Accepted Manuscript Title: A transgenic zebrafish model for monitoring xbp1 splicing and endoplasmic reticulum stress in vivo Author: Junling Li, Zhiliang Chen, Angelo Colorni, Michal Ucko, Shengyun Fang, Shao Jun Du PII: DOI: Reference:

S0925-4773(15)00039-8 http://dx.doi.org/doi:10.1016/j.mod.2015.04.001 MOD 3338

To appear in:

Mechanisms of Development

Received date: Revised date: Accepted date:

6-8-2014 7-4-2015 8-4-2015

Please cite this article as: Junling Li, Zhiliang Chen, Angelo Colorni, Michal Ucko, Shengyun Fang, Shao Jun Du, A transgenic zebrafish model for monitoring xbp1 splicing and endoplasmic reticulum stress in vivo, Mechanisms of Development (2015), http://dx.doi.org/doi:10.1016/j.mod.2015.04.001. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

A transgenic zebrafish model for monitoring xbp1 splicing and endoplasmic reticulum stress in vivo Junling Li1, 2, Zhiliang Chen1, 3, Angelo Colorni5, Michal Ucko5, Shengyun Fang*3, and Shao Jun Du*1

1. Institute of Marine and Environmental Technology, Department of Biochemistry and Molecular Biology, University of Maryland School of Medicine, Baltimore, MD 21201, USA 2. Shandong Medicinal and Biotechnology Center, Shandong Academy of Medical Sciences, Jinan 250062, Shandong Province, P. R. China 3. Center for Biomedical Engineering and Technology, Department of Physiology, University of Maryland School of Medicine, Baltimore, MD 21201, USA 4. Department of cell Biology and Molecular genetics, University of Maryland, College Park, MD 20742, USA. 5. Israel Oceanographic and Limnological Research, National Center for Mariculture, Eilat 88112, Israel *

Corresponding Authors:

Shao Jun Du, Ph.D. Institute of Marine and Environmental Technology Department of Biochemistry and Molecular Biology University of Maryland School of Medicine 701 East Pratt Street Baltimore, MD 21202 USA Tel: 410-234-8854 Fax: 410-234-8896 Email: [email protected] OR Shengyun Fang, MD, Ph.D. Center for Biomedical Engineering and Technology Department of Physiology University of Maryland School of Medicine Baltimore, MD 21201 USA Tel: 410-706-2220 1 Page 1 of 29

Email: [email protected] Short title: xbp1 splicing in zebrafish embryos Keywords: XBP1; endoplasmic reticulum, ER stress; transgenic zebrafish model

2 Page 2 of 29

Highlights 1. A transgenic zebrafish model is generated that expresses a XBP1Δ-GFP fusion protein in response to ER stress. 2. A strong ER stress is detected in zebrafish oocytes, fertilized eggs and embryos. 3. XBP1Δ-GFP expression could be induced by DTT and tunicamycin in transgenic embryos. 4. The transgenic zebrafish provides a useful tool for monitoring xbp1 splicing ad ER stress in vivo.

Graphical Abstract

Abstract Accumulation of misfolded or unfolded proteins in the endoplasmic reticulum (ER) triggers ER stress that initiates unfolded protein response (UPR).

XBP1 is a transcription

factor that mediates one of the key signaling pathway of UPR to cope with ER stress 3 Page 3 of 29

through regulating gene expression.

Activation of XBP1 involves an unconventional

mRNA splicing catalyzed by IRE1 endonuclease that removes an internal 26 nucleotides from xbp1 mRNA transcripts in the cytoplasm.

Researchers have taken advantage of this

unique activation mechanism to monitor XBP1 activation, thereby UPR, in cell culture and transgenic models. Here we report a Tg(ef1:xbp1δ-gfp) transgenic zebrafish line to monitor XBP1 activation using GFP as a reporter especially in zebrafish oocytes and developing embryos. The Tg(ef1:xbp1δ-gfp) transgene was constructed using part of the zebrafish xbp1 cDNA containing the splicing element. ER stress induced splicing results in the cDNA encoding a GFP-tagged partial XBP1 without the transactivation activation domain (XBP1-GFP). The results showed that xbp1 transcripts mainly exist as the spliced activate isoform in unfertilized oocytes and zebrafish embryos prior to zygotic gene activation at 3 hours post fertilization.

A strong GFP expression was observed in unfertilized oocytes, eyes, brain

and skeletal muscle in addition to a weak expression in the hatching gland.

Incubation of

transgenic zebrafish embryos with (dithiothreitol) DTT significantly induced XBP1Δ-GFP expression. Collectively, these studies unveil the presence of maternal xbp1 splicing in zebrafish oocytes, fertilized eggs and early stage embryos. The tg(ef1:xbp1δ-gfp) transgenic zebrafish provides a useful model for in vivo monitoring xbp1 splicing during development and under ER stress conditions.

1. Introduction ER is responsible for synthesis, post-translational modification, and folding of about one third of total cellular proteins that function in the secretory pathway or extracellular space (Federovitch et al., 2005).

To maintain ER homeostasis, cells utilize protective 4 Page 4 of 29

mechanisms, such as chaperones and ER-associated degradation (ERAD), to remove misfolded proteins, thereby preventing protein from accumulation in the ER (Tsai et al., 2002; Stolz and Wolf , 2010; Brodsky et al., 2014).

However, once protein accumulation

exceeds the capacity of ERAD, it induces ER stress that activates a conserved signaling cascade called the “unfolded protein response” (UPR) to restore ER homeostasis (Cox and Walter, 1996; Travers et al., 2000; Gardner et al., 2013; Maly and Papa, 2014). Over the last decade, it has become clear that ER stress contributes to a number of diseases such as Alzheimer’s and Parkinson’s diseases, diabetes, inflammation, cancer, as well as ageing related illnesses (Marciniak and Ron, 2006; Kapoor and Sanyal, 2009; Healy et al., 2009; Homma et al., 2009; Glembotski, 2011; Back and Kaufman, 2012; Imrie and Sadler, 2012; Wang and Kaufman, 2012; Iwawaki and Oikawa, 2013; Roussel et al., 2013; Wang and Kaufman, 2014; Yadav et al., 2014; Dandekar et al., 2015; Dicks et al., 2015). Homeostasis of the ER is regulated by UPR through a complex signaling system emanating from the ER membrane that regulates gene expression in response to increased demands on the protein folding capacity of the ER (Rutkowski and Kaufman, 2004). UPR is mediated by three ER stress sensors that are single ER transmembrane proteins including inositol-requiring enzyme 1 (IRE1), activating transcription factor 6 (ATF6) and pancreatic eIF-2 kinase (PERK) (Ron and Walter, 2007; Walter and Ron, 2011; Wang and Kaufman, 2012).

Under non-stress physiologic conditions, the intraluminal domains

of the stress sensors are associated with the chaperone BiP/grp78, which keeps the stress sensors in their inactive states. When misfolded proteins accumulate in the ER, they compete with the stress sensors to bind with BiP/grp78, thereby releasing the suppression and activating UPR (Ron and Walter, 2007; Walter and Ron, 2011; Wang and Kaufman, 2012). IRE1 is a unique RNase which catalyses a novo splicing of mRNA transcripts encoding X-box binding protein 1 (XBP1), a potent transcription factor that activates gene expression involved in protein folding and degradation (Shen et al., 2001; Yoshida et al., 2001). In response to ER stress, IRE1 dissociates from BiP/grp78 and becomes activated through auto phosphorylation.

The activated IRE1 initiates an unconventional splicing 5 Page 5 of 29

reaction that removes a 26-nucleotide IRE1 target sequence within the xbp1 mRNA transcripts. Removal of the 26 nt intron creates a reading frame shift that allows translation of a longer XBP1 transcription factor that activates gene expression involved in ER stress response (Shen et al., 2001; Yoshida et al., 2001). The unconventional splicing of xbp1 mRNA by IRE1 is a highly specific response to ER stress and appears to be conserved during evolution. The IRE1-catalyzed xbp1 mRNA splicing has been utilized to monitor ER stress in cultured mammalian cells and xbp1δ-gfp transgenic animal models including the free living nematode Caenorhabditis elegans, Drosophila and mice (Iwawaki et al., 2004; Shim et al., 2004; Ryoo et al., 2007; Souid et al., 2007).

These transgenic models have been successfully used to study drug and

genetic mutation induced xbp1 splicing in adult live animals (Iwawaki et al., 2004; Ryoo et al., 2007; Spiotto et al., 2010).

The xbp1δ-gfp constructs used in early studies lacked

the C-terminal coding region and the 3’UTR sequence, and did not completely mimic the splicing of the endogenous xbp1. An improved stress sensing systems has recently been generated by including the full length coding sequence and the 3’UTR sequence in the xbp1δ-gfp reporter construct (Sone et al., 2013). This improved system reveals a novel tissue distribution of IRE1/XBP1 activity during normal Drosophila development (Sone et al., 2013). A strong XBP1Δ-GFP expression is clearly detected in secretary tissues, such as salivary gland, intestine and the male reproductive system in the transgenic Drosophila, mimicking the endogenous gene (Sone et al., 2013). The unconventional splicing of xbp1 has also been characterized in zebrafish during development and UPR in fatty liver diseases (Bennett et al., 2007; Thakur et al, 2011, 2014; Monk et al., 2013, Tsedensodnom et al., 2013; Chen et al., 2014; Vacaru et al., 2014). xbp1 splicing has yet to be analyzed in live zebrafish embryos and larvae. There is a need for a zebrafish model that could be used to monitor xbp1 splicing in vivo and in real time during the early embryonic development.

To characterize xbp1 splicing

during normal development in vertebrates, we characterized xbp1 mRNA transcripts in zebrafish embryos by RT-PCR.

The results showed that unfertilized eggs and early stage

zebrafish embryos expressed high levels of spliced isoform of xbp1 transcripts.

To

6 Page 6 of 29

monitor xbp1 splicing in vivo, we produced a Tg(ef1:xbp1δ-gfp) transgenic zebrafish model. XBP1Δ-GFP expression was clearly detected in oocytes and fertilized eggs of Tg(ef1:xbp1δ-gfp) transgenic zebrafish, suggesting a maternal expression of the spliced isoform of xbp1.

Incubating fish embryos with ER stressing reagents, DTT or

tunicamycin, significantly induced XBP1Δ-GFP expression and xbp1 splicing. Together, these data demonstrated that there is a significant xbp1 splicing in developing oocytes. The Tg(ef1:xbp1δ-gfp) transgenic zebrafish provides a useful model for monitoring xbp1 splicing in vivo and ER stress in response to chemical reagents.

7 Page 7 of 29

2. Results 2.1. Characterization of zebrafish xbp1 gene and its alternatively spliced mRNA transcripts The zebrafish xbp1 gene is located at chromosome five. introns with a size of approximately 5kb (Fig.1A).

It contains 5 exons and 4

A conserved ER stress responsive

element (CCAAT---N9----CCACG ; GGTTA---N9----GGTGC) was identified at the proximal promoter region (Fig. 1A). The zebrafish xbp1 gene encodes two xbp1 mRNA transcripts that differ by a 26 nt IRE1 target sequence located at position 505-530 from the ATG start codon (Fig. 1A).

The xbp1u represents the unspliced isoform which encodes a

protein of 262 amino acids (aa). Under ER stress conditions, removal of the 26 nt sequence by IRE1 results in the production of a spliced xbp1s isoform which encodes a longer and more potent XBP1 protein of 383 aa.

The first 208 aa N-terminal sequences

were identical between the two XBP1 isoforms. Close examination of the 26 bp IRE1 target sequence revealed two 9 nt direct repeats (TCCGCAGCA) in this region (Fig. 1B). The first repeat is located precisely at the 5' flanking region of the 26 bp IRE1 target sequence. end of the 26 bp sequence.

The second repeat is located at the 3'

Interestingly, the first repeat is retained in the spliced isoform

of the xbp1 transcripts (Fig. 1B). Comparison with xbp1 transcripts of other species revealed that the direct repeats and the 26 bp IRE1 target sequence are highly conserved during evolution (Fig. 1B). The significance of the direct repeat at the IRE1 target sequence is unknown. Direct repeats are often found in the sites of DNA recombination. However, it is not clear whether the 9 nt direct repeats are involved in the IRE1 target sequence splicing in xbp1 transcripts.

2.2. Characterization of xbp1 gene expression and splicing during embryonic development To characterize the expression pattern of these two xbp1 isoforms and to study ER stress during development, we analyzed their expression in zebrafish embryos by RT-PCR. The results showed that zebrafish embryos express both xbp1u and xbp1s mRNA 8 Page 8 of 29

transcripts during the early development. The xbp1u and xbp1s expression appears to be dynamic.

The xbp1u and xbp1s transcripts were clearly detected in early stage embryos

from fertilization to 6 hpf (Fig. 1C).

Interestingly, the xbp1s transcripts represented the

dominant isoform in newly fertilized eggs and early stage embryos prior to zygotic gene activation at 3 hpf (Fig. 1C). These data suggest that the xbp1s transcripts were likely derived maternally, because zygotic gene transcription does not start until after midblustula transition (around 3 hpf) in zebrafish embryos (Kane and Kimmel, 1993). xbp1 splicing stress appeared to be significantly down regulated in fish embryos between 9-48 hpf during early development because very little xbp1s transcripts could be detected from 9-48 hpf (Fig. 1C).

Collectively, these data indicate that there is a significant xbp1

splicing in fertilized eggs and early stage embryos.

2.3. Generation of Tg(ef1:xbp1δ-gfp) transgenic zebrafish model for monitoring xbp1 splicing in live zebrafish embryos in vivo. To characterize xbp1 splicing during normal development in live fish embryos, we developed a transgenic zebrafish line expressing XBP1Δ-GFP as an ER stress reporter as previously described in mice and Drosophila.

The Tg(ef1:xbp1δ-gfp) transgene was

constructed by cloning a partial xbp1u coding sequence (373-611) upstream of the EGFP coding sequence (Iwawaki et al., 2004) (Fig. 2A).

The partial XBP1 sequence contained

the 26 nt IRE1 target site but lacked the DNA binding domain required for XBP1 biological function. Under non-stress conditions, the 26-nt IRE1 target sequence in the xbp1δ-gfp fusion gene transcripts could not be spliced out, and consequently no GFP could be expressed. Removal of the IRE1 target sequence by splicing in the xbp1δ-gfp reporter transcripts results in a reading frame shift in xbp1δ-gfp transcripts leading to the expression of the XBP1Δ-GFP fusion protein (Fig. 2A). Three germ line transgenic founders were identified based on the XBP1Δ-GFP expression in their F1 embryos.

Strikingly, a dramatic difference in XBP1Δ-GFP

expression was observed in F1 transgenic fish embryos depending on whether the Tg(ef1:xbp1δ-gfp) transgene was derived maternally or paternally. As shown in Fig. 2B, 9 Page 9 of 29

a strong XBP1Δ-GFP expression was observed in newly fertilized eggs at 1-cell stage and early stage fish embryos if the transgene was derived maternally (Fig. 2B). In contrast to the maternal transgenic line, very little or no XBP1Δ-GFP expression was detected in zebrafish embryos from one cell stage to tail bud stage if the transgene was derived paternally (Fig. 2B).

At 24 hpf, GFP expression appeared in skeletal muscles of

transgenic fish embryos carrying the paternally derived transgene (Fig. 2B).

These data

indicate that xbp1δ-gfp transcripts are likely spliced maternally in transgenic zebrafish oocytes, and zygotic activation of the xbp1δ-gfp splicing starts between 12-24 hpf.

2.4. XBP1 splicing and ER stress in zebrafish oocytes and embryos To further evaluate the distinct pattern of XBP1Δ-GFP expression from the maternal or paternal Tg(ef1a:xbp1δ-gfp) transgene, we characterized the Tg(ef1a:xbp1δ-gfp) splicing by RT-PCR in various stage embryos from fertilization to 48 hpf (Fig. 3). A dramatic difference in xbp1δ-gfp mRNA expression and splicing was observed between maternally or paternally derived Tg(ef1a:xbp1δ-gfp) transgene. In the paternal transgenic line, there was no expression of the Tg(ef1a:xbp1δ-gfp) transgene before the zygotic gene activation at 3 hpf (Fig. 3A), although the endogenous xbp1 gene is strongly expressed (Fig. 3B). The xbp1δ-gfp transcripts primarily existed as the unspliced isoform between 6-12 hpf in the paternal line (Fig. 3A). The spliced isoform appeared at 24 hpf in the paternal transgenic fish embryos (Fig. 3A). In contrast, a robust expression and splicing of the Tg(ef1a: xbp1δ-gfp) transgene was observed between fertilization and 6 hpf in the maternal transgenic line (Fig. 3C), similar to the pattern of endogens xbp1 gene expression and splicing (Fig. 3D). Results from the RT-PCR analyses are consistent the idea that the maternal or paternal Tg(ef1a: xbp1δ-gfp) transgene exhibits distinct pattern of expression and splicing in early stage zebrafish embryos.

It should be noted that the xbp1 splicing

from the endogenous gene was very weak at 24 and 48 hpf (Fig. 3B, D).

In contrast,

xbp1 splicing from the xbp1δ-gfp transgene appeared stronger than the endogenous gene at 24 and 48 hpf regardless the transgene was derived paternally or maternally (Fig. 3A, B). This distinction between the endogenous and exogenous xbp1 splicing in 24 and 48 hpf 10 Page 10 of 29

zebrafish embryos could be caused by their differences in splicing efficiency or mRNA transcripts stability. To determine the XBP1Δ-GFP expression in zebrafish oocytes, zebrafish oocytes were dissected from the gonad of a Tg(ef1a:xbp1δ-gfp) transgenic female.

A strong

XBP1Δ-GFP fluorescence was clearly observed in the oocytes from the transgenic female (Fig. 4A) compared with oocytes from a non-transgenic fish (Fig. 4B).

To test directly

the xbp1 splicing in zebrafish oocytes, we analyzed the expression and splicing of xbp1 transcripts from the endogenous gene in oocytes, and compare with the results from testis and developing embryos. The results showed that zebrafish oocytes mainly expressed the spliced isoform of xbp1s (Fig. 4C).

In contrast, the testis mainly expressed the unspliced

isoform (Fig. 4C). Moreover, expression of BiP/grp78, another ER stress marker, was clearly detected in the zebrafish oocytes but not in testis (Fig. 4C).

Collectively, these

data argue the presence of a strong xbp1 splicing in developing zebrafish oocytes,

2.5. Characterization of Tg(ef1a:xbp1δ-gfp) transgene expression in fish embryos and larvae To better characterize the expression of the Tg(ef1a:xbp1δ-gfp) transgene in developing zebrafish embryos and larvae, homozygous transgenic fish embryos were generated by crossing a transgenic male with a transgenic female.

XBP1Δ-GFP

expression was analyzed in homozygous transgenic embryos and larvae at day 1, 3, and 6 post-fertilization. A strong XBP1Δ-GFP expression was observed in the eyes, brain, skeletal muscle and hatching gland of zebrafish embryos at 24 hpf (Fig. 5A, B). The GFP expression continued to be strong in skeletal muscles at 3 and 6 days post fertilization (dpf) (Fig. 5C, H). To confirm the expression and splicing of the Tg(ef1a:xbp1δ-gfp) transgene in late stage larvae, we analyzed the xbp1δ-gfp transcripts by RT-PCR at 3, 5 and 7 dpf. xbp1δ-gfp expression and splicing were clearly detected upto 7 dpf (Fig. 5I). It has been reported that the elongation factor 1 promoter used to drive the Tg(ef1a:xbp1δ-gfp) transgene expression became less active in certain adult tissues (Thummel et al., 2006).

However, the reduced ef1 promoter activity in adults showed 11 Page 11 of 29

renewed expression after injury (Thummel et al., 2006). To determine whether the Tg(ef1a:xbp1δ-gfp) transgenic line would be a useful injury model, we analyzed GFP expression during fin regeneration at various days after the caudal fin amputation.

The

results showed no detectable GFP expression in the blastema during fin regeneration (Fig. 1S, Supplemental information), suggesting that fin regeneration did not cause ER stress.

2.6. Monitoring ER stress in vivo under stress conditions To determine whether the Tg(ef1a:xbp1δ-gfp) transgenic zebrafish could be used as an in vivo model for monitoring ER stress.

We treated transgenic zebrafish embryos with

Dithiothreitol (DTT), a known reducing reagent that can cause protein misfolding and ER stress.

To eliminate the effect from maternal XBP1Δ-GFP expression, paternal

transgenic embryos were used in the DTT treatment. Compared with the non-treated control (Fig. 6B, F), GFP signal was increased in the DTT treated embryos (Fig. 6C, G). To test whether upregulation of XBP1Δ-GFP expression could be induced by tunicamycin, another ES stress inducer, the Tg(ef1a:xbp1δ-gfp) transgenic fish embryos were treated with tunicamycin between 24-48 hpf.

A small increase of XBP1Δ-GFP expression was

observed for the Tg(ef1a:xbp1δ-gfp) transgene after 24 h of tunicamycin treatment between 24-48 hpf (Fig. 6D, H). To confirm that XBP1Δ-GFP expression was indeed increased in fish embryos treated with DTT or tunicamycin, we analyzed the splicing and expression of the Tg(ef1a:xbp1δ-gfp) transgene by RT-PCR and western blot. The results showed that DTT treatment significantly increased the splicing of the Tg(ef1a:xbp1δ-gfp) transgene and the endogenous xbp1 gene (Fig. 6I), as well as the expression of XBP1Δ-GFP fusion protein (Fig. 6J). In contrast, tunicamycin treatment resulted in only a small increase of xbp1 splicing for both the endogenous gene and the transgene (Fig. 6K), consistent with the weak GFP induction by tunicamycin (Fig. 6D).

To confirm that DTT treatment

indeed induced ER stress, we analyzed the expression of BIP, another ER stress marker, in DTT treated embryos. embryos (Fig. 6L).

An increased BIP expression was detected in the DTT treated

Collectively, these data indicate that Tg(ef1:xbp1δ-gfp) transgenic 12 Page 12 of 29

zebrafish provide a useful model for monitoring ER stress in vivo.

13 Page 13 of 29

3. Discussion In this study, we developed a Tg(ef1a:xbp1δ-gfp) transgenic zebrafish model to monitor xbp1 splicing and ER stress in vivo. We demonstrated that the IRE1/XBP1 pathway is highly active in zebrafish oocytes and early stage embryos. This is supported by the strong XBP1Δ-GFP expression in unfertilized eggs from Tg(ef1a:xbp1δ-gfp) transgenic females and the predominant presence of the spliced xbp1s isoform in unfertilized zebrafish oocytes.

The IRE1/XBP1 pathway is active in fertilized eggs and

early stage embryos prior to 6 hpf. However, it is down regulated between 9-48 hpf analyzed.

The Tg(ef1a:xbp1δ-gfp) transgenic zebrafish provides a useful model for

monitoring chemically induced ER stress in vivo. Incubating fish embryos with DTT significantly induced ER stress and XBP1Δ-GFP expression. Together, these studies uncovered the profound xbp1 splicing in oocytes and developing zebrafish embryos, and established the power of transgenic zebrafish model for monitoring ER stress in vivo.

3.1. xbp1 splicing in oocytes and developing embryos Expression analyses revealed that xbp1 mRNA transcripts were strongly expressed in unfertilized oocytes and early stage zebrafish embryos. This is consistent with previous reports that XBP1 is expressed in oocytes and preimplantation embryos of porcine and mice (Zhang et al., 2012a; 2012b). The presence of the spliced xbp1s isoform as the predominant form in oocytes could be due to the strong demand for protein synthesis and secretion of the oocytes. However, it cannot rule out the possibility of ER stress in zebrafish oocytes during normal physiological conditions. It has been shown that ER stress is a key mechanism mediating fatty acid-induced defects in oocyte developmental potential (Wu et al., 2012).

Inhibiting ER stress enhances oocyte maturation and

embryonic development by preventing ER stress-mediated apoptosis (Zhang et al., 2012a; 2012b). Our studies revealed that zebrafish and mice exhibit a significant difference in the expression of spliced xbp1s in one-cell embryos.

Zebrafish show a predominant

expression of the spliced xbp1s isoform in one-cell embryos whereas mice and porcine 14 Page 14 of 29

one-cell embryos express only the unspliced isoform (Zhang et al., 2012a). The xbp1s transcripts in zebrafish embryos are most likely derived maternally from zebrafish oocytes. In contrast, in mice and porcine oocytes, the maternally spliced xbp1s transcripts are not passed to one-cell embryos although xbp1s is produced maternally in their oocytes at the germinal vesicle stage (Zhang et al., 2012a; 2012b). The spliced xbp1s isoform only become detectable in two-cell mice and porcine embryos, coinciding with the time of major embryonic genome activation. It is not clear why fish and mouse one-cell embryos show different pattern of xbp1 splicing in fertilized eggs at the one cell stage.

It could

reflect their differences in egg size and fertilization process that may have different impact on ER stress. Compared with zebrafish oocytes that are close to 1 mm in size and fertilize externally, mouse oocytes are much smaller and fertilize internally.

Mouse oocytes and

fertilized eggs may be exposed to less stress compared with zebrafish oocytes and fertilized eggs.

Interestingly, stressing mouse one-cell embryos with tunicamycin or

sorbitol in vitro resulted in ER stress induced xbp1 splicing (Zhang et al., 2012a), suggesting that IRE1/XBP1 pathway could be activated by an external stress inducer in mouse one-cell embryos. It has been demonstrated that XBP1 is essential for early embryonic development in Xenopus and mice (Reimold et al., 2000; Zhao et al., 2003; Cao et al., 2006). Complete loss of xbp1 through targeted ablation in mice results in embryonic lethality by embryonic day (E) 14.5. Embryonic lethality is attributed to the development of hypoplastic fetal livers resulting in reduced hematopoiesis and anemia (Zhao et al., 2003). In contrast, knockdown of xbp1 in zebrafish does not lead to abnormal development and embryonic lethality upto 6 days post fertilization (Bennett et al., 2007).

However, the knockdown

study in zebrafish has demonstrated that XBP1 is required for terminal differentiation of hatching gland (Bennett et al., 2007). xbp1 knockdown zebrafish embryos appeared normal but could not hatch (Bennett et al., 2007). Hatching gland contains secretory cells that are responsible for secreting hatching enzyme (chorionase) for solubilization of egg chorion. Consistent with its function in hatching gland development, the Tg(ef1a:xbp1δ-gfp) transgene is expressed in the hatching gland of zebrafish embryos. 15 Page 15 of 29

Previous studies with the xbp1δ-gfp transgenic Drosophila and mice showed a strong xbp1 and XBP1Δ-GFP expression in secretary tissues, such as salivary gland, pancreas, intestine and the male reproductive system (Iwawaki et al., 2004;Ryoo et al., 2007; Sone et al., 2013). In addition, xbp1s splicing and XBP1Δ-GFP expression were also detected in skeletal muscles in mice (Iwawaki et al., 2004).

We demonstrated here a strong

XBP1Δ-GFP expression in the brain, eyes and skeletal muscles of the Tg(ef1a:xbp1δ-gfp) transgenic fish embryos at 24 and 48 hpf.

The strong XBP1Δ-GFP expression in skeletal

muscles is consistent with the pattern of xbp1 expression in these tissues (Thisse et al., 2004).

In situ expression analysis also demonstrated that xbp1 is strongly expressed in

developing pancreas of zebrafish embryos (Thisse et al., 2004). This is consistent with the function of pancreas as an endocrine and exocrine gland producing several important hormones, such as insulin, and secreting digestive enzymes. However, the XBP1Δ-GFP expression was weak in the pancreas of transgenic zebrafish embryos. to the use of a foreign gene promoter from the ef1a gene.

This could be due

It has been reported that the

elongation factor 1 promoter used to drive the Tg(ef1a:xbp1δ-gfp) transgene expression became less active in certain adult tissues (Thummel et al., 2006), preventing the further characterization of XB1-GFP expression and ER stress in the pancreas of adult Tg(ef1a:xbp1δ-gfp) transgenic zebrafish. A recent study in Drosophila demonstrated that the full length XBP1 fused with GFP is a more sensitive ER stress indicator compared with the truncated XBP1 (Sone et al., 2013). We demonstrated in this study that zebrafish embryos expressing the truncated XBP1Δ-GFP reporter was very active in unfertilized eggs and early stage fish embryos. It is difficult to compare the data from these two studies because different promoters and expression systems were used in the zebrafish and fly studies. The Drosophila study utilized the Gal4-UAS system whereas the zebrafish study used the elongation factor 1 promoter directly to express the XBP1Δ-GFP reporter. It remains to be determined whether the full length XBP1 fused with GFP may serve as a more sensitive ER stress indicator in zebrafish.

16 Page 16 of 29

3.2. The potential application of the Tg(ef1a:xbp1δ-gfp) transgenic model in monitoring ER stress Although similar mouse and Drosophila xbp1δ-gfp models have been established (Iwawaki et al., 2004; Ryoo et al., 2007; Sone et al., 2013), the zebrafish model provides several advantages over the previously established mouse and Drosophila models. First, zebrafish embryos develop in water outside of their mothers. Therefore, xbp1 splicing and ER stress can be readily observed in Tg(ef1a:xbp1δ-gfp) transgenic zebrafish embryos. Because the ER process approximately one third of total cellular proteins, this heavy workload is highly susceptible to various insults, such as genetic mutation, oxidative stress, and chemical perturbation, leading to ER stress. Many pollutants are known or have the potential to induce ER stress, such as heavy metals, oxidative agents and pesticides. Therefore, the transgenic zebrafish embryos have great potential to serve as a convenient biosensor for testing water pollution. Second, zebrafish embryos are transparent, allowing fast and direct observation of XBP1Δ-GFP expression under a microscope. This transgenic zebrafish model is expected to have broad applications, including being a model to study the interaction between ER stress and various diseases, such as cancer and various neurodegenerative diseases in vivo. Third, the low cost and easy drug administration make zebrafish bioassays especially suitable for high throughput and large-scale drug screening. Therefore, this transgenic zebrafish provides an especially useful vertebrate model in drug screening for regulators of XBP1 activation and drug toxicity testing (Hill et al., 2005; Parng, 2005; Zon and Peterson, 2005; Rubinstein, 2006). It should be reminded that the XBP1Δ-GFP fusion protein may persist longer than the xbp1δ-gfp mRNA transcripts due to their difference in stability. Therefore, there may be a disconnect between the detection of GFP expression and the splicing of the xbp1δ-gfp transgene. This is an important aspect to consider when analyzing data interpretation generated from using this Tg(ef1:xbp1δ-gfp) transgenic zebrafish model.

17 Page 17 of 29

4. Experimental Procedures 4.1. Ethics Statement This study was carried out in accordance with the recommendation in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. The protocol was approved by the Institutional Animal Care and Use Committee of the University of Maryland (Permit Number: 0513005).

4.2. Fish maintenance and use Zebrafish were maintained at 28.5 C in 10 gallon aquarium supplied with fresh water and air at a photoperiod of 14 hours of light and 10 hours of dark at the Zebrafish Facility at Aquaculture Research Center (ARC) in Columbus Center (Baltimore). Pairs of adult male and female zebrafish were put in 2 liter tanks for natural spawning after lights were turned on in the facility at 9 am. Embryos were collected and raised at 28.5 C. To ease pain and facilitate animal handling, fish embryos over one day old were anaesthetized in 0.6 mM Tricaine that has been buffered to neutral pH around 7. The anaesthetized embryos were sacrificed quickly by keeping on ice and used directly for RNA extraction or fixation.

4.3. Cloning of xbp1 cDNAs from zebrafish The xbp1 cDNAs were cloned by RT-PCR using mRNA extracted from zebrafish embryos at 24hpf. The PCR was carried out with Advantage-2 DNA polymerase using the xbp1-ATG and xbp1-EGFP primers (Table 1).

The PCR products were cloned into

pGEM-T easy vector, and resulting two distinct types of plasmid designated as pGEM-xbp1u and pGEM-xbp1s. Sequence analysis revealed that they represented two different isoforms of xbp1 cDNAs that differ by 26 bp in the middle region of the coding sequence.

The xbp1u represents the longer unspliced isoform with the 26 nucleotide

IRE1 target sequence whereas the xbp1s represents the short spliced isoform without the 26 bp sequence.

18 Page 18 of 29

4.4. RT-PCR analysis The total RNAs were extracted from zebrafish embryos of various stages with or without DTT treatment using Trizol reagent (Invitrogen). Genomic DNA contamination was removed by digestion with RNase-free DNase (Qiagen). First strand cDNAs were synthesized by using the SuperScript III First Strand cDNA Synthesis kit (Invitrogen) using oligo (dT) primers. Expression and splicing of the endogenous xbp1 or xbp1δ-gfp transgene was determined by RT-PCR. The PCR was carried out for 35 cycles using specific primer sets (Table 1) designed for the endogenous xbp1 (P2/P5), the xbp1δ-gfp transgene (P8/R1), BIP (F2/R2) and the elongation factor-1 alpha (EF-1). The PCR products from the unspliced or spliced xbp1δ-gfp transgene are 280 bp and 254 bp, respectively, whereas, the PCR products from the unspliced and spliced endogenous xbp1 gene were 263 bp and 237 bp, respectively. The PCR products were separated on a 3% agarose gel.

4.5. Construction of the Tg(ef1:xbp1δ-gfp) transgene To construct a Tg(ef1:xbp1δ-gfp) fusion gene for transgenic fish production, a 239 bp (position 373-611) partial coding sequence of xbp1u isoform was amplified by PCR with pfu DNA polymerase (Invitrogen) using the xbp1u cDNA as the template. The PCR reaction was carried out using the xbp1-P8 and xbp1-P9 primers (Table 1). The 239 bp xbp1 PCR construct contained 66 bp from exon 3, 167 bp from exon 4 and 6 bp from exon 5. A BamHI site was added to the xbp1-P9 primer for later cloning into the pTol2 vector (pT2AL200R150G) (Urasaki et al., 2006).

The 239 bp xbp1 partial sequence contained

the 26-nt IRE1 target sequence which could be spliced out in response to ER stress. The 239 bp xbp1 PCR product was cloned in frame downstream of the myc-tag coding sequence at the StuI site in the pCS2+-MT vector by blunt end ligation. The resulting plasmid was named pCS2- XBP1myc. The DNA fragment encoding the myc-tagged XBP1myc was released from the pCS2-XBP1myc plasmid by BamH digestion and followed by gel purification using the Qiaquick gel extraction kit (Qiagen). The purified DNA fragment encoding the myc-tagged partial XBP1 was then cloned into the BamHI site 19 Page 19 of 29

upstream of the EGFP coding sequence in the pT2AL200R150G vector (Urasaki et al., 2006). The resulting plasmid was confirmed by DNA sequencing.

The plasmid was

named ef1:xbp1δ-gfp and used to generate the Tg(ef1:xbp1δ-gfp) transgenic lines.

4.6. Microinjection and generation of Tg(ef1:xbp1δ-gfp) transgenic zebrafish The Tg(ef1a:xbp1δ-gfp) DNA construct was dissolved in water to a final concentration of 100ng/l.

The DNA solution (2l) was mixed with an equal volume of

Tol2 transposase mRNA (50ng/l) transcribed in vitro from the pCS-TP plasmid (Urasaki et al., 2006). Approximately 1–2 nl of the DNA/mRNA mixture was injected into zebrafish embryo at one or two-cell stages.

Out of 15 fish screened, three germ line

transgenic founders were identified by direct observation of GFP expression in their F1 generations at 24 hpf.

F1 transgenic larvae were raised to adult hood and crossing with

wild-type fish to generate F2 transgenic lines for later analyses.

4.7. Induction of ER stress by DTT and tunicamycin in zebrafish embryos To induce ER stress in zebrafish embryos by DTT and tunicamycin, Tg(ef1a:xbp1δ-gfp) transgenic zebrafish embryos (100 per Petri dish) were incubated with 0.5mM DTT or 1 g/ml of tunicamycin for 24 hrs from 24 to 48 hpf.

To minimize

the impact from maternal GFP expression, transgenic F2 embryos generated from crossing a Tg(ef1a:xbp1δ-gfp) transgenic male with a non-transgenic female were used in this analysis. XBP1Δ-GFP expression was examined by the direct observation under a fluorescence microscopy after the DTT or tunicamycin treatment.

In addition, the

effects on xbp1 transcripts splicing and XBP1Δ-GFP expression were analyzed by RT-PCR and Western analyses. 4.8. Analysis of protein expression by Western blot Zebrafish embryos were dechorinated manually. The embryos (40 embryos each group) were washed with 1 ml of PBS and crushed gently to remove the yolk by pipetting with a glass pipet in 0.5 ml of PBS. The embryos were collected by a quick spin at 3000 rpm for 1 min. The embryos were washed once with 0.5 ml of PBS and solublized in 100 20 Page 20 of 29

µl of 2×SDS loading buffer (0.125 M Tris-Cl pH 6.8, 4% SDS, 20% Glycerol, 0.2 M DTT, 0.02% Bromophenol Blue). PMSF was added at the final concentration of 1mM to the protein extract to reduce protein degradation. The proteins were denatured by boiling for 3 min and analyzed on a 10% SDS-PAGE. Proteins from 2 embryos were loaded on each lane of the SDS-PAGE gel. Proteins from the gel were transferred onto a PVDF membrane (Immobilion-P, Millipore) by electrophoresis in transfer buffer (20% methanol, 25 mM Tris-HCl, 192 mM glycine, 0.1% SDS). The transfer was carried out for one hour for γ-tubulin and two hours for myc-tagged XBP1Δ-GFP fusion proteins at 400 mAmp. Immunodetection of myc-tagged proteins was carried out using the 9E10 antibody for myc-tagged proteins (DSHB) and followed by corresponding peroxidase-conjugated secondary antibodies (IgG 1:2000, Cell Signaling Technology, Beverly, MA). Immunodetection of -tubulin was carried out with an anti--tubulin antibody (1:5000, Sigma-Aldrich, St. Louis, MO) as loading control. Pierce ECL Western Blotting Substrate (Thermo Scientific, Rockford, IL) was used for detecting HRP-conjugated bound secondary antibodies.

Acknowledgments This research was supported by a research grant MB-8716-08 from United States – Israel Binational Agriculture Research and Development Fund and a grant from the University of Maryland intercenter collaboration.

21 Page 21 of 29

References

Back, S. H., and Kaufman, R. J. (2012). Endoplasmic reticulum stress and type 2 diabetes. Annu Rev Biochem 81, 767-793. Bennett, J. T., Joubin, K., Cheng, S., Aanstad, P., Herwig, R., Clark, M., Lehrach, H., and Schier, A. F. (2007). Nodal signaling activates differentiation genes during zebrafish gastrulation. Dev Biol 304, 525-540. Brodsky, J. L., Merz, A., and Serio, T. (2014). Organelle and proteome quality control mechanisms: how cells are able to keep calm and carry on. Mol Biol Cell 25, 733-734. Cao, Y., Knochel, S., Oswald, F., Donow, C., Zhao, H., and Knochel, W. (2006). XBP1 forms a regulatory loop with BMP-4 and suppresses mesodermal and neural differentiation in Xenopus embryos. Mech Dev 123, 84-96. Chen, Z., Ballar, P., Fu, Y., Luo, J., Du, S. J., Fang, S. (2014). The E3 ubiquitin ligase gp78 protects against ER stress in zebrafish liver.

J Genet Genomics. 41, 357-368

Cox, J. S., and Walter, P. (1996). A novel mechanism for regulating activity of a transcription factor that controls the unfolded protein response. Cell 87, 391-404. Dandekar A, Mendez R, Zhang K. (2015). Cross Talk Between ER Stress, Oxidative Stress, and Inflammation in Health and Disease. Methods Mol Biol. 1292, 205-14 Dicks N, Gutierrez K, Michalak M, Bordignon V, Agellon LB (2015). Endoplasmic reticulum stress, genome damage, and cancer. Front Oncol. 5, 11 Federovitch, C. M., Ron, D., and Hampton, R. Y. (2005). The dynamic ER: experimental approaches and current questions. Curr Opin Cell Biol 17, 409-414. Gardner, B.M., Pincus, D., Gotthardt, K., Gallagher, C.M., Walter, P. (2013). Endoplasmic reticulum stress sensing in the unfolded protein response. Cold Spring Harb Perspect Biol. 5, a013169. Glembotski, C. C. (2011). Functions for the cardiomyokine, MANF, in cardioprotection, hypertrophy and heart failure. J Mol Cell Cardiol 51, 512-517. 22 Page 22 of 29

Healy, S. J., Gorman, A. M., Mousavi-Shafaei, P., Gupta, S., and Samali, A. (2009). Targeting the endoplasmic reticulum-stress response as an anticancer strategy. Eur J Pharmacol 625, 234-246. Hill, A. J., Teraoka, H., Heideman, W., and Peterson, R. E. (2005). Zebrafish as a model vertebrate for investigating chemical toxicity. Toxicol Sci 86, 6-19. Homma, K., Katagiri, K., Nishitoh, H., and Ichijo, H. (2009). Targeting ASK1 in ER stress-related neurodegenerative diseases. Expert Opin Ther Targets 13, 653-664. Imrie, D., and Sadler, K. C. (2012). Stress Management: How the Unfolded Protein Response Impacts Fatty Liver Disease. J Hepatol Iwawaki, T., Akai, R., Kohno, K., and Miura, M. (2004). A transgenic mouse model for monitoring endoplasmic reticulum stress. Nat Med 10, 98-102. Iwawaki T, Oikawa D. (2013). The role of the unfolded protein response in diabetes mellitus. Semin Immunopathol. 35, 333-50 Kane, D. A., and Kimmel, C. B. (1993). The zebrafish midblastula transition. Development. 119, 447-456. Kapoor, A., and Sanyal, A. J. (2009). Endoplasmic reticulum stress and the unfolded protein response. Clin Liver Dis 13, 581-590. Maly D, J., and Papa, F. R. (2014). Druggable sensors of the unfolded protein response. Nat Chem Biol. 10, 892-901 Marciniak, S. J., and Ron, D. (2006). Endoplasmic reticulum stress signaling in disease. Physiol Rev 86, 1133-1149. Monk, K. R., Voas, M. G., Franzini-Armstrong, C., Hakkinen, I. S., Talbot, W. S. (2013). Mutation of sec63 in zebrafish causes defects in myelinated axons and liver pathology. Dis Model Mech. 6, 135-145. Parng, C. (2005). In vivo zebrafish assays for toxicity testing. Curr Opin Drug Discov Devel 8, 100-106. Reimold, A. M., Etkin, A., Clauss, I., Perkins, A., Friend, D. S., Zhang, J., Horton, H. F., Scott, A., Orkin, S. H., Byrne, M. C., Grusby, M. J., and Glimcher, L. H. (2000). An essential role in liver development for transcription factor XBP-1. Genes Dev 14, 23 Page 23 of 29

152-157. Ron, D., and Walter, P. (2007). Signal integration in the endoplasmic reticulum unfolded protein response. Nat Rev Mol Cell Biol 8, 519-529. Roussel BD, Kruppa AJ, Miranda E, Crowther DC, Lomas DA, Marciniak SJ. (2013). Endoplasmic reticulum dysfunction in neurological disease.

Lancet Neurol. 12,

105-18 Rubinstein, A. L. (2006). Zebrafish assays for drug toxicity screening. Expert Opin Drug Metab Toxicol 2, 231-240. Rutkowski, D. T., and Kaufman, R. J. (2004). A trip to the ER: coping with stress. Trends Cell Biol 14, 20-28. Ryoo, H. D., Domingos, P. M., Kang, M. J., and Steller, H. (2007). Unfolded protein response in a Drosophila model for retinal degeneration. EMBO J 26, 242-252. Shen, X., Ellis, R. E., Lee, K., Liu, C. Y., Yang, K., Solomon, A., Yoshida, H., Morimoto, R., Kurnit, D. M., Mori, K., and Kaufman, R. J. (2001). Complementary signaling pathways regulate the unfolded protein response and are required for C. elegans development. Cell 107, 893-903. Shim, J., Umemura, T., Nothstein, E., and Rongo, C. (2004). The unfolded protein response regulates glutamate receptor export from the endoplasmic reticulum. Mol Biol Cell 15, 4818-4828. Sone, M., Zeng, X., Larese, J., and Ryoo, H. D. (2013). A modified UPR stress sensing system reveals a novel tissue distribution of IRE1/XBP1 activity during normal Drosophila development. Cell Stress Chaperones 18, 307-319. Souid, S., Lepesant, J. A., and Yanicostas, C. (2007). The xbp-1 gene is essential for development in Drosophila. Dev Genes Evol 217, 159-167. Spiotto, M. T., Banh, A., Papandreou, I., Cao, H., Galvez, M. G., Gurtner, G. C., Denko, N. C., Le, Q. T., and Koong, A. C. (2010). Imaging the unfolded protein response in primary tumors reveals microenvironments with metabolic variations that predict tumor growth. Cancer Res 70, 78-88. Stolz A, Wolf DH. (2010). Endoplasmic reticulum associated protein degradation: a 24 Page 24 of 29

chaperone assisted journey to hell.

Biochim Biophys Acta. 1803(6):694-705.

Thakur, P. C., Stuckenholz, C., Rivera, M. R., Davison, J. M., Yao, J. K., Amsterdam, A., Sadler, K. C., and Bahary, N. (2011). Lack of de novo phosphatidylinositol synthesis leads to endoplasmic reticulum stress and hepatic steatosis in cdipt-deficient zebrafish. Hepatology. 54, 452-462 Thakur, P. C., Davison, J. M., Stuckenholz, C., Lu, L., Bahary, N. (2014). Dysregulated phosphatidylinositol signaling promotes endoplasmic-reticulum-stress-mediated intestinal mucosal injury and inflammation in zebrafish. Dis Model Mech.7, 93-106. Thisse, B., Heyer, V., Lux, A., Alunni, V., Degrave, A., Seiliez, I., Kirchner, J., Parkhill, J. P., and Thisse, C. (2004). Spatial and temporal expression of the zebrafish genome by large-scale in situ hybridization screening. Methods Cell Biol 77, 505-519. Thummel, R., Burket, C. T., and Hyde, D. R. (2006). Two different transgenes to study gene silencing and re-expression during zebrafish caudal fin and retinal regeneration. ScientificWorldJournal 6 Suppl 1, 65-81. Travers, K. J., Patil, C. K., Wodicka, L., Lockhart, D. J., Weissman, J. S., and Walter, P. (2000). Functional and genomic analyses reveal an essential coordination between the unfolded protein response and ER-associated degradation. Cell 101, 249-258. Tsai, B., Ye, Y., and Rapoport, T. A. (2002). Retro-translocation of proteins from the endoplasmic reticulum into the cytosol. Nat Rev Mol Cell Biol 3, 246-255. Tsedensodnom, O., Vacaru, A. M., Howarth, D. L., Yin, C., and Sadler, K. C. (2013). Ethanol metabolism and oxidative stress are required for unfolded protein response activation and steatosis in zebrafish with alcoholic liver disease. Dis Model Mech. 6,1213-1226. Urasaki, A., Morvan, G., and Kawakami, K. (2006). Functional dissection of the Tol2 transposable element identified the minimal cis-sequence and a highly repetitive sequence in the subterminal region essential for transposition. Genetics 174, 639-649. Vacaru, A. M., Di Narzo, A. F., Howarth, D. L., Tsedensodnom, O., Imrie, D., Cinaroglu, A., Amin, S., Hao, K., Sadle,r K. C. (2014). Molecularly defined unfolded protein 25 Page 25 of 29

response subclasses have distinct correlations with fatty liver disease in zebrafish. Dis Model Mech. 7, 823-835. Walter, P., and Ron, D. (2011). The unfolded protein response: from stress pathway to homeostatic regulation. Science 334, 1081-1086. Wang, S., and Kaufman, R. J. (2012). The impact of the unfolded protein response on human disease. J Cell Biol 197, 857-867. Wang M, Kaufman RJ. (2014). The impact of the endoplasmic reticulum protein-folding environment on cancer development. Nat Rev Cancer. 14, 581-97. Wu, L. L., Russell, D. L., Norman, R. J., and Robker, R. L. (2012). Endoplasmic reticulum (ER) stress in cumulus-oocyte complexes impairs pentraxin-3 secretion, mitochondrial membrane potential (DeltaPsi m), and embryo development. Mol Endocrinol 26, 562-573. Yadav RK, Chae SW, Kim HR, Chae HJ. (2014). Endoplasmic reticulum stress and cancer. J Cancer Prev. 19, 75-88. Yoshida, H., Matsui, T., Yamamoto, A., Okada, T., and Mori, K. (2001). XBP1 mRNA is induced by ATF6 and spliced by IRE1 in response to ER stress to produce a highly active transcription factor. Cell 107, 881-891. Zhang, J. Y., Diao, Y. F., Kim, H. R., and Jin, D. I. (2012a). Inhibition of endoplasmic reticulum stress improves mouse embryo development. PLoS One 7, e40433. Zhang, J. Y., Diao, Y. F., Oqani, R. K., Han, R. X., and Jin, D. I. (2012b). Effect of endoplasmic reticulum stress on porcine oocyte maturation and parthenogenetic embryonic development in vitro. Biol Reprod 86, 128. Zhao, H., Cao, Y., and Grunz, H. (2003). Xenopus X-box binding protein 1, a leucine zipper transcription factor, is involved in the BMP signaling pathway. Dev Biol 257, 278-291. Zon, L. I., and Peterson, R. T. (2005). In vivo drug discovery in the zebrafish. Nat Rev Drug Discov 4, 35-44.

26 Page 26 of 29

Figure Legends Figure 1. A schematic diagram shows zebrafish xbp1 genomic structure and analysis of xbp1 expression and splicing in developing zebrafish embryos. (A). The zebrafish xbp1 gene is approximately 3 kb and composed of five exons. A conserved ER stress responsive element is located at the 5’-flanking region. The grey region within Exon IV indicates the 26 nucleotides (nt) that are spliced out by inositol requiring enzyme 1(IRE1) under ER stress conditions. (B). Comparison of the spliced (XBP1S) and unspliced (XBP1U) xbp1 transcripts from zebrafish (ZF) and mouse reveals a conserved direct repeats (TCCGCAGCA) flanking the 26 bp IRE1 target sequence. (C). RT-PCR analysis shows the expression of two isoforms of xbp1 transcripts in zebrafish embryos at 0.2, 3, 6, 9, 12, 24 and 48 hpf. The spliced isoform appears to be the dominant isoform at 0.2 and 3 hpf. Expression of elongation factor 1-alpha (EF-1) mRNA was used as an internal control. Figure 2. Construction of the Tg(ef1:xbp1δ-gfp) transgene and expression of XBP1Δ-GFP in live zebrafish embryos (A). A partial zebrafish xbp1 cDNA sequence (239 bp) containing the 26 nt IRE1 target sequence was cloned upstream of the EGFP coding sequence. A myc-tag sequence was also included in frame at 5' end of the xbp1 sequence. Under normal conditions without ER stress, the xbp1δ-gfp transcripts are not spliced and protein translation stops prior to the GFP coding sequence. Upon ER stress, the 26 nt IRE1 target sequence is spliced out from the xbp1δ-gfp mRNA transcripts, leading to a frame shift and the expression of the XBP1Δ-GFP fusion protein. (B). Expression of the XBP1Δ-GFP in live zebrafish embryos. XBP1Δ-GFP expression was directly observed in the F1 transgenic zebrafish embryos from one cell stage to 24 hpf. A strong XBP1Δ-GFP expression was observed in transgenic embryos with the transgene derived maternally. In contrast, transgenic embryos with paternally derived transgene showed no XBP1Δ-GFP expression from one cell stage to 12 hpf. A weak expression was detected at 24hpf. Non-transgenic fish embryos were used as control. Figure 3. Expression and splicing of the Tg(ef1a: xbp1δ-gfp) transgene and the xbp1 endogenous gene in zebrafish embryos The expression and splicing of the Tg(ef1:xbp1δ-gfp) transgene (A and C) or the xbp1 endogenous gene (B and D) were analyzed by RT-PCR in paternal (A and B) or maternal (C and D) derived transgenic zebrafish embryos at 0.2, 3, 6, 9, 12, 24 and 48 hpf. Figure 4. Maternal expression of XBP1Δ-GFP and the spliced isoform of xbp1 transcripts (A, B) Direct observation of XBP1Δ-GFP expression in unfertilized zebrafish oocytes dissected from Tg(ef1:xbp1δ-gfp) transgenic (A) or non-transgenic (B) females. (C). RT-PCR analysis of endogenous xbp1 mRNA splicing and BIP expression in unfertilized eggs, testis from adult fish, and embryos at 3 hpf and 24 hpf. 27 Page 27 of 29

Figure 5. XBP1Δ-GFP expression in homozygous transgenic zebrafish larvae (A and B). XBP1Δ-GFP expression in live Tg(ef1a:xbp1δ-gfp) transgenic zebrafish larvae at 24 hpf. (C and D). Non-transgenic fish larvae at 24 hpf with GFP filter (C) or bright filed (D). (E and G). XBP1Δ-GFP expression in Tg(ef1a:xbp1δ-gfp) transgenic zebrafish larvae at 72 (C) or 150 (G) hpf. (F and H). Non-transgenic fish larvae at 72 (F) or 150 (H) hpf. (I and J) Expression and splicing of the Tg(ef1:xbp1δ-gfp) transgene (I) or the elongation factor  (J) at day 3, 5 and 7. Figure 6. Characterization of DTT and tunicamycin induced ER stress in Tg(ef1:xbp1δ-gfp) transgenic zebrafish embryos Transgenic embryos with the paternally derived Tg(ef1:xbp1δ-gfp) transgene were treated with 0.5 mM of DTT, 1 g/ml or 2g/ml of tunicamycin between 24-48 hpf. XBP1Δ-GFP expression and mRNA splicing were analyzed at 48 hpf after the DTT or tunicamycin treatment. (A, E). Side and dorsal views of non-transgenic fish embryos at 48 hpf. (B, F) Side and dorsal view of Tg(ef1:xbp1δ-gfp) transgenic embryos at 48 hpf without DTT treatment. (C, G) XBP1Δ-GFP expression in transgenic embryos at 48 hpf with 24 h of 0.5 mM DTT treatment. (D, H) XBP1Δ-GFP expression in transgenic embryos at 48 hpf with 24 h of 1g/ml tunicamycin treatment (D) or without the treatment (H). (I). RT-PCR analysis shows the expression and splicing of the Tg(ef1:xbp1δ-gfp) transgene and the endogenous xbp1 gene in DTT treated or control paternal transgenic (P-) and wild type (WT-) embryos. P-Endo represents xbp1 expression from the endogenous gene. P-Exo represents xbp1 expression from the exogenous Tg(ef1:xbp1δ-gfp) transgene. WT-Endo represents xbp1 expression from the endogenous gene in wild type embryos. Arrows indicate spliced xbp1 transcripts. (J). Western blot analysis shows the expression of the myc-tagged XBP1Δ-GFP fusion protein derived from the paternal transgenic line at 48 hpf in control and DTT treated embryos. (K). RT-PCR analysis shows the splicing of endogenous and exogenous xbp1 in paternal transgenic zebrafish embryos treated with 1g/ml (TUN-1) and 2g/ml (TUN-2) tunicamycin for 24 hpf between 24-48 hpf. EF1 expression was analyzed as an internal control. Arrows indicate spliced xbp1 transcripts. (L). RT-PCR analysis shows the expression of ER stress markers, xbp1 and BIP, in zebrafish embryos treated with 0.5 mM DTT for 24 hpf between 24-48 hpf.

28 Page 28 of 29

Table 1. List of PCR Primers Primer names

Primer sequence

BIP-F2

AGATCTTCTCCACTGCTTCCGACAA

BIP-R2

TCTACAGCTCGTCCTTCTCTTCGGC

xbp1-ATG

5’-GGATCCAAATGGTCGTAGTTACAGCAGGGAC

xbp1-P2

5'-GCAGGAGATCAGACTCAGAGTCTG

xbp1-P5

5'-CGAGACAAGACGAGTGATCTGCT

xbp1-P8

5'- CTGCTCAGTGAGAATGAGGAGCTG

xbp1-P9

5’- GGATCCGCATCAGACTCAGAGTCTGCAGGGCC

XBP1Δ-GFP-R1

5'-TCCTCGCCCTTGCTCACCATGGT

xbp1-EGFP

5’- GGATCCGCGACGCTAATCAGTTGGGGGAAGA

EF-1F1

5'-GCATACATCAAGAAGATCGGC

EF-1R1

5'-GCAGCCTTCTGTGCAGACTTTG

29 Page 29 of 29