For the kidney validation experiment of the linear amplification protocol, total RNA was isolated by using TRIzol and then frozen at -80°C until use. One microgram of kidney total RNA was used for the first replicate, and 5 mg was used for a second replicate. Two samples from replicate 1 were pooled after the first round to yield »5 mg. One hundred nanograms from each replicate were amplified in rounds two and three. The transcript length ranged from »100 base pairs to >1 kb. One microgram of labeled cDNA from each round was hybridized to the microarray. Ten micrograms of kidney total RNA was used as the unamplified comparison. In total, 16 microarrays were analyzed, because both replicates at each round were labeled with Cy3 and Cy5 fluorescent dye (dye swap). Additional rounds of amplification can reduce 5' complexity because of random priming during cDNA synthesis. However, this trend has little impact on the sensitivity and specificity of the hybridizations because of the 3'-EST design of the 7,872 cDNA microarray. In addition, transcript length of aRNA was comparable between the second and third rounds of amplification (data not shown).
Labeling and Microarray Hybridizations.
One microgram of aRNA was reverse transcribed, labeled, and hybridized to each microarray. In total, 96 microarrays, all embryos and nuclear donor cells with dye-swap, were analyzed. All of the embryos were hybridized versus a standard reference. The standard reference RNA was comprised of total RNA isolated from BL3°, MDBK, and EBTR cell lines and brain tissue from an Angus heifer. More than 85% of the genes on the microarray are expressed in the standard reference. This reference design compares embryo profiles based on the expression of each gene in the embryo versus the standard reference.The aminoallyl labeling and hybridization protocol used were based on the protocol developed by The Institute for Genomic Research (TIGR, ref. 7). One microgram of aRNA was reverse-transcribed by endogenous random priming. For the reaction, 3 ml of exogenous spiking control RNA [soy genes: Rubisco (1,666 pg/ml), AB binding (167 pg/ml), and major latex protein homologue (MSG) (16.7 pg/ml)] were added to the RNA, and the volume was brought up to 17.9 ml and heated at 70°C for 5 min. After snap cooling on ice, 12.1 ml of reverse transcription premix [6 ml, 5´ first-strand buffer/3 ml·1 M DTT/0.6 ml 50´ aminoallyl/dNTP mix (2:3 aa-dUTP:dTTP)/2 ml of SuperScript III (Invitrogen)/ 200 units/ml/0.5 ml of RNasin (Promega) 40 units/ml] was added to each sample. Reverse transcription was carried out at 42°C for »14 h. After cleaning using a PCR purification kit (Qiagen; with homemade phosphate wash and elution buffers), 4.5 ml of the Cy dyes (Amersham Pharmacia Biosciences) rehydrated in DMSO was added to samples in 4.5 ml of 0.1 M Na2CO3 buffer (pH 9.0) and incubated in the dark for 60 min. After a subsequent cleanup, labeling efficiency was calculated, and the labeled probes were dried and rehydrated in 40 ml consisting of H2O, 20 mg of bovine COT1 DNA (Bovine Hybloc, Applied GeneticsFreeport, NY), 20 mg of T7dT24, and 40 mg of dT30. Each sample was combined with the standard reference cDNA that had been labeled with the opposite Cy dye. The probes were denatured for 2-3 min at 95°C, 10 min at 60°C, and 10 min at room temperature immediately before hybridization to the microarray. The microarrays were denatured by boiling, rinsed in ethanol, dried, and then placed in prehybridization buffer (20% formamide/5´ Denhardts/6 × SSC/0.1% SDS/0.05% pyrophosphate/25 mg/ml yeast tRNA) at 42°C for 1 h. After prehybridization, the microarrays were dipped in H2O five times, in isopropanol twice, and then dried by centrifugation. For hybridization, the probe was mixed with an equal volume of 2´ hybridization buffer at 42°C (50% formamide/10 × SSC/0.2% SDS). The microarray was positioned in a hybridization chamber and placed in a dark 42°C water bath for »40 h. Each microarray was washed three times for 4 min each (wash 1, 42°C 1 × SSC/0.1% SDS; wash 2, 25°C 0.2 × SSC/0.5% SDS; wash 3, 25°C 0.05 × SSC) and scanned (Axon genepix 4000B, Axon Instruments, Union City, CA). Scanning normalization of dye intensities was accomplished by using spiking controls and also through the use of genepix pro 4.0 scanning software (Axon Instruments). Each scanned image was examined thoroughly, and dust particles and spots with high background were flagged and removed from analysis. The background and standard deviation were calculated for each raw data file, and only those spots with intensities three standard deviations above background were considered "expressed" and loaded into genespring 6.1 (Agilent Technologies, Palo Alto, CA). Loess normalization was applied to all microarrays before statistical comparison of samples (8). Genes that were present in either the standard reference or sample on 90% of the microarrays were used for further analysis.
Data Analysis.
Both genespring 6.1 and our independent analysis using sas statistical (SAS Institute, Cary, NC) software used one-way ANOVA to determine gene expression differences. There was a very high degree of overlap of the results using the two methods. A two-stage modeling approach as implemented by Wolfinger et al. (9) was used. The first-stage model fitted to the log transformed and normalized ratio of embryo (or kidney, donor cell) and standard reference samples across all genes included the effects of dye, array, and array by dye interaction. The residuals were analyzed by gene in a second-stage model that included the fixed effects of dye and condition and the random effect of array. The heterogeneity of variance model provided a significantly better fit than the homogeneity of variance model for a large number of genes, and thus the heterogeneity of variance model was used for all of the genes across conditions. Three sas models were evaluated to identify the genes differentially expressed across conditions: (i) condition was the round of amplification (0, 1, 2, and 3); (ii) condition was either AI, NT, or IVF embryos and, (iii) condition was either cloned embryos or donor cells. Probability values (P values) were adjusted for multiple comparisons using the false-discovery rate approach. In addition to these statistical criteria, those genes that differed ³2-fold were considered differentially expressed. The correlation coefficients were derived from the hierarchical clusters made by genespring 6.1. The software uses a Pearson correlation »0 to determine similarity using all genes across the samples in the comparison. It randomly selects a sample and then picks the most similar sample to the original one selected and averages their data. The next most similar sample is chosen and so on, until all of the samples have been assigned to clusters, and the difference within and between the clusters is calculated.Quantitative Real-Time RT-PCR.
The sequences of the 13 selected genes that were used on the microarray were repeat-masked (repeatmasker, University of Washington, Seattle) and loaded into the primer express software (Applied Biosystems) for primer and probe design. Primers and 5' FAM-3' TAMRA-labeled probes were synthesized by Sigma-Genosys. Amplified RNA from eight randomly selected NT embryos, eight AI embryos (except for two genes, DNAJC10 and HRH1, where 14 embryos were analyzed), and donor cells was used. RT reactions were carried out as performed for the microarray, except for the use aminoallyl dNTP mix and spiking controls. Five nanograms of cDNA from each sample in triplicate was used for the real-time PCR (Applied Biosystems Prism SDS 7000). For the 25-ml PCR reaction, 1´ Taqman Universal Master Mix, 0.9 mM forward and reverse primers, and 0.25 mM probe was added to 5 ml (1 ng/ml) of sample cDNA. The relative standard curve method was used for quantification (Applied Biosystems Prism SDS 7700 User Bulletin #2). The standard curve was comprised of pooled aRNA from each sample tested. The same standard reference aRNA used in the microarray analysis was used as the calibrator sample. After relative quantities were determined, the ratio of all samples, and the standard reference aRNA was calculated, and the mean for each group was determined and compared for an overall fold change.1. Everts, R. E., Band, M. R., Liu, Z. L., Kumar, C. G., Liu, L., Loor, J. J., Oliveira, R. & Lewin, H. A. (2005) Vet. Immunol. Immunopathol. 105, 235-245.
2. Band, M. R., Olmstead, C., Everts, R. E., Liu, Z. L. & Lewin, H. A. (2002) Anim. Biotechnol. 13, 163-172.
3. Kubota, C., Yang, X., Dinnyes, A., Todoroki, J., Yamakuchi, H., Mizoshita, K., Inohae, S. & Tabara, N. (1998) Mol. Reprod. Dev. 51, 281-286.
4. Kubota, C., Yamakuchi, H., Todoroki, J., Mizoshita, K., Tabara, N., Barber, M. & Yang, X. (2000) Proc. Natl. Acad. Sci. USA 97, 990-995.
5. Nedambale, T. L., Dinnyes, A., Yang, X. & Tian, X. C. (2004) Biol. Reprod. 71, 1671-1676.
6. Baugh, L. R., Hill, A. A., Brown, E. L. & Hunter, C. P. (2001) Nucleic Acids Res. 29, E29.
7. Hegda, P., Qi, R., Abernathy, R., Gay, C., Dharap, S., Gaspard, R, Earle-Hughes, J., Snesrud, E., Lee, N. J. & Quackenbush, J. (2000) BioTechniques 29, 548-562.
8. Clevel, W. S. & Devlin, S.J. (1988) J. Am. Stat. Assoc. 83, 596-610.
9. Wolfinger, R. D., Gibson, G., Wolfinger, E. D., Bennett, L., Hamadeh, H., Bushel, P., Afshari, C. & Paules, R. S. (2001) J. Comput. Biol. 8, 625-637.