Cargo Receptor-Mediated ER Export in Lipoprotein Secretion and Lipid Homeostasis

Abstract

APOB-containing lipoproteins are large, complex lipid carriers that ferry bulk lipids into the circulation via the secretory pathway, originating from the endoplasmic reticulum of specialized cells in the liver or the gut. Elevation of APOB-containing lipoproteins in the plasma represents a major risk factor for cardiovascular diseases. The production of these lipoproteins requires enzyme-catalyzed, cross-membrane transfer of neutral lipids and phospholipids to lipoproteins, in particular onto the structural component APOB. Transport of these lipid-bearing cargos relies on the COPII machinery and employs the transmembrane cargo receptor SURF4 and the small GTPase SAR1B, together constituting a selective transport program. Intriguingly, a number of factors implicated in lipoprotein production are also packaged into COPII vesicles and may be cotransported with APOB. These observations therefore point to a specialized produce-and-export itinerary during the secretion of these lipid-bearing cargos, warranting future investigations into this unique yet pivotal process at the crossroad of cell biology and physiology.

The emergence and establishment of multicellular organisms relies on constant exchange of substance and information among different cell types, in a manner largely mediated by the secretory pathway (Lee et al. 2004; Jensen and Schekman 2011). At the cellular level, vesicle transport via the secretory pathway also enables communication and coordination between various membrane-bound organelles, characterized by their distinct morphology and function (Palade 1975; Bonifacino and Glick 2004). Among these, the endoplasmic reticulum (ER) represents the largest endo-membrane system and the origin of the cellular secretory pathway (Gurkan et al. 2007; Brandizzi and Barlowe 2013; Barlowe and Helenius 2016). Numerous secretory and transmembrane proteins, encoded by ∼30% of the mammalian genome, are initially synthesized in the ER prior to distribution en route to their destined sites of function (Kanapin et al. 2003; Carninci et al. 2005). In specialized cell types in vivo such as the hepatocytes (in the liver) and the enterocytes (in the small intestine), large quantities of lipids are secreted into the circulation via specialized lipid carriers called lipoproteins that are assembled in the ER (Schonfeld et al. 2005; Gillon et al. 2012; Goldstein and Brown 2015). Moreover, dysregulated lipoprotein transport and metabolism are intimately linked with cardiovascular diseases (CVDs), the top killer in the world (Goldstein and Brown 2015). The protein or lipid-bearing client cargos differ drastically in quality and quantity, therefore posing a major challenge to accommodating cargo transport from the very first step, export from the ER.

To exit the ER, most secretory cargos are packaged into transport vesicles generated by the coat protein complex II (COPII). Work over the previous several decades has shown the fundamental mechanism by which the COPII operates, in an evolutionarily conserved manner (Bonifacino and Glick 2004; Gurkan et al. 2007; Zanetti et al. 2012; Lord et al. 2013). In essence, the assembly of the COPII complex begins with the activation of a small Ras-like GTPase named SAR1 by the ER-localized guanine nucleotide exchange factor (GEF) SEC12. GTP-loaded, active SAR1 exposes its amino-terminal amphipathic α-helix that inserts into the ER membrane and initiates the recruitment of the inner coat SEC23/SEC24 heterodimer and subsequently the outer coat SEC31/SEC13 heterotetramer, together generating COPII-coated vesicles for ER export (Fig. 1; Barlowe et al. 1994; Jensen and Schekman 2011).

Figure 1.

Sequential assembly of the coat protein complex II (COPII) coats for transport vesicle production and cargo packaging at the endoplasmic reticulum (ER). The orderly production of COPII-coated vesicles is mediated by the following sequence of events: (1) SAR1 activation, (2) inner-coat (SEC23/24) recruitment and cargo engagement, (3) outer-coat (SEC31/13) recruitment, and (4) vesicle scission and departure for the Golgi.

Activated SAR1 and SEC23/SEC24 form a 15 nm long, bow-tie-shaped “pre-budding complex,” in which SEC23 binds SAR1 and SEC24 is mainly responsible for cargo recruitment. The “pre-budding complex” has a concave inner face enriched with basic residues, conforming to the underlying curved membrane, thereby facilitating the interaction with the ER membrane and transmembrane proteins (Bi et al. 2002; Miller et al. 2003). The heterotetrameric SEC31/SEC13 complex, composed of two SEC31 and two SEC13 subunits, forms the outer layer and may generate cumulative effects via multivalent interactions between SEC31 and SEC23 (Stancheva et al. 2020). SEC31/SEC13 heterotetramers, with ∼30 nm rod-like shapes, polymerize into a cage-like structure, thereby serving as a scaffold to rigidify and shape the vesicles (Stagg et al. 2006; Fath et al. 2007; Čopič et al. 2012; Hutchings et al. 2021). COPII paralogs further increase the combinations of COPII interactions to generate coat diversity for selective transport in vivo (Mancias and Goldberg 2008; Merte et al. 2010; Wansleeben et al. 2010; Jensen and Schekman 2011; Chen et al. 2013). Moreover, the COPII complex contains built-in GTPase activating protein (GAP) activity to inactivate SAR1, thus allowing the disassembly and recycling of the coat machinery after vesicle formation (Yoshihisa et al. 1993; Antonny et al. 2001; Bi et al. 2007).

COPII-coated vesicles are usually 60–90 nm in diameter, sufficient to transport many “conventional” cargos (Barlowe et al. 1994; Gürkan et al. 2006; Zeuschner et al. 2006). However, certain COPII-dependent cargos, such as the over 250 nm-long pro-collagen I (PC1) and the complex lipoproteins, appear to be oversized for the regular COPII vesicles (Hussain 2000; Saito et al. 2009; Gorur et al. 2017; Yuan et al. 2018). While elegant work on the packaging of PC1 into COPII vesicles is summarized in Raote et al. (2022), we focus on the lipid-bearing lipoproteins as an example of ER export and secretion in selective physiological processes. Here we summarize our current understanding of ER export of the unique yet abundant lipid-bearing carriers with pivotal functions in health and disease, by focusing on (1) the cargo, (2) the cargo receptor, and (3) the cargo/receptor complex, as well as future perspectives.

THE CARGO: APOB-CONTAINING LIPOPROTEINS

Among the numerous secretory cargoes, APOB (apolipoprotein B)-containing lipoproteins are distinguished by their structure and function. These lipid carriers are complex particles assembled with a hydrophobic central core containing triglycerides and cholesterol esters, enveloped by a phospholipid monolayer and structurally supported by amphipathic apolipoproteins (Goldstein and Brown 2015). APOB-containing lipoproteins deliver bulk lipids (triglycerides, cholesterol, and phospholipids) into circulation to serve as fuel supplies, structural sources, or signaling molecules. Two types of APOB-containing lipoproteins, very-low-density lipoproteins (VLDLs) and chylomicrons (CMs), transport endogenously synthesized lipids by the liver and dietary lipids absorbed from the small intestine, respectively (Feingold 2000). Importantly, dyslipidemia characterized particularly by an elevated quantity of circulating APOB-containing lipoproteins represents a leading risk factor for cardiovascular diseases (Goldstein and Brown 2015).

The Apolipoprotein B (APOB)

Both CM and VLDL employ the amphipathic apolipoprotein B (APOB) as the primary structural protein, which cannot be exchanged between the lipoprotein particles (Feingold 2000). Two forms of APOB protein, APOB100 and the shorter form APOB48, are produced by one single APOB gene. Human APOB100 is one of the largest known monomeric proteins and contains 4536 amino acids, including an amino-terminal globular domain, two large hydrophobic regions separated by an amphipathic region, and another amphipathic domain at the carboxyl terminus (Ginsberg 2021). APOB48 arises from an RNA editing event catalyzed by the enzyme APOBEC1, leading to a truncated version that contains the amino-terminal 48% portion of APOB100 (Davidson and Shelness 2000). Human APOBEC1 is exclusively expressed in the gut, rendering VLDL to use APOB100 and CM to use APOB48 in humans, respectively. In mice and rats, APOBEC1 is also expressed in the liver and therefore the rodent liver produces both APOB100 and APOB48 to generate VLDLs (Davidson and Shelness 2000). Stoichiometrically, each APOB-containing lipoprotein employs one APOB molecule, thus the level of plasma APOB proteins could better reflect the quantity of lipoproteins and serve as a more reliable indicator for the risk of CVD than the levels of plasma lipids (Ference et al. 2020; Marston et al. 2022).

The Assembly and Lipidation of APOB

The assembly of APOB-containing lipoproteins begins in the rough ER of hepatocytes or enterocytes, in concert with translation of the APOB protein (Pease et al. 1991; Borén et al. 1992). Following the synthesis of the signal peptide spanning residue 1–27, the nascent APOB is cotranslationally translocated into the ER lumen, through the SEC61 translocon complex (Young 1990). The correct folding of APOB in the ER lumen requires cotranslational lipidation, otherwise the misfolded APOB molecules are cotranslationally targeted to proteasomal degradation as well as to other ER-associated degradation (ERAD) pathways (Fisher and Ginsberg 2002; Rutledge et al. 2010). The initial lipid loading requires the ER resident microsomal triglyceride transfer protein (MTP), a heterodimeric complex containing a catalytic MTPα subunit and a protein disulfide isomerase (PDI) subunit (Gordon et al. 1995; Hussain et al. 2003). The role of the PDI subunit in the heterodimeric MTP complex is likely to maintain MTPα in a stable form and also retain the complex in the ER by virtue of its carboxy-terminal KDEL retrieval signal (Hussain et al. 2012; Walsh et al. 2016). The lipid transfer activity of the MTP complex is vital for the stabilization of APOB protein and the subsequent assembly of APOB-containing lipoproteins. Although the exact mechanism by which MTP transfers neutral lipids across the ER membrane to the lipoproteins remains unclear, recent structural analysis revealed that the carboxy-terminal β-sandwich domain of MTPα contains a hydrophobic lipid-binding cavity that appears to accommodate a small number or even one single triglyceride molecule, whereas the rest of the MTPα interacts with ApoB and PDI (Biterova et al. 2019). Given the role of PDI in assisting protein folding, it remains possible that MTPα adopts different protein folding states in the lipid-bound and apo forms, thereby transferring neutral lipids to nascent ApoB to first stabilize the polypeptide and then to form mature, lipid-loaded lipoproteins (Banaszak and Ranatunga 2008). Loss-of-function mutations in the MTP gene result in abetalipoproteinemia, a rare genetic disorder characterized by a deficiency of APOB-containing lipoproteins in the plasma (Wetterau et al. 1992). Genetic inactivation of the murine Mtp gene recapitulates the human rare disease in mouse models (Raabe et al. 1998, 1999). Inhibitors targeting MTP activity have been developed to treat hyperlipidemia (Wetterau et al. 1998).

The addition of bulk lipids to APOB-containing lipoproteins is generally thought to require a second-step lipidation, which appears to take place along the early secretory pathway (ER-to-Golgi) (Olofsson et al. 2000; Brodsky and Fisher 2008; Yao et al. 2013). Whereas the precise site and the exact mechanism of the bulk lipid transfer required for the maturation of APOB-containing lipoproteins remains elusive, the process appears to be independent of MTP (Pan et al. 2002). TM6SF2, a polytopic protein that may cycle between the ER and the Golgi, participates in the maturation of lipoprotein particles. Humans with a TM6SF2 missense mutation (E167K) have increased risk of developing nonalcoholic fatty liver disease (NAFLD), while exhibiting reduced levels of plasma lipids (Holmen et al. 2014; Kozlitina et al. 2014). Inactivation of TM6SF2 in mice or rat recapitulates the fatty liver phenotype and causes a reduction in VLDL size and plasma lipid levels (Li et al. 2020; Luo et al. 2022). These data implicate a role for the transmembrane protein in the lipidation and maturation of VLDLs. CIDEB, an ER/lipid droplet-associated protein, also contributes to VLDL maturation (Ye et al. 2009), thus implicating an active involvement of lipid droplets in supplying bulk lipids to the outbound lipoproteins.

APOB-containing lipoproteins also acquire large amounts of phospholipids, which form a monolayer to cover the surface of the maturing particles in the ER/Golgi lumen. The transfer of phospholipids for lipoprotein production requires the ER-localized phospholipid scramblase TMEM41B (Huang et al. 2021), which will be discussed later. Additional factors including PLTP and the TorsinA-LAP1 complex also participate in the assembly and maturation of lipoproteins in the ER lumen (Jiang et al. 2012; Shin et al. 2019). Furthermore, incorporation of arachidonate into phospholipids in the ER membrane, mediated by the enzyme LPCAT3, increases ER membrane fluidity to facilitate the transfer of triglycerides to lipoproteins (Hashidate-Yoshida et al. 2015; Rong et al. 2015). Taken together, despite great advances in our understanding of the synthesis and assembly of the APOB-containing lipoproteins, numerous questions remain with respect to both the fundamental cell biology of the cross-membrane lipid shuttling and the relevance to prevalent diseases such as NAFLD, obesity, and CVD.

APOB Endocytosis and Human Diseases

One of the classic discoveries that has impacted both cell biology and medicine concerns the endocytic uptake of APOB-containing lipoproteins from the blood (Goldstein and Brown 2009). The work originated from studying a rare genetic disease, familiar hypercholesterolemia (FH), which is inherited in an autosomal-dominant manner. Individuals with homozygous FH (HoFH) exhibit six- to ten-fold elevations in plasma APOB-containing lipoproteins (low density lipoprotein [LDL]) at birth, leading to CVD in early childhood and premature mortality. The more common heterozygous FH (HeFH, occurs in one per ∼500 people) causes ∼two-fold higher levels in plasma LDL as well as higher risks of CVD in these patients (Goldstein and Brown 2015; Foody and Vishwanath 2016).

It is now known that FH is caused by defects in the uptake and clearance of circulating LDL orchestrated by the LDL receptor (LDLR) (Goldstein and Brown 2015). Cell-surface-localized LDLR binds APOB in the circulating LDL, thereby initiating the process known as receptor-mediated endocytosis in which the lipoproteins are selectively taken up by the cells (Brown and Goldstein 1986; Goldstein and Brown 2009; Young and Fong 2012). The receptor engages a cytosolic coat protein called clathrin to generate coated transport vesicles (Pearse 1976; Anderson et al. 1977), in a manner facilitated by adaptor proteins including one named ARH (Garcia et al. 2001; He et al. 2002). The internalized LDLs are eventually targeted to the lysosome, whereas the LDLR dissociates from the cargo due to the low pH in the lumen of the endolysosome system. After dissociation, the freed LDLR recycles back to cell surface and repeats this round trip every 10 min, thus enabling efficient uptake of LDL into cells (Brown et al. 1983). LDLR recycling is regulated by additional factors such as the secretory protein PCSK9, which interacts with and directs LDLR to the lysosome for degradation (Abifadel et al. 2003; Cohen et al. 2006; Lagace et al. 2006; Horton et al. 2009). Collectively, mutations in LDLR, APOB, and PCSK9 account for ∼95% of the FH cases (Soutar and Naoumova 2007; Henderson et al. 2016). Conversely, means to elevate the LDLR and consequently the selective uptake of circulating lipoproteins has led to the development of widely used medications in cardiovascular diseases (Brown and Goldstein 2006; Raal et al. 2012; Stein et al. 2012; Fitzgerald et al. 2017).

THE CARGO RECEPTOR: THE ER TRANSMEMBRANE PROTEIN SURF4

Compared to the endocytic trafficking that enables APOB-containing lipoproteins to enter the cells from the plasma membrane, the secretory route that escorts these lipid carriers from the ER is much less well understood. Moreover, the notion of receptor-mediated secretion was thought to represent a boutique mechanism inadequate for high-capacity transport (Warren and Mellman 1999), and evidence for the requirement of selective receptor(s) in high-capacity transport is still emerging (Nichols et al. 1998; Zhang et al. 2003; Wang et al. 2021b), especially in higher eukaryotes. Based on the evidence summarized below, however, one might reason that a dedicated cargo receptor could connect to the COPII machinery and initiate specific transport programs, thereby ensuring the delivery of the specialized yet abundant lipid carriers with vital roles in physiology.

Human Genetics of Lipoprotein Secretion

Human genetics provided important hints for the potential selectivity in the ER export and secretion of APOB-containing lipoproteins. Germline mutations in SAR1B, one of the two SAR1 paralogs in humans, result in chylomicron retention disease (CMRD, or Anderson disease) (Jones et al. 2003; Annesi et al. 2007). These patients present a distinct defect in fat absorption due to reduced chylomicron secretion by intestinal enterocytes, which can be managed by a low-fat diet to help the individuals thrive (Jones et al. 2003; Annesi et al. 2007; Levy et al. 2019). Of note, when murine Sar1b was selectively ablated in the liver, the mutant mice displayed a ∼90% reduction in plasma lipids and lipoprotein levels at fasted states, due to blockade of hepatic VLDL secretion. Meanwhile, protein secretion from the liver was largely unaffected by Sar1b deficiency (Wang et al. 2021c). These observations uncovered a segregation of lipoproteins from general protein transport at the entrance of the secretory pathway, and indicated the involvement of additional factor(s) that link the cytosolic transport machinery and the lumenal lipid carriers.

Despite much progress aided by the powerful genetics of yeast Saccharomyces cerevisiae, identifying cargo receptors and pairing with their cognate cargos upon ER export remain challenging (Barlowe 2003; Lee et al. 2004; Baines and Zhang 2007; Dancourt and Barlowe 2010), given the intrinsic cycling of the COPII complex and the dynamic nature of vesicle transport (Antonny et al. 2001; Nie et al. 2018). The restricted production and secretion of lipoproteins in vivo by the liver and the gut pose additional technical challenges for molecular and cell biology investigations. By contrast, great efforts have been made to understand the genetic basis of plasma lipid levels as common yet quantitative traits in humans (Teslovich et al. 2010; Willer et al. 2013; Liu et al. 2017; Lu et al. 2017). Unlike the linkage-based Mendelian diseases such as CMRD or FH in which a single disease gene can be pinpointed, the large-scale genome-wide association studies (GWAS) in populations mostly identify noncoding or regulatory sequence variants. Moreover, these variants are innumerable and often in linkage disequilibrium with the putative causal gene(s). Hence, a combination of genetic, biochemical, and cell biology approaches may need to converge to uncover lynchpins in selective processes involved in ER export and secretion in vivo.

SURF4 and Lipoprotein Secretion

SURF4 has emerged from proximity-based proteomics aiming to identify partners of SAR1B, whereas population-based genetics discovered a strong association between plasma LDL-cholesterol and sequence variants in SURF4 in humans (Wang et al. 2021c). The noncoding variant alters SURF4 gene expression in a manner correlating with plasma LDL levels, indicating that even quantitative changes in SURF4 level could impact lipid homeostasis. Indeed, complete inactivation of hepatic Surf4 depletes plasma lipids and lipoproteins to near zero in fasted mice due to a selective blockade of lipoproteins in the ER and hepatic Surf4 haplo-insufficiency correspondingly lowers plasma lipids. Hence, the genetic data in humans and mice demonstrate a pivotal role of SURF4 in lipoprotein transport and lipid homeostasis in vivo, and the dosage-sensitive effects of SURF4 in lipid transport further supports a model of receptor-dependent lipoprotein delivery.

SURF4 in mammals is a 269 amino acid protein with eight transmembrane regions as predicted by AlphaFold, orthologous to the COPII cargo receptor Erv29 in yeast (Belden and Barlowe 2001; Otte and Barlowe 2004). SURF4 oligomerizes and can be efficiently packaged into COPII-coated transport vesicles destined for the Golgi (Wang et al. 2021c), and retrieved back to the ER from the Golgi via its carboxy-terminal COPI-sorting motif (Jackson et al. 2012). The Caenorhabditis elegans ortholog SFT-4 is essential for secretion of some yolk proteins including VIT2, a lipid carrier protein homologous to APOB (Saegusa et al. 2018). In cultured cells such as HEK293, SURF4 also mediates the secretion of PCSK9, erythropoietin (EPO), and an array of soluble cargos with amino-terminal IPV-like tripeptides (Emmer et al. 2018; Yin et al. 2018; Lin et al. 2020). More recently, SURF4 was demonstrated to mediate the ER export of sonic hedgehog (Shh) from cells and form a relay with proteoglycans (Tang et al. 2022), although inactivation of murine Surf4 in neuronal systems in vivo did not present any obvious developmental defects (Y Guo and X-W Chen, pers. comm.). The ER retrieval of SURF4 is also involved in the proper shuttling of STING, presumably after the ER export of the key regulator in innate immunity (Deng et al. 2020; Mukai et al. 2021).

Despite affecting a broad range of cargos in cell culture, SURF4 exhibits a striking priority for lipoproteins in vivo in the liver. Analysis of the SURF4 interactome in mouse liver shows that SURF4 specifically enriches apolipoproteins including APOB and lipoprotein-associated proteins over other abundant hepatic secretory proteins such as albumin or α1 antitrypsin (Wang et al. 2021c). Along the same line, hepatic SURF4 deficiency causes little alteration of general protein secretion from the liver, while the secretion of lipoproteins becomes diminished. Furthermore, Surf4 liver-specific knockout mice appear grossly healthy except for mild lipid accumulation in hepatocytes, with no obvious signs of NASH or liver damage. These data are therefore consistent with the notion that, in physiological settings, the ER export of lipoproteins is a selective process. The capacity of lipoprotein transport could be ensured by the efficient recycling of SURF4 back to the ER, enabling subsequent rounds of lipoprotein export.

Taken together, SURF4-mediated ER export of lipoproteins appears to resemble many aspects of receptor-mediated endocytosis of LDL (Brown and Goldstein 1986). In this regard, both the “out-bound” and “in-bound” transport of APOB-containing lipoproteins employ transmembrane receptors to mediate the specific recognition of lipoproteins and their packaging into coated vesicles, while the transport efficiency is further enhanced by receptor recycling (Goldstein and Brown 2009). The numerous regulators centered on LDLR, in both endocytosis and lipid homeostasis (Goldstein and Brown 2009), further suggests the existence of additional partners for SURF4 and more broadly the presence of other receptor-mediated ER export events for selective cargos in specific physiological processes (Fig. 2).

Figure 2.

A working model of cargo receptor-mediated endoplasmic reticulum (ER) export of lipoproteins (very-low-density lipoprotein [VLDL]) following their production initiated in the ER lumen of hepatocytes. Hepatic production and transport of amphipathic apolipoprotein B (APOB)-containing lipoproteins requires the synthesis and stabilization of APOB, the continuous lipidation of APOB into lipoproteins (VLDLs), the transport of VLDLs via a selective coat protein complex II (COPII)-dependent ER export program, which is characterized in part by the cargo receptor SURF4 and the small GTPase SAR1B, and the recycling of key factors back to the ER for subsequent continuous rounds of transport. The APOB polypeptide is colored in red.

THE CARGO-RECEPTOR COMPLEX

Isolation of SURF4-associated proteins by selective immunoprecipitation from mouse liver captures multiple proteins that are likely involved in the assembly and transport of lipoproteins, and their presence in the SURF4 complex is increased when lipoproteins are stalled in the ER by Sar1b deficiency. This observation suggests that additional factors may be recruited and coupled spatially to assemble a “functional receptome” (Fig. 3), which could coordinate both lipoprotein production and transport in a manner maximizing the efficiency of lipid ferrying. Current data supports the presence of enzymatic factors that catalyze the production of lipoproteins, and auxiliary adaptors that facilitate their transport.

Figure 3.

A putative functional cargo-receptor complex that may enable a “produce-and-export” mechanism for bulk lipid transport from the endoplasmic reticulum (ER). (Left) Hepatic SURF4 complex isolated from wild-type or liver-specific SAR1B deficient mice. Arrow indicates amphipathic apolipoprotein B (APOB). (Right) Proposed coupling between the cargo receptor SURF4 and factors including coreceptor(s), enzymes mediating lipid synthesis and transfer, and other regulatory proteins modulating the production or transport of APOB-containing lipoproteins.

Enzymatic Factors

The assembly of lipoproteins requires increasing the amount of amphipathic phospholipids to cover and stabilize the expanding surface. Phospholipid biosynthesis is strictly asymmetric and takes place exclusively at the cytosolic leaflet of the ER membrane, whereas lipoproteins assemble in the ER lumen (Vance 2014, 2015). Hence, the production of APOB-containing lipoproteins exemplifies a long-standing question in cell biology and metabolism—transbilayer movement of phospholipids at the biogenic ER membrane (Pomorski and Menon 2016). TMEM41B, a long-sought ER-localized phospholipid scramblase (Ghanbarpour et al. 2021; Huang et al. 2021), forms a ternary complex with both SURF4 and APOB and is required for supplying phospholipids for the biogenesis of APOB-containing lipoproteins. Loss of hepatic TMEM41B eliminates plasma lipids, due to complete absence of mature lipoproteins within the ER lumen, prior to their secretion mediated by SURF4 (Huang et al. 2021).

TMEM41B, as well as related proteins such as VMP1, belongs to the VTT family (Okawa et al. 2021), which has been implicated in a variety of biological processes including lipid metabolism, autophagy, and viral infection (Moretti et al. 2018; Morita et al. 2018, 2019; Morishita et al. 2019; Shoemaker et al. 2019; Hoffmann et al. 2021; Reinisch et al. 2021; Schneider et al. 2021). Consistent with the fundamental and universal nature of phospholipid mobilization and membrane homeostasis, TMEM41B deficiency leads to extremely rapid progression of liver diseases and triggers drastic morphological alterations of the hepatic ER membrane. These pathologies reflect the detrimental consequents arising from imbalanced inner and outlet leaflets of the ER bilayer characterized by the uneven distribution of phospholipids due to defective transbilayer lipid scrambling (Huang et al. 2021). Intriguingly, VMP1 appears to play a lesser and perhaps indirect role in lipoprotein biogenesis and lipid homeostasis, whereas combined deficiency of both VMP1 and TMEM41B greatly exaggerate pathologies at the cellular and whole-body level (D Huang and X-W Chen, unpubl.). These results suggest ubiquitous yet selective roles of the newly discovered ER scramblases in cells, pointing to a novel produce-and-protect mechanism for organelle function and integrity (Huang et al. 2021).

TMEM41B and other factors implicated in lipoprotein production, including CIDEB and TM6SF2, interact with APOB and can be packaged into COPII vesicles and shuttle between the ER and the Golgi, likely due to sorting motifs in these transmembrane proteins (Ye et al. 2009; Su et al. 2019; Huang et al. 2021; Luo et al. 2022). The enrichment of enzymatic factors with the lipoprotein in COPII transport vesicles raises the possibility of cotransport maturation of lipoproteins en route to the Golgi and indicates other enzymatic factors that remain to be identified in the carrier vesicles.

Auxiliary Cofactors

Despite the profound impact of SURF4 in lipid homeostasis and its well-recognized identity as a COPII cargo receptor in cells, the molecular aspects of cargo recognition are far from clear. While being efficiently packaged into COPII vesicles, SURF4 and its yeast ortholog Erv29 appear to lack well-defined sorting motifs for the COPII complex (Barlowe 2003). This raises the possibility of potential coreceptors of SURF4, which may further assign cargo selectivity. To date, only a handful of adaptors have been identified to accompany transmembrane proteins out of the ER. Cornichon proteins (Erv14 in yeast) in the TMED superfamily (p24 family in yeast) are required for the proper delivery of membrane proteins including TGFα, AMPA receptors, and GPCRs (Bökel et al. 2006; Schwenk et al. 2009; Sauvageau et al. 2014). Of note, some Erv14 clients such as the ABC transporters and Yor1 contain their own sorting motifs, but still employ Erv24 for efficient ER export (Louie et al. 2012; Pagant et al. 2015).

A number of regulatory factors in membrane transport have also been implicated in the transport of lipoproteins. These include TANGO1 (the cargo receptor of procollagen as supersized COPII cargos), TALI/Mea6 (a TANGO1 like protein generated from a chimeric transcript of Mia2 and cTAGE5), and KLHL12 (an E3 ubiquitin ligase that regulates the shape of the COPII coats) (Jin et al. 2012; Butkinaree et al. 2014; Santos et al. 2016; Wang et al. 2016). Whether any or all these factors, or even additional trafficking regulators (Cai et al. 2007), may pair with SURF4 to generate transport vesicles of sufficient size would be of great interest, perhaps toward a general understanding of the transport of uncanonical cargos in vivo.

CONCLUDING REMARKS AND FUTURE PERSPECTIVES

Despite being a major determinant of circulating lipid levels and intimately associated with the risks of cardiovascular diseases, the secretion of APOB-containing lipoproteins remains an elusive yet challenging problem, especially considering the high quantity, large size, and restricted tissue distribution of these lipid-bearing cargos (Ginsberg 2021). The current data summarized above support a model of selective transport of these unique yet abundant cargos in vivo, characterized in part by the transmembrane cargo receptor SURF4 and likely assisted by other factors that together form a functional cargo receptor complex.

While the model of receptor-mediated ER export of lipoproteins may provide a “solution” for the issue of selectivity (Wang et al. 2021a,c), it undoubtedly creates a cohort of new “problems,” including the molecular basis of lipoprotein recognition, the coordination with lipid synthesis and transfer, and the mechanism of cargo dislodging, etc. Moreover, how the receptor and its partners engage and mobilize a sufficient amount of COPII coats to ensure the transport capacity for the privileged lipid carriers in vivo remains unknown. Of note, combined haploinsufficiency of SURF4 and the small GTPase SAR1B shows a synergistic effect in plasma lipid lowering, resulting in profound protection from atherosclerosis (Wang et al. 2021c). These data further suggest a yet-to-be defined dosage mechanism in lipoprotein ER export catalyzed by COPII (X Wang and X-W Chen, unpubl.). Last, whether these mechanistic studies might uncover novel targets or even therapeutics for lipid lowering or liver diseases deserves future investigation, particularly when building on the landmark work of receptor-mediated lipoprotein endocytosis and COPII biology.