Enhanced ER-associated degradation of HMG CoA reductase causes embryonic lethality associated with Ubiad1 deficiency

UbiA prenyltransferase domain-containing protein-1 (UBIAD1) synthesizes the vitamin K subtype menaquinone-4 (MK-4). Previous studies in cultured cells (Schumacher et al., 2015) revealed that UBIAD1 also inhibits endoplasmic reticulum (ER)-associated degradation (ERAD) of ubiquitinated HMG CoA reductase (HMGCR), the rate-limiting enzyme of the mevalonate pathway that produces cholesterol and essential nonsterol isoprenoids. Gene knockout studies were previously attempted to explore the function of UBIAD1 in mice; however, homozygous germ-line elimination of the Ubiad1 gene caused embryonic lethality. We now report that homozygous deletion of Ubiad1 is produced in knockin mice expressing ubiquitination/ERAD-resistant HMGCR. Thus, embryonic lethality of Ubiad1 deficiency results from depletion of mevalonate-derived products owing to enhanced ERAD of HMGCR rather than from reduced synthesis of MK-4. These findings provide genetic evidence for the significance of UBIAD1 in regulation of cholesterol synthesis and offer the opportunity in future studies for the discovery of new physiological roles of MK-4.


Introduction
Vitamin K refers to a group of lipophilic molecules that serve as a cofactor for g-carboxyglutamyl carboxylase, which converts specific glutamate residues in a limited set of proteins to g-carboxyglutamate (Shearer and Newman, 2014;Shearer and Okano, 2018). This post-translational modification is obligatory for biological functions of resultant vitamin K-dependent proteins (VKDPs), some of which play key roles in blood coagulation. Other VKDPs are implicated in processes ranging from bone and cardiovascular mineralization to energy metabolism and inflammation (Booth, 2009;Shearer and Okano, 2018). In addition, vitamin K may exert direct effects on gene expression, signal transduction, and cellular regulation.
All vitamin K forms include a common 2-methyl-1,4-naphthoquinone ring structure known as menadione (MD) ( Figure 1A) and are distinguished from one another by length and saturation of the side chain attached at the 3-carbon position on the ring (Shearer and Newman, 2014). MD is a provitamin form of vitamin K as the side chain is required for vitamin K activity (Buitenhuis et al., 1990). Phylloquinone (PK, also known as vitamin K 1 ) ( Figure 1A) contains a phytyl side chain, whereas menaquinones (MKs, collectively referred to as vitamin K 2 ) contain a side chain with 5-carbon isoprenyl units and are named according to the number of these units (e.g., MK-n) ( Figure 1A). PK is UbiA prenyltransferase domain-containing protein-1 (UBIAD1), a member of the UbiA superfamily of prenyltransferases (Li, 2016), transfers the 20-carbon geranylgeranyl group from geranylgeranyl pyrophosphate (GGpp) to PK-derived MD, thereby producing MK-4 (Hirota et al., 2013;Nakagawa et al., 2010;Figure 1B). The function of UBIAD1 appears to extend beyond its role in MK-4 synthesis, as indicated by association of mutations in human UBIAD1 with Schnyder corneal dystrophy (SCD) (Orr et al., 2007;Weiss et al., 2007). This rare, autosomal-dominant disease is characterized by progressive corneal opacification that results from accumulation of cholesterol. In 2015, we showed that SCD-associated UBIAD1 inhibits the sterol-accelerated, endoplasmic reticulum (ER)-associated degradation (ERAD) of 3-hydroxy-3-methylglutaryl coenzyme A reductase (HMGCR) (Schumacher et al., 2015), one of several feedback mechanisms that converge on the enzyme to assure cholesterol homeostasis (Brown and Goldstein, 1980).
Sterols also cause HMGCR to bind UBIAD1 (Schumacher et al., 2015). This binding inhibits the ERAD of HMGCR at a post-ubiquitination step in the reaction, thereby permitting continued synthesis of mevalonate for incorporation into nonsterol isoprenoids even when intracellular sterols are abundant (Schumacher et al., 2018). GGpp triggers release of UBIAD1 from HMGCR, which allows for maximal ERAD of HMGCR and translocation of UBIAD1 from the ER to the medial-trans Golgi. SCD-associated mutations cluster around the membrane-embedded active site of UBIAD1 (Cheng and Li, 2014;Huang et al., 2014), indicating they may disrupt sensing of GGpp. Indeed, SCD-associated UBIAD1 is refractory to GGpp-induced release from HMGCR and becomes sequestered in the ER (Schumacher et al., 2016). The resultant inhibition of HMGCR ERAD leads to enhanced synthesis and intracellular accumulation of cholesterol (Schumacher et al., 2018).
To explore the in vivo function of UBIAD1, efforts were attempted to generate mice lacking Ubiad1 (Nakagawa et al., 2014). However, mouse embryos homozygous for Ubiad1 deficiency failed to survive past embryonic day 7.5. We recently observed that the ERAD of HMGCR was enhanced in transformed human fibroblasts lacking UBIAD1 (Schumacher et al., 2018). This observation led us to speculate that embryonic lethality of Ubiad1 deficiency in mice results from mevalonate depletion due to accelerated ERAD of HMGCR rather than from reduced synthesis of MK-4. We evaluate this notion here by determining whether ubiquitination/ERAD-resistant HMGCR rescues embryonic lethality of Ubiad1-deficiency.
Having established the absence of UBIAD1 protein and its enzymatic product (MK-4) in Ubiad1deficient mice, we next compared blood chemistries between Ubiad1 +/+ : :Hmgcr Ki/Ki and Ubiad1 -/-: : Hmgcr Ki/Ki animals. Serum cholesterol levels were not significantly different between the two groups  In Figure 4D, we conducted a second blood chemistry analysis on male Ubiad1 +/+ : :Hmgcr Ki/Ki and Ubiad1 -/-: :Hmgcr Ki/Ki mice. Similar to results of Figure 4C, serum levels of AST and ALT were elevated 5-fold and 1.8-fold, respectively, in the absence of Ubiad1 ( Figure 4D). Serum lactate dehydrogenase (LDH) was elevated approximately 2-fold in Ubiad1-deficient mice; however, a more prominent elevation (7.5-fold) of serum creatine kinase (CK) was observed. Finally, a slight reduction in the amount of serum lipase and amylase was present in Ubiad1-deficient mice; serum albumin remained unchanged. Serum from female Ubiad1 -/-: :Hmgcr Ki/Ki mice exhibited similar characteristics ( Figure 4-figure supplement 1B).
To further characterize Ubiad1 -/-: :Hmgcr Ki/Ki mice, we conducted a complete histological analysis of all tissues from the animals. Surprisingly, abnormalities were observed in only two tissues of Ubiad1-deficient mice -skeletal muscle and bone. Hematoxylin and eosin (H and E)-staining of gastrocnemius ( Figure 5A, panels 1-4) and quadriceps muscles (panels 5-8) from male Ubiad1-deficient mice revealed occasional degenerating myofibers with macrophage infiltration as well as frequent myofibers with centrally-localized nuclei. Similar results were observed in gastrocnemius and quadriceps muscles isolated from female mice ( Figure 5-figure supplement 1). Overall, these histological findings are indicative of ongoing muscle injury and correlate to the elevated serum CK levels observed in Figure 4D. We used H and E together with Safranin O staining to examine growth plates in femurs from both male ( Figure 5B) and female (Figure 5-figure supplement 2) Ubiad1 +/ + : :Hmgcr Ki/Ki and Ubiad1 -/-: :Hmgcr Ki/Ki mice. The results reveal that Ubiad1 deficiency led to the disorganization of cells within proliferative and hypertrophic zones of the growth plate, persistence of cartilage within trabeculae, and a mild decrease in the number of boney trabeculae.

Discussion
The genetic ablation of UBIAD1 in transformed human fibroblasts led to enhanced ERAD of HMGCR and reduced cholesterol synthesis and intracellular accumulation of cholesterol (Schumacher et al., 2018). These observations prompted us to consider the possibility that embryonic lethality associated with Ubiad1 deficiency in mice resulted from mevalonate depletion. Indeed, homozygous Hmgcr deficiency caused early embryonic lethality in mice (Ohashi et al., 2003), establishing that mevalonate-derived metabolites are crucial for embryonic development. We previously generated Hmgcr Ki/Ki mice, which harbor knockin mutations in the Hmgcr gene that prevent sterol-induced ubiquitination and subsequent ERAD of HMGCR (Hwang et al., 2016). As a result of resistance to ERAD, HMGCR protein accumulated in tissues of Hmgcr Ki/Ki mice that stimulated the overproduction of cholesterol and likely other sterol and nonsterol isoprenoids. Hence, we reasoned that overproduction of sterol and nonsterol isoprenoids in Hmgcr Ki/Ki mice would rescue embryonic lethality associated with Ubiad1 deficiency. Our current studies show that homozygous Ubiad1 deletion was produced at expected Mendelian ratios in Hmgcr Ki/Ki , but not in WT mice (Table 1). This important observation not only highlights the crucial role for UBIAD1 in regulation of HMGCR ERAD, but also 3 continued pooled and subjected to SDS-PAGE, followed by immunoblot analysis using antibodies against endogenous HMGCR, UBIAD1, SREBP-1, SREBP-2, calnexin, and LSD-1. Although shown in separate panels, LSD-1 is a loading control for nuclear SREBP immunoblots. (C) Male and female Ubiad1 +/+ : : Hmgcr Ki/Ki , Ubiad1 +/-: :Hmgcr Ki/Ki , and Ubiad1 -/-: :Hmgcr Ki/Ki littermates (eight mice/group) were weaned at 4 weeks of age, fed chow diet ad libitum, and weighed for seven consecutive weeks, after which they were sacrificed. Error bars, S.E. *, p<0.05 and ****, p<0.0001. The online version of this article includes the following source data and figure supplement(s) for figure 3: Source data 1. Body weights of Ubiad1 -/-: : Hmgcr Ki/Ki mice.  confirms that abrogating the reaction allows production of a mevalonate-derived metabolite(s) that rescues embryonic lethality associated with Ubiad1 deficiency. Notably, administration of MK-4, ubiquinone-10 (Nakagawa et al., 2014), or cholesterol (data not shown) to Ubiad1 +/mice prior to intercrossing and throughout pregnancy failed to rescue embryonic lethality of Ubiad1 deficiency. While the identity of the mevalonate-derived metabolite(s) that rescues embryonic development     remains unknown, genes encoding HMGCR, enzymes required for GGpp and Fpp synthesis, and prenylation of small GTPases are essential for migration of primordial germ cells during embryonic development (Kunwar et al., 2006). Thus, depletion of GGpp and/or Fpp owing to enhanced ERAD of HMGCR and reduced prenylation of small GTPases may contribute to embryonic lethality associated with Ubiad1 deficiency. Although ubiquitination/ERAD-resistant HMGCR rescues embryonic lethality of Ubiad1 deficiency (Table 1), Ubiad1 -/-: :Hmgcr Ki/Ki mice 8-12 weeks of age were 20-40% smaller than their Ubiad1 +/+ : : Hmgcr Ki/Ki and Ubiad1 +/-: :Hmgcr Ki/Ki littermates ( Figure 3B and C). It will be important in future studies to determine whether Ubiad1-deficient mice consume less food, have defects in nutrient absorption, or exhibit increased energy expenditure. UBIAD1 protein and its enzymatic product MK-4 were not detected in tissues of Ubiad1-deficient mice ( Figures 3B, 4A and B). However, it is important to note that despite the absence of MK-4, Ubiad1-deficient mice did not exhibit typical signs of vitamin K deficiency (i.e., excessive hemorrhaging). The diet used in this study is supplemented with MD, which does not exhibit vitamin K activity (Buitenhuis et al., 1990) and is not converted to MK-4 in absence of UBIAD1. This indicates that the diet and/or gut microbiota provide sufficient amounts of vitamin K to support g-glutamyl carboxylation of coagulation factors in Ubiad1 -/-: :Hmgcr Ki/Ki mice, suggesting their failure to thrive may result from disruption of carboxylation-independent activities of MK-4. Figure 4B shows that similar to previous results (Harshman et al., 2016;Okano et al., 2008), levels of MK-4 were highest in the pancreas and brain. Thus, failure of Ubiad1-deficient mice to thrive may result from reduced production of MK-4 in these or other tissues of the animals.
Recent studies show that inducible knockout of Ubiad1 in adult mice caused death within 60 days (Nakagawa et al., 2019). The most striking abnormality of these mice was a remarkable reduction in pancreas size resulting from apoptotic disappearance of acinar cells. This observation prompted the authors to conclude that UBIAD1-mediated synthesis of MK-4 is essential for survival of pancreatic acinar cells. Our current studies reveal that the pancreas of Ubiad1 -/-: :Hmgcr Ki/Ki mice failed to produce MK-4 ( Figure 4B); however, the organ was normal in size and exhibited no gross abnormalities (data not shown). These findings argue that ubiquitination/ERAD-resistant HMGCR allows for production of sterol and/or nonsterol isoprenoids distinct from MK-4 that are essential for subsistence of pancreatic acinar cells.
Blood chemistry analyses were conducted to determine whether Ubiad1 deficiency results in damage of the liver and/or other organs. Compared to Ubiad1 +/+ : :Hmgcr Ki/Ki littermates, Ubiad1 -/-: :Hmgcr Ki/Ki mice exhibited elevated levels of ALT (1.6-fold) and AST (>5 fold) in the serum ( Figure 4C and D; Figure 4-figure supplement 1). Elevated levels of serum aminotransferases are routinely applied as biomarkers for hepatocyte injury. However, livers of Ubiad1-deficient mice did not feature gross abnormalities upon histological analysis (data not shown). Further examination revealed Ubiad1 deficient mice exhibited a 2-fold increase in serum LDH and a 4-7.5-fold increase in CK that was accompanied by a 35% decrease in ALP (Figure 4 and Statins, competitive inhibitors of HMGCR, are widely prescribed to lower plasma levels of cholesterol-rich low-density lipoprotein and reduce atherosclerotic cardiovascular disease. However, a significant fraction of patients undergoing statin therapy develop myopathy; a small portion of these patients progress to rhabdomyolysis (Thompson et al., 2003;Ward et al., 2019). Statin-induced myopathy has been attributed to depletion of mevalonate-derived metabolites resulting from inhibition of HMGCR. Skeletal muscle-specific knockout of HMGCR in mice causes severe myopathy that is rescued by mevalonate (Osaki et al., 2015). The observation that Ubiad1 -/-: :Hmgcr Ki/Ki mice exhibit signs of muscle injury suggests statin-induced myopathy may in part, result from MK-4 depletion. Support for this possibility requires MK-4 rescue experiments in Ubiad1-deficient mice and determination of whether Hmgcr Ki/Ki mice with skeletal muscle-specific knockout of Ubiad1 develop myopathy.
Evidence indicates that vitamin K modulates bone homeostasis and metabolism through two mechanisms. One mechanism is mediated by osteocalcin and matrix Gla protein (Fusaro et al., 2017), two VKDPs that play key roles in bone formation and mineralization. The second mechanism is mediated by the nuclear receptor known as steroid and xenobiotic receptor (SXR) in humans and pregnane X receptor (PXR) in mice. These promiscuous nuclear receptors are activated by a wide variety of xenobiotics and regulate genes involved in metabolism and clearance of the substances (Kliewer, 2003). Pxr-deficient mice present with osteopenia accompanied by reduced bone formation and increased bone resorption (Azuma et al., 2010). Considering that MK-4 has been reported to bind to and activate PXR (Tabb et al., 2003), it will be important to determine whether Ubiad1deficiency phenocopies Pxr-deficiency with regard to bone homeostasis.
The characterization of genetically-manipulated mice underscores the physiological significance of UBIAD1 as an inhibitor of HMGCR ERAD. We recently generated mice (designated Ubiad1 Ki/Ki ) harboring a knockin mutation that changes asparagine-100 (N100) to serine (N100S) (Jo et al., 2019). The N100S mutation in mouse UBIAD1 corresponds to SCD-associated N102S mutation in human UBIAD1 that abolishes sensing of membrane-embedded GGpp. UBIAD1 (N100S) was sequestered in ER membranes to inhibit ERAD of HMGCR, causing the protein's accumulation and overproduction of sterol and nonsterol isoprenoids in the liver and other tissues of Ubiad1 Ki/Ki mice. Significant corneal opacification was observed in Ubiad1 Ki/Ki mice greater than 50 weeks of age (Hmgcr Ki/Ki mice used in the current study were not aged and thus, not examined for corneal opacification). Considered together with current studies, these findings unequivocally position UBIAD1 as a major regulator of HMGCR and mevalonate metabolism in vivo, provide new links between synthesis of sterols and MK-4, and establish Ubiad1 -/-: :Hmgcr Ki/Ki mice as a model of MK-4 deficiency. Further analysis of these mice may reveal new physiological roles for MK-4 and additional pathways modulated by the vitamin K subtype.

Mice
Previously described Hmgcr Ki/Ki mice harbor homozygous nucleotide mutations in the Hmgcr gene that change lysine residues 89 and 248 to arginine (Hwang et al., 2016). These mutations prevent Insig-mediated ubiquitination and subsequent ERAD of HMGCR in the liver and other tissues of the knockin mice. Ubiad1 -/+ and Ubiad1 +/-: :Hmgcr Ki/Ki mice (C57BL/6N background) were generated using WT and Hmgcr Ki/Ki mice, respectively, using CRISPR/Cas9 technology in the Transgenic Core Facility at UT Southwestern Medical Center. The guide RNAs were designed to generate a deletion in exon 1 of the Ubiad1 gene, resulting in production of a truncated, nonfunctional protein (see Figure 2). F 0 founders were used to produce F 1 offspring that carried the Ubiad1-deficient allele through the germline. Pairs (male and female) of Ubiad1 +/and Ubiad1 +/-: :HmgcrKi Ki/Ki mice were intercrossed for production of homozygous Ubiad1-deficient mice in the WT or Hmgcr Ki/Ki background. To genotype Ubiad1-deficient animals, genomic DNA from tails was used for PCR with the primers indicated in the Key resources table against the mouse Ubiad1 sequence. The genotype of Hmgcr Ki/Ki mice was determined as described previously (Hwang et al., 2016

Subcellular fractionation and immunoblot analysis
Approximately 80 mg of frozen tissue was homogenized in 500 ml buffer (10 mM HEPES-KOH, pH 7.6, 1.5 mM MgCl 2 , 10 mM KCl, 5 mM EDTA, 5 mM EGTA, and 250 mM sucrose) supplemented with a protease inhibitor cocktail consisting of 0.1 mM leupeptin, 5 mM dithiothreitol, 1 mM PMSF, 0.5 mM Pefabloc, 5 mg/ml pepstatin A, 25 mg/ml N-acetyl-leu-leu-norleucinal, and 10 mg/ml aprotinin. The homogenates were then passed through a 22-gauge needle 10-15 times and subjected to centrifugation at 1,000 X g for 5 min at 4˚C. The 1,000 X g pellet was resuspended in 500 ml of buffer (20 mM HEPES-KOH, pH 7.6, 2.5% (v/v) glycerol, 0.42 M NaCl, 1.5 mM MgCl 2 , 1 mM EDTA, 1 mM EGTA) supplemented with the protease inhibitor cocktail, rotated for 30 min at 4˚C, and centrifuged at 100,000 X g for 30 min at 4˚C. The supernatant from this spin was precipitated with 1.5 ml cold acetone at À20˚C for at least 30 min; the precipitated material was collected by centrifugation, resuspended in SDS-lysis buffer (10 mM Tris-HCl, pH 6.8, 1% (w/v) SDS, 100 mM NaCl, 1 mM EDTA, and 1 mM EGTA), and designated the nuclear extract fraction. The post-nuclear supernatant from the original spin was used to prepare the membrane fraction by centrifugation at 100,000 X g for 30 min at 4˚C. Each membrane fraction was resuspended in 100 ml SDS-lysis buffer. Protein concentration of nuclear extract and membrane fractions were measured using the BCA Kit (ThermoFisher Scientific). Prior to SDS-PAGE, aliquots of the nuclear extract fractions were mixed with 5X SDS-PAGE loading buffer to achieve a final concentration of 1X. Aliquots of the membrane fractions were mixed with an equal volume of buffer containing 62.5 mM Tris-HCl, pH 6.8, 15% (w/v) SDS, 8 M urea, 10% (v/v) glycerol, and 100 mM DTT, after which 5X SDS loading buffer was added to a final concentration of 1X. Nuclear extract fractions were boiled for 5 min, and membrane fractions were incubated for 20 min at 37˚C prior to SDS-PAGE. After SDS-PAGE, proteins were transferred to Hybond C-Extra nitrocellulose filters (GE Healthcare, Piscataway, NJ). The filters were incubated with the antibodies described below and in the figure legends. Bound antibodies were visualized with peroxidase-conjugated, affinity-purified donkey anti-mouse or anti-rabbit IgG (Jackson ImmunoResearch Laboratories, Inc, West Grove, PA) using the SuperSignal CL-HRP substrate system (ThermoFisher Scientific) according to the manufacturer's instructions. Gels were calibrated with prestained molecular mass markers (Bio-Rad, Hercules, CA). Filters were exposed to film at room temperature. Antibodies used for immunoblotting to detect mouse SREBP-1 (rabbit monoclonal IgG-20B12), SREBP-2 (rabbit monoclonal IgG-22D5), HMGCR (IgG-839c), and UBIAD1 (rabbit polyclonal IgG-205) were previously described (Engelking et al., 2005;Jo et al., 2011;McFarlane et al., 2014;Rong et al., 2017). Rabbit polyclonal anti-calnexin IgG was purchased from Novus Biologicals (Centennial, CO). Rabbit polyclonal anti-LSD1 IgG was obtained from Cell Signaling (Danvers, MA). All antibodies were used at a final concentration of 1-5 mg/ml; the anti-calnexin antiserum was used at a dilution of 1:5000.
Blood chemistry, MK-4 measurement, and histological analysis Blood was drawn from the vena cava after mice were anesthetized in a bell jar atmosphere containing isoflurane. Serum was immediately separated and analyzed or stored at À80˚C until use. Blood chemistries (cholesterol, triglycerides, AST, ALT, ALP, nonesterified fatty acids, etc.) were measured in the Metabolic Phenotyping Core Facility at UT Southwestern Medical Center. MK-4 levels in mouse tissues was measured as follows. Approximately 100 mg of tissue from Ubiad1 +/+ : :Hmgcr Ki/Ki and Ubiad1 -/-: :Hmgcr Ki/Ki mice was homogenized in phosphate-buffered saline (PBS) using a Powergen homogenizer (Fisher Scientific). The internal standard, vitamin K 1(25) , was added to homogenates generated from the kidney, pancreas, and spleen. The concentration of MK-4 in these homogenates was subsequently determined by reverse-phase HPLC as described previously (Booth et al., 2008) using a C30 column that allows improved resolution. The MK-4 content of the liver, brain, and adipose tissue was determined as described (Fu et al., 2009;Harshman et al., 2016) by LC-MS using deuterium-labeled vitamin K 1 as an internal standard.
The histological analysis of tissues from Ubiad1 +/+ : :Hmgcr Ki/Ki and Ubiad1 -/-: :Hmgcr Ki/Ki mice was conducted by the Pathology Core at UT Southwestern Medical Center.

Quantitative real-time PCR (qRT-PCR)
Total RNA was prepared from mouse tissues using the RNA STAT-60 kit (TEL-TEST 'B', Friendswood, TX). Equal amounts of RNA from individual mice were treated with DNase I (DNA-free, Ambion/Life Technologies, Grand Island, NY). First strand cDNA was synthesized from 10 mg of DNase I-treated total RNA with random hexamer primers using TaqMan Reverse Transcription Reagents (Applied Biosystems/Roche, Branchburg, NJ). Specific primers for each gene were designed using Primer Express software (Life Technologies) or Primer Bank of Harvard University. The real-time RT-PCR reaction was set up in a final volume of 20 ml containing 20 ng of reverse-transcribed total RNA, 167 nM of the forward and reverse primers, and 10 ml of 2X SYBR Green PCR Master Mix (Life

Data availability
All data generated or analysed during this study are included in the manuscript and supporting files.