REPRODUCTIONRESEARCH

Transcriptional signatures throughout development: the effects of mouse embryo manipulation in vitro

Sky K Feuer1, Xiaowei Liu1, Annemarie Donjacour1, Rhodel Simbulan1, Emin Maltepe2 and Paolo Rinaudo1

1Department of Obstetrics, Gynecology and Reproductive Sciences and 2Department of Pediatrics, University of California at San Francisco, San Francisco, California, USA Correspondence should be addressed to P Rinaudo; Email: [email protected]

Abstract

Stressful environmental exposures incurred early in development can affect postnatal metabolic health and susceptibility to non- communicable diseases in adulthood, although the molecular mechanisms by which this occurs have yet to be elucidated. Here, we use a mouse model to investigate how assorted in vitro exposures restricted exclusively to the preimplantation period affect transcription both acutely in embryos and long term in subsequent offspring adult tissues, to determine if reliable transcriptional markers of in vitro stress are present at specific developmental time points and throughout development. Eachin vitro fertilization or embryo culture environment led to a specific and unique blastocyst transcriptional profile, but we identified a common 18-gene and 9-pathway signature of preimplantation embryo manipulation that was present in all in vitro embryos irrespective of culture condition or method of fertilization. This fingerprint did not persist throughout development, and there was no clear transcriptional cohesion between adult IVF offspring tissues or compared to their preceding embryos, indicating a tissue-specific impact ofin vitro stress on gene expression. However, the transcriptional changes present in each IVF tissue were targeted by the same upstream transcriptional regulators, which provide insight as to how acute transcriptional responses to stressful environmental exposures might be preserved throughout development to influence adult gene expression. Reproduction (2017) 153 107–122

Introduction suggests that common mechanisms may be involved in both sensing and transducing the environment stimuli, There is robust epidemiological and animal evidence leading to a reprogrammed epigenetic, transcriptional that exposure to different environmental conditions and/or metabolic state that could potentiate an during early development affects postnatal growth, increased susceptibility for metabolic disease metabolism and disease susceptibility in adulthood. throughout postnatal life. As a result, understanding As dictated by the Developmental Origins of Health the mechanisms by which the developing embryo and and Disease (DOHaD) hypothesis, poor or suboptimal fetus assimilate environmental input into a programmed developmental experiences—including nutritional, metabolic response is relevant across multiple DOHaD oxidative or in vitro stress—can predispose chronic fields. diseases, including hypertension and components of the Appropriate to this discussion is the evidence that metabolic syndrome (Gillman 2005). preimplantation development is an important period Arguably the most pressing question in the DOHaD of environmental sensitivity and that stresses incurred field is identifying how the molecular changes within this window can affect glucose metabolism, -cell occurring in a developing organism secondary to β function, blood pressure and fat deposition in adulthood adverse environmental or nutritional exposures alter (Kwong et al. 2000, Fernandez-Gonzalez et al. 2004, the growth and metabolic trajectories across the life Watkins et al. 2010, Rexhaj et al. 2013, Donjacour course. Interestingly, many models of DOHaD exhibit et al. 2014). The preimplantation period is a particularly common long-term outcomes, including glucose advantageous model for investigating DOHaD-related intolerance, hypertension and vascular dysfunction, questions because exposures are limited to a time frame irrespective of both the form of developmental stress of 4–5 days and embryos exhibit cellular uniformity and the timing of exposure (Simmons et al. 2001, (only 2 cell types are formed after 4 days in culture: the Jungheim et al. 2010, Watkins et al. 2011). This

© 2017 Society for Reproduction and Fertility DOI: 10.1530/REP-16-0473 ISSN 1470–1626 (paper) 1741–7899 (online) Online version via www.reproduction-online.org Downloaded from Bioscientifica.com at 09/26/2021 04:56:31AM via free access

10.1530/REP-16-0473 108 S Feuer and others inner cell mass and trophectoderm)—thereby providing Rinaudo et al. 2006, Giritharan et al. 2007, 2010, an opportunity to connect the precise variations in 2012, Feuer et al. 2014). These experiments have been exposure to outcome. In addition, embryo manipulation conducted by several scientists in multiple locations, holds wide clinical significance due to the routine use using various microarray platforms and analysis software. of assisted reproductive technologies (ART) such as Further, there have been significant advancements in in vitro fertilization (IVF), which has resulted in the birth our knowledge of the genome over the past decade. of over 5 million children as of 2012 (ICMART 2012). This manuscript provides a systematic analysis and It is also clear that increasing the severity of a particular integration of the gene expression changes present in stressor leads to progressively worse phenotypes in (1) blastocysts exposed to assorted culture conditions adulthood. For example, mouse IVF and embryo culture and methods of fertilization and (2) selected adult performed under clinically optimized conditions affect tissues from offspring generated by in vitro fertilization. the expression of nearly 300 genes in blastocysts but do The ultimate goal of this study was to compare the not alter adult glucose tolerance. Conversely, increasing transcriptional profiles to identify common signatures embryo culture stress with a higher oxygen tension of in vitro embryo manipulation and to determine if a

(20% O2) or an inferior culture medium exacerbates relationship exists between the acute transcriptional the transcriptional changes (over 2000 misexpressed response to in vitro stress and subsequent adult gene genes in blastocysts) and results in significant glucose expression. intolerance in adulthood (Fig. 1) (Donjacour et al. 2014, Feuer et al. 2014). Methods Over the past 15 years, the Rinaudo laboratory has performed microarray studies in mice to examine Animals how the many different components of in vitro All animals were maintained according to institutional embryo manipulation (culture medium composition, regulations and NIH guidelines, under a constant 12 h light/ oxygen tension and method of fertilization) have darkness cycle with ad libitum access to water and standard influenced transcriptional profiles in both blastocysts chow (PicoLab #5058). The strains used in this study included and adult offspring tissues (Rinaudo & Schultz 2004, vasectomized CD-1 males, B6D2F1/J males, CF-1 females and C57Bl/6J males and females. Beginning at 24 weeks, the The in vitro embryo 2133 postnatal IVF and control inbred C57Bl/6J cohorts were placed on a high-fat diet (Research Diets, Inc. #D12492) until time of death at 29 weeks. The postnatal IVF and control outbred CF1 × B6D2F1 cohorts were maintained on a standard chow

583 diet (20% protein, 9% fat, LabDiet) until time of death at # genes altered 264 in blastocysts 40 weeks. vs. in vivo Culture medium: KAA KAA WM Oxygen percentage: 5% O2 20% O2 20% O2 Embryo generation, culture and collection A detailed methodology of the embryo collection and transfer Adult offspring techniques has been published and may be found in the 400 400 following: (Rinaudo & Schultz 2004, Rinaudo et al. 2006, Giritharan et al. 2007, 2010, 2012, Feuer et al. 2014). Briefly, 300 300 CF-1 females aged 6–8 weeks were injected with 5 IU PMSG 200 200 followed 46–68 h later by 5 IU hCG to induce superovulation. Glucose (mg/dl) 100 Glucose (mg/dl) 100 For in vitro culture (IVC) experiments, superovulated dams IVF KAA 5% O2 IVF WM 20% O2 CTRL CTRL were mated overnight post-hCG injection with B6D2F1/J 30 60 90 120 30 60 90 120 males. The next morning, fertilized zygotes were flushed from Time (min) Time (min) ampullae, washed and cultured to the blastocyst stage at 37°C

Figure 1 Embryos with more severe changes in gene expression after under Ovoil (Vitrolife, #10029) with 5% CO2 in a modular in vitro manipulation will manifest abnormal glucose tolerance as adults. humidified chamber. At the two-cell stage, all unfertilized Different culture media and oxygen tensions in combination have oocytes were removed. Culture conditions included either distinct effects on blastocyst gene expression and adult glucose Whitten’s medium (WM (Whitten 1971); made in house) or tolerance. The number indicates statistically different gene expression potassium simplex optimization medium supplemented with compared to blastocysts conceived naturally and flushed from the amino acids (KAA (Ho et al. 1995); Millipore, MR-106-D), uterus. In general, suboptimal conditions such as Whitten’s medium with 5% or 20% oxygen (4 possible conditions). (WM) or higher oxygen concentration (20% O ) will more severely 2 For in vitro fertilization (IVF) experiments, 13–15 h after hCG affect embryo transcription and result in worse offspring glucose phenotypes (red panel). Conversely, optimized conditions such as K administration, the cumulous-oocyte complexes were isolated simplex optimized medium (KAA) and physiologic oxygen concentration from ampullae and incubated 4–6 h in HTF medium (Millipore,

(5% O2) have a lesser impact on blastocyst transcription with normal MR-070-D) with capacitated (1 h) cauda epididymal sperm from offspring glucose tolerance (green panel). IVF, in vitro fertilization. B6D2F1/J males. Fertilized zygotes were washed and cultured

Reproduction (2017) 153 107–122 www.reproduction-online.org

Downloaded from Bioscientifica.com at 09/26/2021 04:56:31AM via free access Transcriptome of embryo correlates to manipulation 109 to the blastocyst stage in WM and 20% oxygen as described ICM samples were immediately transferred to cell lysis previously. A second IVF cohort was generated using C57Bl/6J buffer provided in the PicoPure RNA Isolation Kit (Molecular animals and cultured under KAA and 5% oxygen conditions. Devices) and frozen at −80°C. For intracytoplasmic sperm injection (ICSI) experiments, oocytes were obtained the morning after superovulation as described for IVF. Cauda epididymides were isolated from Embryo transfer B6D2F1/J males aged 10–11 weeks, punctured gently with For postimplantation cohorts, pseudopregnancy was induced 30-G needles and incubated 10 min in a microcentrifuge tube by mating naturally cycling CF-1 females to vasectomized at 37°C in pre-warmed 1 mL EGTA Tris–HCl buffered solution CD-1 males, confirmed by the presence of a copulation (10 mM Tris–HCl pH 8.2, 50 mM EGTA and 50 mM NaCl). ICSI plug the next morning (considered day 0.5). Late-cavitating was performed using the top 800 µL sperm suspension, and blastocysts were transferred to the uterine horns of recipients oocytes containing the first polar body were microinjected on day 2.5 of pseudopregnancy. For control experiments, using a Piezo drill (PMM Controller, Prime Tech) as described superovulated dams were mated overnight; embryonic day 3.5 by Li and coworkers (2009). Fertilized zygotes were washed blastocysts (96 h after hCG administration) were flushed from and cultured to the blastocyst stage in WM and humidified air. the uterine horns and transferred immediately to the uterine To obtain late-cavitating blastocysts of similar morphology, horns of CF-1 recipients. This experimental strategy controls embryos were harvested at different time points: after for litter size and any effects of superovulation or the embryo 96–100 h (IVC), 106–110 h (IVF) or 112 h (ICSI) and snap- transfer procedure, both of which can influence imprinted gene frozen for microarray experiments. The in vivo-derived expression (Fortier et al. 2008, Rivera et al. 2008). Resulting control group was represented by blastocysts isolated from pups were not cross-fostered, as this procedure may alter adult naturally mated superovulated dams 96 h after hCG injection phenotypes and imprinted gene expression (Hager et al. 2009, (CF-1 × B6D2F1/J, or C57Bl/6J, as indicated in the text). At all Matthews et al. 2011). times, the experimental groups were compared to the same in vivo control breeds. All embryo generation experiments Microarray preparation were performed ≥4 times. We have previously shown that IVF and in vivo embryos Comprehensive descriptions of RNA extraction, amplification, derived by these protocols contain a similar number of inner fragmentation and hybridization to Affymetrix GeneChips for cell mass (ICM) cells (12.8 ± 0.4 vs 13.8 ± 0.5, not significant) each microarray experiment may be found in the following: (Giritharan et al. 2012). ICM was isolated from CF-1 × B6D2F1 (Rinaudo & Schultz 2004, Rinaudo et al. 2006, Giritharan control and IVF blastocysts cultured in WM and 20% oxygen et al. 2007, 2010, 2012, Feuer et al. 2014). Table 1 outlines the by immunosurgery (n = 3 times per treatment group). Briefly, individual microarrays presented or re-analyzed in this study. trophectoderm cells were lysed by incubating embryos 30 min For embryo studies, total RNA was extracted either with in WM containing 20 µg/mL anti-mouse rabbit antibodies TRIzol containing 2 mL Pellet Paint (Novagen) or a PicoPure (Sigma) and 30 min in WM with 5 µg/mL rabbit complement RNA Isolation Kit (Arcturus) from 3-4 pooled embryo replicates (Invitrogen) at 37°C. ICM samples were cleaned of destroyed comprising 80 blastocysts (IVC), 10 blastocysts (IVF or ICSI) or TE via repeated pipetting using a glass pipette with a 30–40 µm 40 ICM samples. RNA samples were submitted for preparation diameter under a dissecting microscope. Upon collection, and hybridization to the University of Pennsylvania Microarray

Table 1 Summary of the microarray experiments used in this study. Experiment Fertilization Culture Strain Tissue Platform Location Publication

Effect of embryo Natural (IVC) KAA 5% O2 CF1 × B6D2F1/J Blastocsyt Affymetrix University of Rinaudo and Schultz culture WM 5% O2 MOE 430A Pennsylvania (2004), Rinaudo et al. KAA 20% O2 (22,690 (2006) WM 20% O2 probes) Effect of Natural (IVC) WM 20% O2 CF1 × B6D2F1/J Blastocsyt Affymetrix University of Giritharan et al. (2007, fertilization IVF MOE 430.2 California, 2010) ICSI (45,101 San Francisco probes) Long-term IVF KAA 5% O2 C57Bl/6J Liver Affymetrix University of Feuer et al. (2014) effects: adult 29 week Gonadal fat Mouse 1.0 California, tissues female Skeletal muscle ST array San Francisco Pancreatic islets (>770,000 probes) Long-term IVF WM 20% O2 CF1 × B6D2F1/J Heart Affymetrix University of Unpublished effects: adult 40 week Mouse 1.0 California, tissues male ST array San Francisco (>770,000 probes) Cell type IVF WM 20% O2 CF1 × B6D2F1/J Inner cell mass Affymetrix University of Giritharan et al. (2012) specificity (ICM) MOE 430.2 California, (45,101 San Francisco probes) www.reproduction-online.org Reproduction (2017) 153 107–122

Downloaded from Bioscientifica.com at 09/26/2021 04:56:31AM via free access 110 S Feuer and others

Facility (IVC) or the Genomic Core Facility at the University of functional associations of genes (http://ingenuity.com). The California, San Francisco (IVF, ICSI and ICM). analysis includes canonical pathways overrepresented in gene In adult offspring, microarray analysis was performed on two lists, how genes and pathways integrate into broader biological postnatal cohorts: (1) 29-week-old C57Bl/6J female gonadal networks, as well as predicted upstream regulators responsible fat, liver, skeletal muscle and pancreatic islets derived from IVF for the cascade of gene expression changes. Only data KAA-5% oxygen or naturally (n = 3 per condition, with each experimentally observed in animal tissues were considered, animal providing the 4 tissues to minimize variation), and and fold-change thresholds were set such that the number of (2) 40-week-old CF1 × B6D2F1/J male cardiac tissue derived entities contributing to each analysis remained between the from IVF WM-20% oxygen or naturally (n = 3 per condition). recommended 100 and 2000. Animals contributing to these analyses were selected from at Heatmaps were generated using GENE-E software developed least two separate litters of 5–10 pups per condition. Total RNA by the Broad Institute (available at http://www.broadinstitute. was extracted and purified from previously frozen tissues using org/cancer/software/GENE-E/). the RNeasy mini kit (Qiagen), then submitted to the Gladstone Genomics Core Facility at the University of California, San Francisco for labeling, hybridization and scanning of the Microarray validation by real-time quantitative RT-qPCR microarrays. Whole frozen islets were also sent to the core To validate the microarray data, blastocysts derived in vivo facility for RNA isolation and amplification prior to microarray or cultured from the zygotic stage (IVC, WM with 20% processing. oxygen or KAA with 5% oxygen) were collected as described previously, and real-time quantitative RT-qPCR was conducted Microarray data analysis on 3–4 independent biological replicates containing 10 or more pooled blastocysts. Total RNA was extracted using the All microarray data were analyzed with GeneSpring GX 13.1 PicoPure RNA Isolation Kit (Applied Biosystems) and reverse software (Agilent Technologies). Scanned arrays were uploaded transcribed to cDNA using an iScript cDNA synthesis kit (Bio- into GeneSpring for background adjustment, summarization, Rad Laboratories). Quantification of gene transcripts was log transformation and baseline transformation. Samples were performed in duplicate with SyBr Green PCR Supermix using interpreted by culture medium, oxygen, method of fertilization 0.2 embryo equivalents of cDNA from each treatment group and cell type, via individual microarray platform. In addition to per reaction. Amounts of Arrdc4 (Forward: 5′-CCC TTA TTG β-actin, Gapdh signal is used as an internal quality control in ACT CCT ATG CT and reverse: 3′-CTT CTC CGT TAC AGT both Affymetrix MOE 430A and 430.2 arrays, and its intensity AGC CC); Slc7a3 (Forward: 5′-TTC TGG CCG AGT TGT CTA level can lead to the inclusion vs exclusion of specific samples TGT TTG and reverse: 3′-AGT GCG GTT CTG TGG CTG TCT as outliers in a dataset during a quality control analysis. We C (Bloise et al. 2012)); Socs3 (Forward: 5′-GCA GGA GAG recently discovered that Gapdh is altered in IVF blastocysts CGG ATT CTA CT and reverse: 3′-ACG CTC AAC GTG AAG compared to naturally derived control blastocysts (data not AAG TG); and Fgf4 (Forward: 5′-CCG GTG CAG CGA GGC shown), suggesting that previous analysis of these microarrays GTG GT and reverse: 3′-GGA AGG AAG TGG GTG ACC TTC may have led to the incorrect inclusion/exclusion of data. As a AT (Rappolee et al. 1994)) transcripts were normalized to levels result, these experiments were re-analyzed excluding Gapdh of histone 2A (H2A) transcript (Forward: 5′-ACA TGG CGG signal as a quality control influence. The principal component CGG TGC TGG AGT A and reverse: 3′-CGG GAT GCG CGT analysis (PCA) algorithm in GeneSpring was applied to all CTT CTT GTT). H2A was selected as a reference because it is specimens grouped by individual ART condition (including stably and reliably expressed across preimplantation embryo culture medium, oxygen percentage and method of fertilization) development (Jeong et al. 2005, Kuijk et al. 2007). using all genes and expressed sequence tags, separated by Primers were designed using PerlPrimer software (http:// microarray platform, to evaluate for patterns in gene expression perlprimer.sourceforge.net/) unless otherwise indicated. ART and underlying cluster structures. Fold-changes were calculated blastocyst transcriptional results are reported as normalized to based on the normalized scores instead of the raw expression gene expression in vivo. data. To identify differentiated genes between two groups (e.g. ART vs in vivo), an un-paired t-test with a significance threshold of P value <0.05 was applied to compare ART to control samples. The Effect-of-Culture and Cell-Type-Specificity Results experiments were analyzed using a 30% cutoff with Benjamini– Hochberg correction; the Effect-of-Fertilization experiments More severe preimplantation manipulation increases were analyzed using a 50% cutoff with Benjamini–Hochberg the levels of blastocyst transcriptional alteration correction; and the long-term adult tissue data were analyzed We examined the impact of in vitro manipulation on using a 30% cutoff without correction. blastocyst gene expression by testing either the specific effects of different preimplantation embryo culture Gene functional analysis and pathway analysis conditions or by varying the method of fertilization and Post-processing of the resulting gene lists was conducted using maintaining the culture conditions constant (Fig. 2). Ingenuity Pathway Analysis (IPA, June 2015 release), a structured First, superovulated and naturally fertilized zygotes knowledge repository that identifies biologically significant were cultured in either WM or KAA, and 5% or 20% relationships and networks based on previously characterized oxygen (4 conditions total) and compared to in vivo-

Reproduction (2017) 153 107–122 www.reproduction-online.org

Downloaded from Bioscientifica.com at 09/26/2021 04:56:31AM via free access Transcriptome of embryo correlates to manipulation 111

Effect-of-Culture Experiments Common fertilization method, different embryo culture conditions

ACDifferentially Expressed Transcripts B Principal Component Analysis Concordant Gene Misexpression vs in vivo 1500 All probes IVC IVC 2-fold 1200 Y-Axis KAA 5% O2 WM 5% O2

egulated (264 genes) (348 genes) 55 78 900 15 600 1 81 95 1491 ranscripts Upr 300 544 # T

20% O2 77 2 37

egulated 300 46 30 IVC IVC 600 KAA 20% O 327 WM 20% O Z-Axis 2 2 X-Axis (583 genes) (2133 genes) 900

anscripts Downr WM KAA WM KAA

# Tr 20% O2 5% O2

D IVC WM 20% O2 IVC KAA 20% O2 IVC WM 5% O2 IVC KAA 5% O2 196 altered pathways 109 altered pathways 41 altered pathways 35 altered pathways Predominant networks: Predominant networks: Predominant networks: Predominant networks:

biochemistry, amino acid metabolism production, molecular transport organismal injury & abnormalities repair, nucleic acid metabolism, small molecule biochemistry & assembly, hematological system biochemistry, infectious diseases system development & function, development & function musclular system development & cell organization & assembly, cell morphology function, tissue development connective tissue development & function development

Effect-of-Fertilization Experiments Different fertilization methods, common embryo culture conditions

E Differentially Expressed Transcripts FGPrincipal Component Analysis Concordant Gene Misexpression vs in vivo Y-Axis 3000 All probes IVC 2-fold WM 20% O2 egulated 1500 362 (883 genes)

400 70 347 ICSI 104 anscripts Upr 200

# Tr 54 65 5316 IVF ICSI WM 20% O WM 20% O 300 X-Axis 2 2 egulated (293 genes) not included: (5832 genes) ICSI KAA 20% O 600 Z-Axis 2

1500 H ICSI WM 20% O2 IVF WM 20% O2 IVC WM 20% O2

anscripts Downr 5000 48 altered pathways 42 altered pathways 119 altered pathways ICSI IVC IVF Predominant networks: Predominant networks: Predominant networks: # Tr

WM & 20% O2 small molecule biochemistry synthesis, humoral immune response

morphology, cellular movement metabolism, molecular transport metabolism, small molecule biochemistry

cell cycle death & survival

Figure 2 Effect of different embryo culture conditions and types of fertilization on blastocyst gene expression. Microarrays were performed on blastocysts either generated naturally and cultured in WM or KAA with 5% or 20% oxygen (Effect-of-Culture experiments, A, B, C and D) or produced by IVF, ICSI or naturally and cultured in WM and 20% oxygen (Effect-of-Fertilization experiments, E, F, G and H), with naturally derived blastocysts as controls. (A) Number of transcripts altered after each culture condition shows a strong impact of suboptimal WM and high oxygen on transcription. (B) Principal component analysis (PCA) indicates that oxygen concentration has a more robust effect on gene expression than culture medium composition. (C) Venn diagram of concordant gene misexpression across the 4 culture conditions. (D) Ingenuity pathway analysis (IPA) highlighting the predominant networks (network score ≥30) associated with the gene expression changes. (E, F, G and H) Same as A, B, C and D for the Effect-of-Fertilization experiments. (E) There was a severe effect of fertilization by ICSI on blastocyst gene expression, which (F) contributed significantly to PCA clustering. (G) Venn diagram comparing the concordance of gene misexpression after each fertilization type, with (H) top IPA networks associated with the gene lists. ICSI, intracytoplasmic sperm injection; IVC, in vitro culture; IVF, in vitro fertilization; KAA, potassium simplex optimization medium with amino acids; WM, Whitten’s medium. www.reproduction-online.org Reproduction (2017) 153 107–122

Downloaded from Bioscientifica.com at 09/26/2021 04:56:31AM via free access 112 S Feuer and others generated blastocysts from superovulated dams (flushed blastocyst gene expression. We chose 4 genes (Arrdc4, blastocyst control). Both WM and higher oxygen Fgf4, Slc7a3 and Socs3) for validation in naturally- tensions are considered stressful; conversely, KAA and derived blastocysts cultured from the zygotic stage (IVC) 5% oxygen are optimal for mouse embryo culture and in either KAA-5% oxygen or WM-20% oxygen, which represent current IVF clinical practices (Schwarzer et al. confirmed the microarray data for all conditions except

2012, Chronopoulou & Harper 2015). This gradation for Slc7a3 expression in IVC KAA-5% O2 blastocysts in stress was reflected in the number of transcripts (Fig. 3B). Further, 9 pathways associated with growth, altered by each culture condition, with WM and higher cancer, pluripotency and response to stress (Fig. 3C) oxygen levels increasingly perturbing blastocyst gene were collectively enriched in the in vitro manipulated expression (Fig. 2A). Principal component analysis embryos. revealed a marked effect of oxygen on transcription, significantly more pronounced than the relative influence culture medium composition (Fig. 2B, red circle). Each The impact of IVF on gene expression is cell condition begat a unique blastocyst transcriptome, with type-specific gene expression perturbations associated with distinct Because blastocysts are composed of both trophectoderm pathways and gene networks (Fig. 2C and D). 77 genes and the inner cell mass (ICM), we further probed our were commonly misexpressed in all four in vitro culture embryo data by comparing the blastocyst IVF WM-20% (IVC) conditions and were enriched for stress response oxygen transcriptome to blastocyst ICM derived from pathways, cell cycle control, cancer and pluripotency the same conditions (Supplementary Table 3). Of the signaling (Supplementary Fig. 1, Supplementary Table 1, 293 genes and 42 pathways affected in the blastocysts, see section on supplementary data given at the end of only 39 genes and 2 pathways were similarly altered this article). in the ICM (Fig. 3D), confirming that controlling for We next evaluated the influence of different different cell types is highly relevant when analyzing fertilization techniques by comparing blastocyst the influence of in vitro embryo manipulation. Only 6 transcriptomes among embryos fertilized via ICSI, IVF or of the 39 genes, as well as both pathways (cell cycle: naturally (in vitro culture only, IVC) and cultured from G1/S checkpoint regulation and p53 signaling) were zygote to blastocyst in WM and 20% oxygen, vs in vivo. also highlighted in the ‘Common Fingerprint’ of in vitro ICSI—a more mechanically intrusive procedure—had a embryo manipulation (Fig. 3E and F, red text). dramatic effect on transcription, altering the expression of over 7000 transcripts corresponding to 5832 genes (Fig. 2E, F and G). Unexpectedly, the IVC condition had There is no common gene signature of in vitro embryo a more prominent effect on gene expression than IVF. As manipulation in adult tissues with the effect-of-culture experiments, each fertilization Previously, we examined the long-term effects of IVF method had a separate impact on blastocyst gene (KAA and 5% O2) on adult offspring liver, skeletal expression, with altered gene lists associated with unique muscle, gonadal fat and pancreatic islet transcriptomes pathways and networks (Fig. 2H and Supplementary in an inbred C57Bl/6J mouse model and found no Table 2). 104 genes showed concordant misexpression concordant changes or common biological themes in each fertilization condition and were associated with (Feuer et al. 2014). These particular tissues were selected a wide range of cellular processes including cytoskeletal for their involvement in either secretion of or response dynamics, metabolite biosynthesis, the cell cycle and to insulin, as these IVF offspring show altered growth growth (Supplementary Fig. 2). curves, pancreatic beta cell hyperinsulinemia, and are predisposed to glucose intolerance (Fig. 1). The gene expression changes were subtle ( 95% under 2-fold Blastocysts exhibit a common gene signature of in vitro > change), and re-analysis with updated software confirmed embryo manipulation that only one gene, Gm14403, and no pathways were Following the observation that each fertilization method altered in all four IVF tissues compared to naturally or culture condition resulted in unique gene expression conceived controls (Fig. 4 and Supplementary Table 4), changes, we searched for evidence of a ubiquitous indicating a tissue- and/or cell type-specific impact of transcriptomic fingerprint indicative of a shared effect IVF on gene expression. This was corroborated by the of embryo manipulation. We identified a list of 18 genes fact that removal of the pancreatic islet data (a highly concordantly misexpressed in all in vitro conditions complex tissue with many cell types) led to modest compared with in vivo blastocyst gene expression overlap (38 genes and 3 pathways) among fat, liver and (Fig. 3A). Importantly, all of the genes were similarly muscle IVF transcriptomes (Fig. 4C and E). Interestingly, increased or decreased compared to in vivo embryos pathway analysis revealed that the 38 concordantly except for Camk1 (increased in the ICSI microarray only), misexpressed genes were largely associated with glucose suggesting a common effect of in vitro manipulation on metabolic flux (through glycolysis, the TCA cycle and

Reproduction (2017) 153 107–122 www.reproduction-online.org

Downloaded from Bioscientifica.com at 09/26/2021 04:56:31AM via free access Transcriptome of embryo correlates to manipulation 113

A Fertilization B Validation by RT-PCR 3 2 IVC WM 20% O2 MA IVC 2 2 IVC KAA 5% O2 MA 2 IVC WM 20% O2 PCR 2 2 2 IVC KAA 5% O2 PCR CTRL 5% O 20% O 2

KAA WM 5% O KAA WM 20% O IVC WM 20%IVF O WM 20%ICSI O WM 20% O Arrdc4 B4galt6 1 Blnk

Camk1 Relative expression Cdh2 Dusp9 Eomes Arrdc4 Fgf4 Slc7a3 Socs3 Fbln1 Fgf4 Grhl1 Hif2a C Fertilization Lap3

2 IVC 2 Mettl17 2 Slc7a3 2

2 2 Socs3 2 Tspan1 5% O 20% O

Ube2a A Ube2l3 KAA WM 5% O KA WM 20% O IVC WM 20%IVF O WM 20%ICSI O WMCanonical 20% O pathway Fold change v. CTRL p53 signaling Cell cycle: G1/S checkpoint regulation -4 3 Role of Oct4 in mammalian ESC pluripotency Molecular mechanisms of cancer E Pancreatic adenocarcinoma signaling HGF signaling GADD45 signalinng

Blastocyst ICM Blastocyst ICM Hypoxia signaling in the cardiovascular system Akr1c21 Hist1h2bc/e/g Role of Jak1 and Jak3 in c cytokine signaling Ass1 Iqcg p-value B4galt6 Mast4 Bbx Nr0b1 0.3 0.05 <0.001 Bpnt1 Oas1d|Oas1e C130026I21Rik Parvb Ccng1 Pdrg1 Cd81 Phlda3 D IVF WM 20% O2 Cdh2 Pramel4 Blastocyst ICM Cdkn1a Ptprs Ddit4l Rap2b Dhx16 Rpl5 Genes 254 346 39 Dpep3 Rsu1 Pathways 40 13 Dusp9 Runx1t1 2 Dynlt3 Slc7a3 E2f8 Timd2 Eda2r Trp53inp1 Eomes Ube2a F Exoc4 Wdr35 Gpx3 Canonical pathway Fold change v. CTRL Blastocyst ICM Cell cycle: G1/S checkpoint regulation p53 signaling -4 3

Figure 3 A common, but cell type-specific, fingerprint ofin vitro embryo manipulation. (A) A core list of 18 genes were altered by in vitro embryo manipulation, regardless of the culture conditions or the method of fertilization. (B) Real-time qPCR verification of the microarray data.

Four of the 18 genes were selected for analysis in IVC blastocysts cultured in KAA-5% O2 (green) or WM-20% O2 (red), and the graph shows their expression relative to controls (normalized to 1) from both the microarray (MA) and PCR data. (C) 9 pathways were collectively enriched in association with the individual microarray experiments evaluating the impacts of fertilization conditions and culture conditions on blastocyst transcription. (D, E and F) A cell type-specific effect ofin vitro fertilization. (D) Venn diagram showing the differential effects of IVF WM-20% oxygen culture conditions on blastocyst vs inner cell mass (ICM) gene expression and pathway association. (E) Only 39 genes and (F) two pathways were concordantly affected in both blastocysts and ICM, in spite of similar conditions of fertilization and culture in vitro. Genes and pathways also present in the ‘Common Fingerprint’ are highlighted in red. Legends (fold-changes and P value) are applicable to all heatmaps in the figure. www.reproduction-online.org Reproduction (2017) 153 107–122

Downloaded from Bioscientifica.com at 09/26/2021 04:56:31AM via free access 114 S Feuer and others

A B C Differentially Expressed Transcripts Concordant Gene Misexpression

Fat Liver Muscle 1600 0610031J06Rik All probes IVF KAA 5% O2 IVF KAA 5% O2 Arel1 2-fold Gonadal fat Skeletal muscle egulated 1200 Arhgdia|Rnf11 (1003 genes) (2179 genes) Arid4b|Zfp932 692 2198 Atp6v0c 800 Bcap29 142 15 274 C1galt1c1 Casp9 ranscripts Upr 400 Chd7 # T 448 1433 638 Dlst Dock6 Dusp3 1 Dyrk1b 400 64 106 egulated Ebag9 8 3 Echdc1 800 Eno1b IVF KAA 5% O2 35 IVF KAA 5% O2 Ets3 1200 Pancreatic islets Liver Fkbp3 (580 genes) (1898 genes) Gm14403 Gnpnat1 ranscripts Downr 1600 Ift74

# T Muscle Liver Fat Islets Mir680-1 IVF KAA 5% O2 E Myo9a Ndst1 Fat Liver Muslce Pathway overlap Orc4 D Complement system Pdcd10 FAT MUSCLE Huntington's disease signaling Phf2 Pmvk IVF KAA 5% O2 IVF KAA 5% O2 IL-1 signaling Rprl3 95 altered pathways 89 altered pathways P-value Scarf1 Predominant networks: Predominant networks: Stxbp3a 0.3 0.05 <0.001 Tab2 biochemistry, energy production to-cell signaling & interaction Taf1d Thsd1 cell-to-cell signaling & interaction, 1) skeletal muscle system, F Pathway enrichment: 38 concordant genes Timp3 development & function of the 2) hematological system Tmbim6 cardiovascular system PKR in interferon induction and antiviral response Tmem243 disease, organismal injury & 2-ketoglutaration dehydrogenase complex Tmx3 diseases, endocrine system disorders abnormalities UDP-N-acetyl-D-glucosamine biosynthesis Trappc9 UDP-N-acetyl-D-galactosamine biosynthesis Fold change v. CTRL Ubiquinol-10 biosynthesis ISLETS LIVER Mevalonate pathway -2 2 IVF KAA 5% O2 IVF KAA 5% O2 Granzyme B signaling Parkinson's disease 19 altered pathways 58 altered pathways Geranylgeranyldiphosphate biosynthesis Predominant networks: Predominant networks: Endoplasmic reticulum stress pathway TCA cycle Tumoricidal function of hepatic natural killer cells free radical scavenging, therapeutics, connective tissue Triacylglycerol degradation small molecule disorders Gluconeogenesis biochemistry Glycolysis small molecule biochemistry, Cell cycle control of chromosomal replication hematological disease Superpathway of cholesterol biosynthesis Sonic hedgehog signaling

Figure 4 Adult IVF transcriptional signatures are tissue specific. Microarray comparison between female 29-week gonadal fat, liver, skeletal muscle and pancreatic islets derived from IVF KAA-5% oxygen animals and in vivo flushed blastocyst controls. (A) Number of transcripts altered in each IVF tissue. The majority of changes were modest and ≤2-fold different from controls. (B) Venn diagram comparing the concordance of gene misexpression after IVF across the 4 tissues. (C) Fold-change directionality of concordant IVF gene misexpression among fat, liver and muscle, with the one gene altered in all 4 tissues (Gm14403) indicated in red. (D) Ingenuity pathway analysis (IPA) highlighting the predominant networks (network score >30) associated with the gene expression changes in each tissue, and (E) the commonly altered pathways between IVF fat, liver and muscle gene lists. (F) Top canonical pathways enriched in the 38 genes misexpressed in IVF fat, liver and muscle. mitochondrial function, gluconeogenesis and UDP-N- microarray data between IVC embryos and acetylglucosamine biosynthesis, Fig. 4F). their analogous adult IVF liver, skeletal muscle, gonadal fat and pancreatic islets to determine if different tissues were vulnerable to retaining the There is no obvious transcriptional link between in vitro embryos and their corresponding adult tissues in vitro embryo signature (Fig. 5 and Supplementary Table 5). There were minimal commonalities Given that in vitro embryo manipulation results between the two time points, with fewer than 2% of in transcriptional changes in both embryos the genes affected in each adult IVF tissue similarly and adult tissues, we next inquired if the gene altered in the blastocyst. Further, the shared gene misexpression in adult tissues could be attributed misexpression between embryos and adult tissues to or explained by the transcriptional alterations was frequently altered in opposite directions. in blastocysts. We compared the KAA-5% oxygen Pathway analysis demonstrated that although the

Reproduction (2017) 153 107–122 www.reproduction-online.org

Downloaded from Bioscientifica.com at 09/26/2021 04:56:31AM via free access Transcriptome of embryo correlates to manipulation 115

A Blastocyst Liver B Blastocyst Muscle by differences in mouse strain and conception condition (CF-1 × B6D2F1/J IVC blastocysts, C57Bl/6J IVF adult 230 1869 Genes 236 2708 34 28 tissues). We therefore expanded our analysis by 31 54 Pathways 25 79 4 10 investigating whether more stringent and homogenous conditions would reveal a relationship between the acute transcriptional response to in vitro stress and

Blastocyst Liver Blastocyst Liver Blastocyst Muscle Blastocyst Muscle 4933426M11Rik Itfg1 Ap5m1 Itm2b subsequent gene expression in adulthood. Microarray Acp5 Mogs Arrdc4 Katnb1 Acpp Msx1 Bcap29 Lima1 was performed on 40-week-old IVF and control cardiac Arrdc4 Ndrg1 Bhlhe40 Ppic B230118H07Rik Nid1 Blnk Pqlc1 tissue and compared to data from ICM or whole Bcap29 Nmd3 Bmp4 Pstk blastocysts derived from analogous conditions (outbred Bmp4 Nup133 Coro1b Rhou Carkd Pros1 Cul4b Sall4 CF-1 × B6D2F1 tissues conceived by IVF with WM and Ccng1 Pstk Dhx16 Tmem39a Dnajb1 Reep3 Enpp1 Tmem40 20% oxygen). We selected the heart because in addition Dpm1 Rfc4 Foxh1 Tmem41b Dynll2 Serpinh1 Hbegf Tuba4a to glucose intolerance, these mice exhibit cardiac left Ephx2 Slc4a2 Hdlbp Ube2a Grn Slc6a13 Itfg1 Zfand5 ventricular hypertrophy (Donjacour et al. 2014), a Gstm1 Sparc H1f0 Spp1 condition that has also been reported in IVF/ICSI infants Hexa Tmem39a (Valenzuela-Alcaraz et al. 2013). Relative to controls,

Blastocyst Muscle Canonical pathway Molecular mechanisms of cancer 1443 transcripts corresponding to 1361 genes were Cell cycle: G1/S checkpoint regulation Canonical pathway Blastocyst Liver Estrogen-mediated S-phase entry altered in the IVF hearts, and only 16 genes exhibited Cell cycle: G1/S checkpoint regulation Huntington's disease signaling Huntington's disease signaling Small cell lung cancer signaling fold-changes greater than ±2-fold (Fig. 6A). 99 pathways Molecular mechanisms of cancer Protein ubiquitination pathway were significantly enriched in the transcriptional Protein ubiquitination pathway Cyclins and cell cycle regulation Prostate cancer signaling profiles, with overrepresentation in the inflammatory Cell cycle regulation by BTG family proteins Aldosterone signaling in epithelial cells C Blastocyst Fat response, hematological system development and function, cancer, cellular assembly and organization, as 249 997 D 15 Blastocyst Islets well as cell-to-cell signaling, interaction and movement 32 93 3 Genes (Fig. 6B, C and Supplementary Table 6). Pathways 261 577 3 35 19 When we compared the IVF WM-20% oxygen 0 transcriptional profiles between the early embryo and

Blastocyst Gonadal fat Blastocyst Gonadal fat adult heart, there again was minimal concordance in Amz2 Depdc7 Apaf1 Dock9 Blastocyst Islets gene misexpression: 22 genes and 4 pathways were Bcap29 Enpp1 2810025M15Rik Cd81 Fcgr3 Cd81 commonly altered in blastocysts and heart, whereas Crtap Lcp1 Tex19.1 28 genes and 1 pathway were shared between ICM Dcaf11 Lgals3 Marcksl1 and heart (Fig. 6D, E, F and Supplementary Table 5). Nid1 Fold change v. CTRL P-value Wbp5 Only 4 genes (Ccng1, Rsu1 Runx1t1 and Trp53inp1)

-4 3 0.3 0.05 <0.001 Blastocyst GonadalCanonical fat pathway were misexpressed in blastocysts, ICM and heart tissue Coagulation system (Fig. 4E, red asterisks). Surprisingly, more than half of Extrinsic prothrombin activation pathway Huntington's disease signaling the genes with shared misexpression were altered in Figure 5 No common in vitro signature of KAA and 5% oxygen opposite ways between the embryos and heart. between IVC blastocysts and IVF adult tissues. Comparison of microarray data between IVC KAA-5% oxygen CF1 × B6D2F1/J blastocysts and IVF KAA-5% oxygen C57Bl/6J 29 weeks female (A) A set of common upstream regulators are predicted liver, (B) skeletal muscle, (C) gonadal fat and (D) pancreatic islets, vs to drive the transcriptional changes in embryos and in vivo-derived controls. Venn diagrams show overlap of gene adult tissues misexpression and ingenuity pathway analysis, with heatmaps depicting the fold-change directionality of concordant gene Ingenuity pathway analysis offers an upstream misexpression as well as pathway enrichment between blastocysts and regulator prediction algorithm that evaluates corresponding adult tissues. Genes and pathways also present in the gene lists and microarray fold-change to identify embryo ‘Common Fingerprint’ of in vitro manipulation are highlighted molecules (transcriptional regulators, growth factors, in red. Legends are applicable to all heatmaps in the figure. receptors, transporters, etc.) with direct actions on the differentially expressed genes. Based on fold- gene misexpression was highly tissue specific, some change directionality of the downstream target genes, pathways were consistently enriched in both KAA-5% IPA predicts whether upstream regulator inhibition O2 blastocysts and adult tissues, including cell cycle or activation might govern the differences observed. G1/S checkpoint regulation, molecular mechanisms Given the significant cell type specificity observed in of cancer, protein ubiquitination pathway and each IVF tissue transcriptome, we hypothesized that Huntington’s disease signaling. common regulators might control the gene expression Although this initial comparison found negligible changes, but based on tissue or cell type, this would transcriptional linkage between in vitro embryos and manifest in unique transcriptional signatures (Fig. 7 adult tissues, the findings could have been influenced and Supplementary Tables 1–6). www.reproduction-online.org Reproduction (2017) 153 107–122

Downloaded from Bioscientifica.com at 09/26/2021 04:56:31AM via free access 116 S Feuer and others

A B C Top canonical pathway HEART Heart IVF WM 20% O Chd7 Toll-like receptor signaling 2 Cnnm2 HMGB1 signaling 99 altered pathways Rnu2 MIF-mediated glucocorticoid regulation Predominant networks: Lrrc52 IL-10 signaling Il17ra MIF regulation of innate immunity movement, hematological system Figure 6 No common IVF signature between Stxbp4 development & function Tmem176a RANK signaling in osteoclasts embryos and adult tissues from homogenous Chi3l1 Role of IL-17A in arthritis cancer, cellular assembly & organization conditions. (A, B and C) Misregulation in adult Ccl7 Glioma invasiveness signaling Ccl19 Pancreatic adenocarcinoma signaling male 40-week cardiac tissue derived from IVF Il1rb NGF signaling WM-20% oxygen mice. (A) Of the 1361 genes Thbs1 UDP-N-acetyl-D-glucosamine biosynthesis Dennd4a NF-kB signaling E significantly misexpressed in IVF heart Gfpt2 IL-6 signaling compared to controls, only 16 showed a Ptgs2 LXR/RXR activation ICM Heart Ngf PEDF signaling Alppl2 fold-change greater than ±2-fold. (B) Ingenuity Blastocyst Heart Ilk signaling Ccng1 Arhgef2 pathway analysis of canonical pathways most IL-17 signaling * Efhd1 Brd8 LPS-stimulated MAPK signaling Ergic1 significantly enriched P( 0.001) in the altered Ccng1 < * Fads1 Cd68 IVF heart genes and (C) their associated Fbxl20 Dyrk2 Gab2 D Etv3 networks (network score ≥25). (D) Venn Gfpt2 Fbln1 Blast Heart ICM Heart Hk1 diagram indicating the low number of genes Fbln5 Mmp14 Hif2a (top number) and number of pathways Myh9 271 1339 Genes 357 1333 Hsd17b11 Pitx2 (bottom number) overlapping between the IVF 22 28 Immt 38 95 Pathways 14 98 Plau 4 1 Lbx1 WM-20% oxygen blastocyst (blue) or ICM Ptprj Mtus2 Rftn1 (green) transcriptomes and corresponding Pdgfrb Rsu1 Pgm2l1 * adult heart (grey). (E) The genes similarly Runx1t1 Pip5k1a * Ryr1 misexpressed between blastocyst and heart, or Ptp4a3 F Serping1 Rsu1 ICM and heart, and their directionality of * Slc44a1 Runx1t1 Ssh2 change. Red asterisks indicate genes * Ssr4 Blastocyst Heart ICM Heart Tbc1d13 Tpd52 commonly altered in both blastocysts and HGF signaling Protein Kinase A signaling Tbx18 Trp53inp1 Pancreatic adenocarcinoma signaling * Tfpi ICM, vs heart. (F) Overlap of pathways Signaling by Rho family GTPases Tnrc18 associated with the altered genes between the Thrombin signaling P-value Fold change v. CTRL * Trp53inp1 Tspan8 embryo data and the heart. Legends are 0.3 0.05 <0.001 -4 3 Zic5 applicable to all heatmaps in the figure.

Analysis of the transcriptional alterations present in shared by the ART embryo expression profiles and the blastocysts after each condition of in vitro manipulation condition-specific embryo vs adult comparisons. identified a core group of 15 regulators—comprising kinases, enzymes, growth factors, phosphatases, ligand-dependent nuclear receptors and transcriptional Discussion regulators—that collectively function directly upstream This study used a mouse model to determine if there are of the altered genes, 10 of which additionally target common mechanisms involved in the preimplantation the misregulated ICM genes (Fig. 7A). The predicted embryo’s response to different in vitro exposures activation or inhibition of these regulators was largely frequently used in assisted reproductive technologies, concordant across all of the in vitro embryo conditions. and if there exists a relationship between these acute Focusing on the embryo-to-adult comparisons (Fig. 7B transcriptional responses and gene expression in and 7C), there were 7 putative regulators of the genes subsequent offspring adult tissues. This work has wide altered in blastocysts, ICM and adult male heart derived clinical relevance not only because well over 5 million from the IVF WM-20% oxygen conditions, compared children have been born from ART (ESHRE 2014) but to in vivo controls (Fig. 7B). Based on the directionality also because it remains unclear how environmentally of the gene expression changes, the majority of these induced developmental reprogramming can predispose putative regulators were predictably activated (e.g. to abnormal metabolic phenotypes in adulthood. The functionally increased) in blastocysts and heart, but main findings of the manuscript are that (1) mouse inhibited in the ICM. blastocysts exhibit a common gene signature of in vitro Regarding the KAA-5% oxygen blastocysts and their embryo manipulation; (2) blastocyst transcriptional corresponding adult female fat, liver, and muscle, 21 alterations are increased with greater preimplantation regulators operate upstream of the misexpressed genes at manipulation or stress; (3) the impact of embryo both time points (Fig. 7C). Interestingly, these molecules manipulation on gene expression is cell type specific; showed predicted inhibition in blastocyst and liver (4) there is minimal overlap between IVF transcriptional profiles, but activation in fat and muscle, suggesting profiles in different adult tissues and (5) there is no that differential mediation of gene expression changes obvious transcriptional link between in vitro embryos by common regulators is tissue-specific.Supplementary and subsequent adult tissues derived from similar Figure 3 summarizes the predicted upstream regulators in vitro conditions. Finally, (6) we identified a set of

Reproduction (2017) 153 107–122 www.reproduction-online.org

Downloaded from Bioscientifica.com at 09/26/2021 04:56:31AM via free access Transcriptome of embryo correlates to manipulation 117

Shared in ART embryos Shared in the embryo-to-adult conditions

A B C ICM Fertilization IVF WM 20% O2 KAA 5% O2

2 IVC 2 Figure 7 Common upstream regulators may 2 2 IVC IVF

2 govern the transcriptional changes present in 2 2 2 in vitro embryos and adult tissues. Ingenuity Blastocyst ICM Heart pathway analysis identifies regulators CEBPA functioning upstream of the altered genes, and IVF WM 20% O KAA 5% OWM 5% O KAA 20% OWM 20% O IVC WM 20%IVF O WM 20%ICSI O WM 20% O Cg CD38 TGFB1 predicts whether these regulators are impaired Blastocyst Gonadal FaLivert Skeletal Muscle CEBPB TGFBR2 Cg (negative z-score, blue) or activated (positive Cg TNF CSF2 z-score, red) based on fold-change CTNNB1 TP53 CTNNB1 DICER1 TP73 misexpression. (A) Shared upstream regulators IFNG IKBKB IKBKB of the ART embryo expression profiles. 15 MYC IL10RA molecules were predicted to function TGFB1 IL1B TGFBR2 upstream of the transcriptional alterations IL4 TP53 present in the in vitro blastocyst, 10 of which IL5 AHR IL6 were similarly highlighted in ICM. (B and C) CHUK KRAS Upstream regulators present in the embryo-to- IKBKG MYC PPARA adult condition-specific comparisons. (B) 7 NFE2L2 PTEN NOS2 regulators commonly target the altered genes PPARA in IVF WM-20% oxygen blastocysts, ICM and PRL adult heart, and (C) 21 regulators collectively SOX4 mediate the gene expression changes STAT3 Activation z-score TGFB1 observed in KAA-5% oxygen blastocysts predicted predicted inhibition activation TNF and corresponding adult female fat, liver -3.5 3.5 TP53 and muscle. common upstream regulators that could drive the oxidative stress is exemplified by changes in the hypoxia- transcriptional changes present in IVF embryos and inducible transcription factor Hif2α, the expression of adult tissues. genes involved in ubiquitination and protein turnover The first notable finding is that although each (Ube2a, Ube2l3, Lap3, Socs3, Arrdc4, (Shang & Taylor in vitro condition led to a specific and unique blastocyst 2011)), and in particular Arrdc4 transcription. transcriptional profile, we identified a common Hif2α is a well-known transcriptional regulator of 18-gene and 9-pathway signature of preimplantation cellular and organismal responses to oxygen deprivation embryo manipulation that was present in all in vitro (Semenza 2016). Oxygen deprivation is sensed by embryos irrespective of culture condition or method of mitochondria and converted to a reactive oxygen fertilization. Coupled with the variation in time, location, species-dependent signal that results in HIFα subunit scientist and microarray platform for these experiments, stabilization (Chandel 2010). Although transcriptional the emergence of a core fingerprint ofin vitro embryo induction of HIFα subunit gene expression under manipulation is particularly robust. Further, the 18 hypoxic conditions is less widely reported, it has been genes were altered in the same direction except for demonstrated in hematopoietic stem cells that redox Camk1 expression. Camk1 operates within the calcium- stress can trigger Hif2α expression to help modulate triggered signaling cascade; a hallmark of mammalian redox balance via the homeodomain transcription fertilization is the presence of repetitive Ca2+ oscillations factor Meis1 (Kocabas et al. 2012, Miller et al. 2016). from the site of sperm penetration. The unique increase in Along these lines, the arrestin family member Arrdc4 is Camk1 expression in ICSI blastocysts compared to other of interest because it also plays roles in cell metabolism ART conditions might be explained by data showing that and is related to molecules such as Txnip, which can ICSI induces subtle changes in oocyte Ca2+ oscillatory influence cellular redox status in response to glucose patterns compared to conventional insemination (Sato availability (Zhou & Chng 2013) and thus integrate et al. 1999, Miao et al. 2012). cellular oxidative and metabolic states (Patwari & Lee In addition to several key transcriptional regulators 2012). Txnip expression is upregulated in IVF embryos with crucial roles in development and cellular (Feuer et al. 2014), and other models have correlated this differentiation (Eomes, Fgf4, Grhl1, Hif2α), many of increase with impaired glucose tolerance (Parikh et al. the genes commonly affected by in vitro manipulation 2007), oxidative stress (Schulze et al. 2004), apoptosis reflect an increase in reactive oxygen species (ROS). (Chen et al. 2008) and diabetes pathogenesis (Shalev Indeed, there is abundant evidence that embryo culture 2008). The induction of these transcripts suggests a induces an increase in ROS (Goto et al. 1993, Cebral perturbation of the metabolic and redox axes in early et al. 2007, Martin-Romero et al. 2008). In our data set, embryos conceived and cultured in vitro. In further www.reproduction-online.org Reproduction (2017) 153 107–122

Downloaded from Bioscientifica.com at 09/26/2021 04:56:31AM via free access 118 S Feuer and others support of this, GADD45 and p53 signaling—identified disorders such as insulin resistance (Moskalev et al. as one of the most significantly misregulated pathways in 2012), alterations in this signaling pathway during an our dataset—are implicated in cellular and organismal epigenetically vulnerable period might have lasting responses to perturbations in cell metabolism or redox consequences for metabolic health. status (Zhuang et al. 2012, Salvador et al. 2013). This is highly relevant, as it is widely believed that The remaining conserved transcripts were repressed epigenetic changes mediate developmental plasticity in embryos after in vitro manipulation. Many of these, and contribute significantly to the programming of including Eomes, Fgf4, Socs3 and Dusp9 are implicated environmental signals (Gluckman et al. 2011). Given in placental development or are critical for trophoblast that both DNA methylation and histone modifications stem cell function (Maltepe & Fisher 2015). We have are extensively remodeled in the preimplantation documented changes in mouse IVF placentae in prior embryo (Reik 2007) and because culture conditions studies (Delle Piane et al. 2010, Bloise et al. 2012, 2014), can affect chromatin marks (Doherty et al. 2000), it is which suggests that dysregulation of critical placental genes possible that IVF-induced changes in transcriptional and is conserved across IVF techniques and could potentially epigenetic regulation are responsible for propagating contribute to alterations to placental development and the adult DOHaD phenotypes. function in this setting. Further, ART-induced remodeling Although each in vitro condition exerted a marked of the placental landscape could contribute to the adult effect on blastocyst transcription, we additionally found metabolic phenotypes observed in these mice. Another that more severe transcriptional changes occurred noteworthy gene in this common fingerprint is Ube2a, with increased preimplantation stress. In particular, which catalyzes the monoubiqitination of histone H2B culture of embryos using atmospheric oxygen had at Lys-120 to form H2BK120ub1; this marks epigenetic a dramatic effect on gene expression (Fig. 2A and B). transcriptional activation, elongation by RNA pol II, This was compounded by the addition of a suboptimal telomeric silencing and is a requirement for H3K4 me culture medium (WM), whereas the relative influence and H3K79me formation. Together, these genes suggest a of culture medium composition alone on transcriptional common fingerprint ofin vitro embryo manipulation that profiles was small in comparison. Further, the Effect-of- links cellular differentiation, oxidative stress, glucose Fertilization experiments (Fig. 2E, F, G and H) showed an metabolism and insulin resistance with epigenetic outstanding effect of ICSI compared to IVC or IVF. The changes. severity of the stress (oxygen and ICSI) was also reflected The pathways with significant overrepresentation after in the degree of fold-change for the 18-gene signature of any in vitro manipulation are also implicated in cellular in vitro manipulation in embryos. Interestingly, embryo development and differentiation, proliferation and the culture of naturally fertilized zygotes had a stronger cell cycle (including cancer signaling), as well as stress impact on gene expression than in vitro fertilization response signaling via p53 and GADD45. Given that and culture in the same conditions (Fig. 2G, 883 vs DOHaD-related phenotypes are often metabolic, it is 293 altered genes). This highlights the remarkable notable that there was no enrichment for any particular vulnerability of zygotes to the stresses incurred through metabolic pathways. However, many of the pathways the embryo isolation process, and may be related to the with enrichment reveal interesting connections between timing of zygotic genome activation. Indeed, transfer the acute responses to in vitro embryo manipulation of mouse zygotes after in vitro exposure to different and long-term metabolic phenotypes. First, oxygen nutritional milieu during the pronuclear stage alters birth homeostasis is important for cardiovascular system weight in a condition-specific manner Banrezes( et al. patterning during embryogenesis, including adaptive 2011). A notable physiological feature that separates the responses in signaling and redox homeostasis (reviewed IVC group is that these in vivo-generated embryos are in Simon et al. 2008); adult mice generated by IVF exposed to seminal fluid, which can elicit unique gene

WM-20% O2 conditions exhibit cardiac hypertrophy expression changes (Schjenken & Robertson 2015). (Donjacour et al. 2014). GADD45 and p53 signaling In addition to gene expression, protocol-specific effects reflect cell stress and DNA damage, with known have also been described for blastocyst cell number and roles in the regulation of the cell cycle, senescence, lineage ratio, implantation efficiency, fetal and placental survival and apoptosis. The link between GADD45 growth, postnatal growth and adult glucose metabolism and apoptosis is significant, as GADD45 induction is (Scott et al. 2010, Kohda et al. 2011, Schwarzer et al. known to downregulate pro-apoptotic JNK signaling 2012). Along these lines, it is likely that additional ART and therefore suggests a pro-survival function in these procedures such as in vitro or in vivo oocyte maturation can embryos (De Smaele et al. 2001). Further, GADD45 distinctly modify embryo, fetal and postnatal phenotypes. proteins promote active DNA demethylation and thus Moreover, this would suggest that the recipient uterine mediate gene activation, which is particularly important environment could facilitate supplementary changes to for cell differentiation and transcriptional regulation transcriptional signatures of transferred embryos, further during development (Schafer 2013). Given GADD45 compounding the effects of ART techniques on gene function in the stress response and role in age-related expression. Overall, these findings are in agreement

Reproduction (2017) 153 107–122 www.reproduction-online.org

Downloaded from Bioscientifica.com at 09/26/2021 04:56:31AM via free access Transcriptome of embryo correlates to manipulation 119 with the ‘quiet embryo’ hypothesis (Leese et al. 2008) changes are resolved during development. However, and argue strongly for avoiding extensive embryo the presence of distinct adult phenotypes (e.g. the manipulation whenever possible. manifestation of glucose intolerance in mouse IVF Another key finding of this analysis is that significant offspring) makes this scenario unlikely. Another cell-type transcriptional specificity of IVF is already explanation is that the molecular changes present observable at the blastocyst stage, where only 2 cell in blastocysts after preimplantation disturbance are types are present. In fact, comparison of blastocyst vs differentially affected by organogenesis, growth factors or ICM profiles derived from the same in vitro conditions sexually dimorphic signals occurring during later stages of showed an overlap of only 39 genes, 6 of which are development. As a result, cell physiology and metabolism present in the common fingerprint. Such nominal within each developing tissue is altered in accordance concordance demonstrates that individual cell types with new, tissue-specific developmental cues. are primed to differentially respond to environmental Fitting within this framework, we have previously perturbation, and that pooling multiple cell types may investigated whether specific genes altered in IVF embryos obscure investigations into DOHaD mechanisms. The exhibit tissue-specific maintenance of transcriptional relevance of cell type is further supported by the lack of a and epigenetic changes throughout development. We collective IVF fingerprint across the adult tissues, and the showed that expression of the glucose-sensitive gene evidence that pancreatic islets—a highly differentiated Txnip was significantly increased in IVF blastocysts, and complex tissue—exhibited fewer transcriptional and selectively increased in female IVF fat and muscle tissues pathway changes after IVF than other tissues. Interestingly, but not liver, and that this dysregulation was associated the adult transcriptional changes in IVF offspring were with enriched H4 acetylation at the Txnip promoter (Feuer minimal (over 95% of the gene misexpression was ≤2 et al. 2014). In the current study, Txnip is upregulated in fold-change different). This was true for tissues derived all ART blastocysts except for the ICSI and the IVC KAA- from optimized IVF KAA-5% oxygen conditions (islets, 20% oxygen conditions. In another study, we reported liver, fat and muscle) as well as for cardiac tissue that glucocorticoid receptor (GR) expression is increased from suboptimal IVF WM-20% O2 offspring. Other in IVF KAA-5% O2 embryos and male offspring fat, but models of developmental environmental perturbation not liver or muscle (Simbulan et al. 2015). have similarly reported only subtle and tissue-distinct Finally, it is possible that the gene expression changes transcriptional profiles. For example, investigation of in adult IVF tissues occur secondary to an altered the transgenerational effects of gestational vinclozolin function of key upstream transcriptional regulators, exposure demonstrated that F3 generation rats exhibited which would not necessarily be identifiable in a unique, tissue-specific changes to adult expression, with microarray profile. Transcriptional regulators are able to no common pathway overrepresentation (Skinner et al. control diverse processes with cell type and temporal 2012). Another study showed gene expression changes specificity through combinatorial interplay with other in ICSI but not IVF mouse neonates and no transcriptional transcriptional factors and modulators. In this manner, or phenotypic differences by 8 weeks of age (Kohda altering the function of a transcriptional regulator would et al. 2011). Given the lack of a common signature in exert a variety of different effects depending upon the adults, it is not surprising that the transcriptional changes cell type, which could explain the stark tissue specificity observed in manipulated embryos did not persist through observed in the ART expression profiles. Several of the development. proposed upstream regulators are highly relevant to The reasons for the tissue-specific variation and DOHaD phenotypes and provide grounds for directly the lack of stability of the IVF-induced transcriptional testable hypotheses in future investigations. CEBPA/B changes throughout in utero and postnatal development play essential and redundant functions during early are unknown. Because only 4–8 epiblast cells contribute embryogenesis and placentation (Begay et al. 2004), and to the adult body (Soriano & Jaenisch 1986, Morris & CEBPA enhances iPS cell reprogramming efficiency (Di Zernicka-Goetz 2012), persistence and penetrance of Stefano et al. 2014). CEBPA is crucial for liver and lung early gene expression patterns into adulthood could development, adipocyte terminal differentiation, the depend on which particular founder cells contributed establishment and maintenance of energy homeostasis, to organogenesis. Separately, the necessity of pooling lipid storage and gluconeogenesis—all through blastocysts to assess gene expression could have combinatorial interactions with other transcription captured embryos capable of adapting to a stressful factors such as Myc or members of the PPAR family in vitro environment but not necessarily able to develop (Wang et al. 1995, Flodby et al. 1996, Wu et al. 1999). beyond the blastocyst stage. This would introduce a Context-specific interactions might explain why some strong transcriptional bias in embryos compared to transcriptional regulators exhibited predicted activation postnatal tissues, although given the robust concordance in certain tissues and inhibition in others. Further, the in gene expression among embryos derived from different Cepba locus encodes a functional RNA that sequesters in vitro conditions, we believe this is an unlikely DMNT1 to ensure robust Cebpa transcription through possibility. It is also possible that some ART-induced local inhibition of Cepba gene methylation, thus www.reproduction-online.org Reproduction (2017) 153 107–122

Downloaded from Bioscientifica.com at 09/26/2021 04:56:31AM via free access 120 S Feuer and others directly linking Cepba activity to site-specific genomic how acute transcriptional responses to stressful methylation (Di Ruscio et al. 2013). environmental exposures might be preserved throughout Separately, NFE2L2 (also known as NRF2) is involved development to influence adult gene expression and in antioxidant defense mechanisms and binds DNA manifest metabolic phenotypes. Our results suggest that antioxidant response elements (AREs) to regulate the the activity of these regulators may be reliable markers adaptive response to oxidative stress (reviewed in Ma of in vitro stress present at specific developmental points 2013). NFE2L2 was predictedly inhibited in our ART extending into adulthood. datasets, which is particularly relevant because mice deficient in NFE2L2 exhibit increased sensitivity to oxidative stress. There is ample evidence in the literature Supplementary data that embryo culture induces an increase in reactive oxygen This is linked to the online version of the paper at http://dx.doi. species (ROS) (Goto et al. 1993, Cebral et al. 2007, Martin- org/10.1530/REP-16-0473. Romero et al. 2008), and ROS induction is a central event precipitating diabetes pathogenesis (Nishikawa et al. 2000, Sakai et al. 2003). ROS can also impair telomerase Declaration of interest activity, with telomere attrition being a predictive marker The authors declare that there is no conflict of interest that for cardiovascular dysfunction, metabolic syndrome and could be perceived as prejudicing the impartiality of the other age- and DOHaD-related pathologies (Hallows research reported. et al. 2012). It is therefore possible that an increase in ROS levels is the initial stimulus that reprograms the ART embryo, which employs transcription factors like Funding NFE2L2, CEBPs (Manea et al. 2014) and p53 to exert This work was supported by R01: HD 082039 and ASRM award acute transcriptional responses and subsequent chromatin to PFR. SKF was supported by the NIH training fellowship remodeling that persists throughout development, leading 5T32HD007263-32. to different cell type-contextual expression patterns. Among the potential caveats of this paper is the fact that the results presented are specific to data from References one laboratory and one species. Although this enables Banrezes B, Sainte-Beuve T, Canon E, Schultz RM, Cancela J & Ozil JP direct comparisons between datasets, small technical 2011 Adult body weight is programmed by a redox-regulated and discrepancies between different laboratories can affect energy-dependent process during the pronuclear stage in mouse. PLoS ONE 6 e29388. (doi:10.1371/journal.pone.0029388) the results and subsequent conclusions. Similarly, Begay V, Smink J & Leutz A 2004 Essential requirement of CCAAT/enhancer choice of mouse strain(s) can influence gene expression binding proteins in embryogenesis. Molecular and Cellular Biology 24 signatures (Turk et al. 2004). The mouse data provided 9744–9751. (doi:10.1128/MCB.24.22.9744-9751.2004) Bloise E, Lin W, Liu X, Simbulan R, Kolahi KS, Petraglia F, Maltepe E, may not apply to other species and thus should be Donjacour A & Rinaudo P 2012 Impaired placental nutrient transport in confirmed in, for example, bovine Smith( et al. 2009) mice generated by in vitro fertilization. Endocrinology 153 3457–3467. or sheep (Wei et al. 2016) models. Finally, given the (doi:10.1210/en.2011-1921) progressive maturation of translation machinery in the Bloise E, Feuer SK & Rinaudo PF 2014 Comparative intrauterine development and placental function of ART concepti: implications early embryo, changes in RNA may not be reflected by for human reproductive medicine and animal breeding. Human changes in protein abundance (Seydoux et al. 1996). Reproduction Update 20 822–839. (doi:10.1093/humupd/dmu032) To conclude, we have provided a systematic, integrated Cebral E, Carrasco I, Vantman D & Smith R 2007 Preimplantation embryotoxicity after mouse embryo exposition to reactive oxygen analysis of the acute and long-term transcriptional effects species. Biocell 31 51–59. of different mechanisms of in vitro embryo manipulation Chandel NS 2010 Mitochondrial regulation of oxygen sensing. Advances commonly used in assisted reproduction. The results of in Experimental Medicine and Biology 661 339–354. (doi:10.1007/978- this study indicate that the expression of 18 genes may 1-60761-500-2_22) Chen J, Saxena G, Mungrue IN, Lusis AJ & Shalev A 2008 Thioredoxin- be used to evaluate ART embryo health and could be interacting protein: a critical link between glucose toxicity and beta-cell developed into a new tool for identifying embryos with apoptosis. Diabetes 57 938–944. (doi:10.2337/db07-0715) the greatest implantation potential. Because fertilization Chronopoulou E & Harper JC 2015 IVF culture media: past, present and and embryo culture practices vary worldwide, this list future. Human Reproduction Update 21 39–55. (doi:10.1093/humupd/ dmu040) of 18 genes is valuable for its robustness and potential De Smaele E, Zazzeroni F, Papa S, Nguyen DU, Jin R, Jones J, Cong R application across the entire ART field. Thisin vitro & Franzoso G 2001 Induction of gadd45beta by NF-kappaB fingerprint does not persist throughout development, downregulates pro-apoptotic JNK signalling. Nature 414 308–313. (doi:10.1038/35104560) and there is no clear transcriptional cohesion between Delle Piane L, Lin W, Liu X, Donjacour A, Minasi P, Revelli A, Maltepe E & adult IVF offspring tissues compared to their preceding Rinaudo PF 2010 Effect of the method of conception and embryo transfer embryos. However, the unique transcriptional signatures procedure on mid-gestation placenta and fetal development in an IVF present in each IVF tissue are collectively targeted by mouse model. Human Reproduction 25 2039–2046. (doi:10.1093/ humrep/deq165) the same upstream transcriptional regulators, which Di Ruscio A, Ebralidze AK, Benoukraf T, Amabile G, Goff LA, Terragni J, provides insight into DOHaD mechanisms regarding Figueroa ME, De Figueiredo Pontes LL, Alberich-Jorda M et al. 2013

Reproduction (2017) 153 107–122 www.reproduction-online.org

Downloaded from Bioscientifica.com at 09/26/2021 04:56:31AM via free access Transcriptome of embryo correlates to manipulation 121

DNMT1-interacting RNAs block gene-specific DNA methylation. Nature Jorgensen SB, O’Neill HM, Sylow L, Honeyman J, Hewitt KA, Palanivel R, 503 371–376. (doi:10.1038/nature12598) Fullerton MD, Oberg L, Balendran A, Galic S et al. 2013 Deletion of Di Stefano B, Sardina JL, van Oevelen C, Collombet S, Kallin EM, skeletal muscle SOCS3 prevents insulin resistance in obesity. Diabetes Vicent GP, Lu J, Thieffry D, Beato M & Graf T 2014 C/EBPalpha poises 62 56–64. (doi:10.2337/db12-0443) B cells for rapid reprogramming into induced pluripotent stem cells. Jungheim ES, Schoeller EL, Marquard KL, Louden ED, Schaffer JE & Nature 506 235–239. (doi:10.1038/nature12885) Moley KH 2010 Diet-induced obesity model: abnormal oocytes and Doherty AS, Mann MR, Tremblay KD, Bartolomei MS & Schultz RM persistent growth abnormalities in the offspring. Endocrinology 151 2000 Differential effects of culture on imprinted H19 expression in the 4039–4046. (doi:10.1210/en.2010-0098) preimplantation mouse embryo. Biology of Reproduction 62 1526– Kocabas F, Zheng J, Thet S, Copeland NG, Jenkins NA, DeBerardinis RJ, 1535. (doi:10.1095/biolreprod62.6.1526) Zhang C & Sadek HA 2012 Meis1 regulates the metabolic phenotype Donjacour A, Liu X, Lin W, Simbulan R & Rinaudo PF 2014 In vitro and oxidant defense of hematopoietic stem cells. Blood 120 4963–4972. fertilization affects growth and glucose metabolism in a sex-specific (doi:10.1182/blood-2012-05-432260) manner in an outbred mouse model. Biology of Reproduction 90 80. Kohda T, Ogonuki N, Inoue K, Furuse T, Kaneda H, Suzuki T, Kaneko- (doi:10.1095/biolreprod.113.113134) Ishino T, Wakayama T, Wakana S, Ogura A et al. 2011 Intracytoplasmic ESHRE 2014, posting date. ART fact sheet. [Online.] sperm injection induces transcriptome perturbation without any Fernandez-Gonzalez R, Moreira P, Bilbao A, Jimenez A, Perez-Crespo M, transgenerational effect. Biochemical and Biophysical Research Ramirez MA, Rodriguez De Fonseca F, Pintado B & Gutierrez-Adan A Communications 410 282–288. (doi:10.1016/j.bbrc.2011.05.133) 2004 Long-term effect of in vitro culture of mouse embryos with serum Kuijk EW, du Puy L, van Tol HT, Haagsman HP, Colenbrander B & Roelen BA on mRNA expression of imprinting genes, development, and behavior. 2007 Validation of reference genes for quantitative RT-PCR studies in PNAS 101 5880–5885. (doi:10.1073/pnas.0308560101) porcine oocytes and preimplantation embryos. BMC Developmental Feuer SK, Liu X, Donjacour A, Lin W, Simbulan RK, Giritharan G, Piane LD, Biology 7 58. (doi:10.1186/1471-213X-7-58) Kolahi K, Ameri K, Maltepe E et al. 2014 Use of a mouse in vitro Kwong WY, Wild AE, Roberts P, Willis AC & Fleming TP 2000 Maternal fertilization model to understand the developmental origins of health undernutrition during the preimplantation period of rat development and disease hypothesis. Endocrinology 155 1956–1969. (doi:10.1210/ causes blastocyst abnormalities and programming of postnatal en.2013-2081) hypertension. Development 127 4195–4202. Flodby P, Barlow C, Kylefjord H, Ahrlund-Richter L & Xanthopoulos KG Leese HJ, Baumann CG, Brison DR, McEvoy TG & Sturmey RG 2008 1996 Increased hepatic cell proliferation and lung abnormalities in mice Metabolism of the viable mammalian embryo: quietness revisited. Molecular deficient in CCAAT/enhancer binding protein alpha. Journal of Biological Human Reproduction 14 667–672. (doi:10.1093/molehr/gan065) Chemistry 271 24753–24760. (doi:10.1074/jbc.271.40.24753) Li MW, Willis BJ, Griffey SM, Spearow JL & Lloyd KC 2009 Assessment Fortier AL, Lopes FL, Darricarrere N, Martel J & Trasler JM 2008 of three generations of mice derived by ICSI using freeze-dried sperm. Superovulation alters the expression of imprinted genes in the Zygote 17 239–251. (doi:10.1017/S0967199409005292) midgestation mouse placenta. Human Molecular Genetics 17 Ma Q 2013 Role of nrf2 in oxidative stress and toxicity. Annual Review 1653–1665. (doi:10.1093/hmg/ddn055) of Pharmacology and Toxicology 53 401–426. (doi:10.1146/annurev- Gillman MW 2005 Developmental origins of health and disease. pharmtox-011112-140320) New England Journal of Medicine 353 1848–1850. (doi:10.1056/ Maltepe E & Fisher SJ 2015 Placenta: the forgotten organ. Annual Review NEJMe058187) of Cell and Developmental Biology 31 523–552. (doi:10.1146/annurev- Giritharan G, Talbi S, Donjacour A, Di Sebastiano F, Dobson AT & cellbio-100814-125620) Rinaudo PF 2007 Effect of in vitro fertilization on gene expression and Manea SA, Todirita A, Raicu M & Manea A 2014 C/EBP transcription factors development of mouse preimplantation embryos. Reproduction 134 regulate NADPH oxidase in human aortic smooth muscle cells. Journal 63–72. (doi:10.1530/REP-06-0247) of Cellular and Molecular Medicine 18 1467–1477. (doi:10.1111/ Giritharan G, Li MW, De Sebastiano F, Esteban FJ, Horcajadas JA, jcmm.12289) Lloyd KC, Donjacour A, Maltepe E & Rinaudo PF 2010 Effect of ICSI on Martin-Romero FJ, Miguel-Lasobras EM, Dominguez-Arroyo JA, Gonzalez- gene expression and development of mouse preimplantation embryos. Carrera E & Alvarez IS 2008 Contribution of culture media to oxidative Human Reproduction 25 3012–3024. (doi:10.1093/humrep/deq266) stress and its effect on human oocytes. Reproductive BioMedicine Online Giritharan G, Delle Piane L, Donjacour A, Esteban FJ, Horcajadas JA, 17 652–661. (doi:10.1016/S1472-6483(10)60312-4) Maltepe E & Rinaudo P 2012 In vitro culture of mouse embryos Matthews PA, Samuelsson AM, Seed P, Pombo J, Oben JA, Poston L & reduces differential gene expression between inner cell mass Taylor PD 2011 Fostering in mice induces cardiovascular and metabolic and trophectoderm. Reproductive Sciences 19 243–252. dysfunction in adulthood. Journal of Physiology 589 3969–3981. (doi:10.1177/1933719111428522) (doi:10.1113/jphysiol.2011.212324) Gluckman PD, Hanson MA & Low FM 2011 The role of developmental Miao YL, Stein P, Jefferson WN, Padilla-Banks E & Williams CJ 2012 plasticity and epigenetics in human health. Birth Defects Research Part Calcium influx-mediated signaling is required for complete mouse egg C Embryo Today 93 12–18. (doi:10.1002/bdrc.20198) activation. PNAS 109 4169–4174. (doi:10.1073/pnas.1112333109) Goto Y, Noda Y, Mori T & Nakano M 1993 Increased generation of reactive Miller ME, Rosten P, Lemieux ME, Lai C & Humphries RK 2016 Meis1 oxygen species in embryos cultured in vitro. Free Radical Biology and is required for adult mouse erythropoiesis, megakaryopoiesis and Medicine 15 69–75. (doi:10.1016/0891-5849(93)90126-F) hematopoietic stem cell expansion. PLoS ONE 11 e0151584. Hager R, Cheverud JM & Wolf JB 2009 Change in maternal environment (doi:10.1371/journal.pone.0151584) induced by cross-fostering alters genetic and epigenetic effects Morris SA & Zernicka-Goetz M 2012 Formation of distinct cell types in on complex traits in mice. Proceedings of the Royal Society 276 the mouse blastocyst. Results and Problems in Cell Differentiation 55 2949–2954. (doi:10.1098/rspb.2009.0515) 203–217. Hallows SE, Regnault TR & Betts DH 2012 The long and short of it: the role Moskalev AA, Smit-McBride Z, Shaposhnikov MV, Plyusnina EN, of telomeres in fetal origins of adult disease. Journal of Pregnancy 2012 Zhavoronkov A, Budovsky A, Tacutu R & Fraifeld VE 2012 Gadd45 638476. (doi:10.1155/2012/638476) proteins: relevance to aging, longevity and age-related pathologies. Ho Y, Wigglesworth K, Eppig JJ & Schultz RM 1995 Preimplantation Ageing Research Reviews 11 51–66. (doi:10.1016/j.arr.2011.09.003) development of mouse embryos in KSOM: augmentation by amino Nishikawa T, Edelstein D, Du XL, Yamagishi S, Matsumura T, acids and analysis of gene expression. Molecular Reproduction and Kaneda Y, Yorek MA, Beebe D, Oates PJ, Hammes HP et al. 2000 Development 41 232–238. (doi:10.1002/mrd.1080410214) Normalizing mitochondrial superoxide production blocks three ICMART 2012 International Committee Monitoring Assisted Reproductive pathways of hyperglycaemic damage. Nature 404 787–790. Technology (ICMART) World Report: Preliminary 2008 Data. (doi:10.1038/35008121) Jeong YJ, Choi HW, Shin HS, Cui XS, Kim NH, Gerton GL & Jun JH 2005 Palanivel R, Fullerton MD, Galic S, Honeyman J, Hewitt KA, Jorgensen SB & Optimization of real time RT-PCR methods for the analysis of gene Steinberg GR 2012 Reduced Socs3 expression in adipose tissue protects expression in mouse eggs and preimplantation embryos. Molecular female mice against obesity-induced insulin resistance. Diabetologia 55 Reproduction and Development 71 284–289. (doi:10.1002/mrd.20269) 3083–3093. (doi:10.1007/s00125-012-2665-3) www.reproduction-online.org Reproduction (2017) 153 107–122

Downloaded from Bioscientifica.com at 09/26/2021 04:56:31AM via free access 122 S Feuer and others

Parikh H, Carlsson E, Chutkow WA, Johansson LE, Storgaard H, Poulsen P, glucocorticoid receptor expression in fat tissue. Journal of Developmental Saxena R, Ladd C, Schulze PC, Mazzini MJ et al. 2007 TXNIP regulates Origins of Health and Disease 1–10. peripheral glucose metabolism in humans. PLoS Medicine 4 e158. Simmons RA, Templeton LJ & Gertz SJ 2001 Intrauterine growth retardation (doi:10.1371/journal.pmed.0040158) leads to the development of type 2 diabetes in the rat. Diabetes 50 Patwari P & Lee RT 2012 An expanded family of arrestins regulate 2279–2286. (doi:10.2337/diabetes.50.10.2279) metabolism. Trends in Endocrinology and Metabolism 23 216–222. Simon MC, Liu L, Barnhart BC & Young RM 2008 Hypoxia-induced (doi:10.1016/j.tem.2012.03.003) signaling in the cardiovascular system. Annual Review of Physiology 70 Rappolee DA, Basilico C, Patel Y & Werb Z 1994 Expression and function 51–71. (doi:10.1146/annurev.physiol.70.113006.100526) of FGF-4 in peri-implantation development in mouse embryos. Skinner MK, Manikkam M, Haque MM, Zhang B & Savenkova MI 2012 Development 120 2259–2269. Epigenetic transgenerational inheritance of somatic transcriptomes and Reik W 2007 Stability and flexibility of epigenetic gene regulation epigenetic control regions. Genome Biology 13 R91. (doi:10.1186/gb- in mammalian development. Nature 447 425–432. (doi:10.1038/ 2012-13-10-r91) nature05918) Smith SL, Everts RE, Sung LY, Du F, Page RL, Henderson B, Rodriguez- Rexhaj E, Paoloni-Giacobino A, Rimoldi SF, Fuster DG, Anderegg M, Zas SL, Nedambale TL, Renard JP, Lewin HA et al. 2009 Gene expression Somm E, Bouillet E, Allemann Y, Sartori C & Scherrer U 2013 Mice profiling of single bovine embryos uncovers significant effects of generated by in vitro fertilization exhibit vascular dysfunction and in vitro maturation, fertilization and culture. Molecular Reproduction shortened life span. Journal of Clinical Investigation 123 5052–5060. and Development 76 38–47. (doi:10.1002/mrd.20927) (doi:10.1172/JCI68943) Soriano P & Jaenisch R 1986 Retroviruses as probes for mammalian Rinaudo P & Schultz RM 2004 Effects of embryo culture on global pattern development: allocation of cells to the somatic and germ cell lineages. of gene expression in preimplantation mouse embryos. Reproduction Cell 46 19–29. (doi:10.1016/0092-8674(86)90856-1) 128 301–311. (doi:10.1530/rep.1.00297) Turk R, t Hoen PA, Sterrenburg E, de Menezes RX, de Meijer EJ, Boer JM, Rinaudo PF, Giritharan G, Talbi S, Dobson AT & Schultz RM 2006 Effects of van Ommen GJ & den Dunnen JT 2004 Gene expression variation oxygen tension on gene expression in preimplantation mouse embryos. between mouse inbred strains. BMC Genomics 5 57. (doi:10.1186/1471- Fertility and Sterility 86 1252–1265, 1265.e1251–1236. 2164-5-57) Rivera RM, Stein P, Weaver JR, Mager J, Schultz RM & Bartolomei MS Valenzuela-Alcaraz B, Crispi F, Bijnens B, Cruz-Lemini M, Creus M, 2008 Manipulations of mouse embryos prior to implantation result in Sitges M, Bartrons J, Civico S, Balasch J & Gratacos E 2013 Assisted aberrant expression of imprinted genes on day 9.5 of development. reproductive technologies are associated with cardiovascular Human Molecular Genetics 17 1–14. (doi:10.1093/hmg/ddm280) remodeling in utero that persists postnatally. Circulation 128 Sakai K, Matsumoto K, Nishikawa T, Suefuji M, Nakamaru K, Hirashima Y, 1442–1450. (doi:10.1161/CIRCULATIONAHA.113.002428) Kawashima J, Shirotani T, Ichinose K, Brownlee M et al. 2003 Wang ND, Finegold MJ, Bradley A, Ou CN, Abdelsayed SV, Wilde MD, Mitochondrial reactive oxygen species reduce insulin secretion Taylor LR, Wilson DR & Darlington GJ 1995 Impaired energy by pancreatic beta-cells. Biochemical and Biophysical Research homeostasis in C/EBP alpha knockout mice. Science 269 1108–1112. Communications 300 216–222. (doi:10.1016/S0006-291X(02)02832-2) (doi:10.1126/science.7652557) Salvador JM, Brown-Clay JD & Fornace AJ Jr 2013 Gadd45 in stress Watkins AJ, Lucas ES, Torrens C, Cleal JK, Green L, Osmond C, Eckert JJ, signaling, cell cycle control, and apoptosis. Advances in Experimental Gray WP, Hanson MA & Fleming TP 2010 Maternal low-protein Medicine and Biology 793 1–19. diet during mouse pre-implantation development induces vascular Sato MS, Yoshitomo M, Mohri T & Miyazaki S 1999 Spatiotemporal dysfunction and altered renin-angiotensin-system homeostasis in the analysis of [Ca2+]i rises in mouse eggs after intracytoplasmic sperm offspring. British Journal of Nutrition 103 1762–1770. (doi:10.1017/ injection (ICSI). Cell Calcium 26 49–58. (doi:10.1054/ceca.1999.0053) S0007114509993783) Schafer A 2013 Gadd45 proteins: key players of repair-mediated DNA Watkins AJ, Lucas ES, Wilkins A, Cagampang FR & Fleming TP 2011 demethylation. Advances in Experimental Medicine and Biology 793 Maternal periconceptional and gestational low protein diet affects 35–50. (doi:10.1007/978-1-4614-8289-5_3) mouse offspring growth, cardiovascular and adipose phenotype at 1 year Schjenken JE & Robertson SA 2015 Seminal fluid signalling in the of age. PLoS ONE 6 e28745. (doi:10.1371/journal.pone.0028745) female reproductive tract: implications for reproductive success and Wei X, Xiaoling Z, Kai M, Rui W, Jing X, Min G, Zhonghong W, Jianhui T, offspring health. Advances in Experimental Medicine and Biology 868 Xinyu Z & Lei A 2016 Characterization and comparative analyses of 127–158. (doi:10.1007/978-3-319-18881-2_6) transcriptomes for in vivo and in vitro produced peri-implantation Schulze PC, Yoshioka J, Takahashi T, He Z, King GL & Lee RT 2004 conceptuses and endometria from sheep. Journal of Reproduction and Hyperglycemia promotes oxidative stress through inhibition of Development 62 279–287. (doi:10.1262/jrd.2015-064) thioredoxin function by thioredoxin-interacting protein. Journal of Whitten WK 1971 Nutrient requirements for the culture of preimplantation Biological Chemistry 279 30369–30374. (doi:10.1074/jbc.M400549200) embryos in vitro. Advances in Bioscience 6 129–141. (doi:10.1016/ Schwarzer C, Esteves TC, Arauzo-Bravo MJ, Le Gac S, Nordhoff V, Schlatt S b978-0-08-017571-3.50013-9) & Boiani M 2012 ART culture conditions change the probability of mouse Wu Z, Rosen ED, Brun R, Hauser S, Adelmant G, Troy AE, McKeon C, embryo gestation through defined cellular and molecular responses. Darlington GJ & Spiegelman BM 1999 Cross-regulation of C/EBP alpha Human Reproduction 27 2627–2640. (doi:10.1093/humrep/des223) and PPAR gamma controls the transcriptional pathway of adipogenesis Scott KA, Yamazaki Y, Yamamoto M, Lin Y, Melhorn SJ, Krause EG, and insulin sensitivity. Molecular Cell 3 151–158. (doi:10.1016/S1097- Woods SC, Yanagimachi R, Sakai RR & Tamashiro KL 2010 Glucose 2765(00)80306-8) parameters are altered in mouse offspring produced by assisted Zhou J & Chng WJ 2013 Roles of thioredoxin binding protein (TXNIP) reproductive technologies and somatic cell nuclear transfer. Biology of in oxidative stress, apoptosis and cancer. Mitochondrion 13 163–169. Reproduction 83 220–227. (doi:10.1095/biolreprod.109.082826) (doi:10.1016/j.mito.2012.06.004) Semenza GL 2016 Dynamic regulation of stem cell specification and Zhuang J, Ma W, Lago CU & Hwang PM 2012 Metabolic regulation of maintenance by hypoxia-inducible factors. Molecular Aspects of oxygen and redox homeostasis by p53: lessons from evolutionary Medicine 47–48 15–23. (doi:10.1016/j.mam.2015.09.004) biology? Free Radical Biology and Medicine 53 1279–1285. Seydoux G, Mello CC, Pettitt J, Wood WB, Priess JR & Fire A 1996 (doi:10.1016/j.freeradbiomed.2012.07.026) Repression of gene expression in the embryonic germ lineage of C. elegans. Nature 382 713–716. (doi:10.1038/382713a0) Shalev A 2008 Lack of TXNIP protects beta-cells against glucotoxicity. Biochemical Society Transactions 36 963–965. (doi:10.1042/BST0360963) Shang F & Taylor A 2011 Ubiquitin-proteasome pathway and cellular Received 3 June 2016 responses to oxidative stress. Free Radical Biology and Medicine 51 First decision 4 August 2016 5–16. (doi:10.1016/j.freeradbiomed.2011.03.031) Simbulan RK, Liu X, Feuer SK, Maltepe E, Donjacour A & Rinaudo P 2015 Revised manuscript received 19 October 2016 Adult male mice conceived by in vitro fertilization exhibit increased Accepted 26 October 2016

Reproduction (2017) 153 107–122 www.reproduction-online.org

Downloaded from Bioscientifica.com at 09/26/2021 04:56:31AM via free access