Cloning and spatiotemporal expression of Xenopus laevis Apolipoprotein CI

Apolipoprotein CI (ApoCI) belongs to the Apolipoprotein superfamily, members of which are involved in lipid transport, uptake and homeostasis. Excessive ApoCI has been implicated in atherosclerosis and Alzheimer’s disease in humans. In this study we report the isolation of Xenopus laevis apoCI and describe the expression pattern of this gene during early development, using reverse transcription polymerase chain reaction and whole mount in situ hybridization. Xenopus apoCI is enriched in the dorsal ectoderm during gastrulation, and is subsequently expressed in sensory placodes, neural tube and cranial neural crest. These data suggest as yet uncharacterized roles for ApoCI during early vertebrate embryogenesis.


Introduction
Apolipoproteins are protein constituents that bind to lipids via their amphipathic α-helical domains, forming lipoproteins of varying size and density [1]. They function in lipid metabolism by facilitating the transport, redistribution, and uptake of lipids between different tissues and by acting as cofactors for enzymes involved in lipid metabolism. The major apolipoproteins include apoAI, apoAII, apoAIV, apoB, apoCI, apoCII, apoCIII and apoE.
Human APOCI, at 6.6KDa, is the smallest member of the APOC subfamily; it is synthesized as a mature 57-residue peptide after cleavage of an N-terminal 26 amino acid signal peptide [2]. In circulation, APOCI is found as a surface component of High Density Lipoproteins (HDLs), Very Low Density Lipoproteins (VLDLs), and chylomicrons, and is exchanged between these forms during lipid metabolism [3].
APOCI is best known for its role in lipoprotein metabolism, primarily affecting intravascular turnover of triglyceride-rich lipoproteins (TRLs) that transport endogenous fat from the liver (VLDLs) and dietary fat from the intestine (chylomicrons). ApoCI inhibits hepatic remnant clearance by preventing ApoE-mediated binding and uptake of VLDLs by LDL receptor Related Protein (LRP), VLDL receptor (VLDLr) and LDL receptor (LDLr), and by inhibition of lipoprotein lipase activity [4][5][6][7]. Consistently, both endogenous mouse ApoCI expression in an ApoE deficient background and overexpression of human APOCI in either a wild type or an ApoE deficient background can lead to hypertriglyceridemia and atherosclerosis in mice [5, 8,  9]. Moreover, in humans, high APOC1 content of TRLs is considered a risk factor for early atherosclerosis [10].
Human APOCI is strongly expressed in the liver, and is also expressed in the lung, skin, spleen, adipose tissue, and brain [11]. In the brain, a polymorphism in the APOCI allele that leads to an increase in APOCI expression has been implicated as a risk factor in the pathogenesis of Alzheimer's disease (AD) [12][13][14]; consistently, expression of human APOCI in mice can lead to learning and memory impairment, perhaps as a result of alterations in brain lipid metabolism or by interfering with ApoE binding to β-amyloid peptides [12,13,15,16].
Despite our understanding of apolipoprotein function in lipid homeostasis and disease, the role of this protein family during early development has not been extensively addressed. Here, we describe the cloning and expression analysis of apoCI in embryos of the frog Xenopus laevis; our studies suggest important and perhaps conserved roles for ApoCI during gastrulation and subsequent patterning of the vertebrate ectoderm.

Materials and methods
All procedures were approved by the Queens College IACUC committee, protocol #160 1. Description of procedures. The animal system used for these studies is the African clawed frog, Xenopus laevis. Adult female and male frogs (over 1 year old) were used to produce embryos for the proposed experiments. I maintain a colony of about 135 female and 35 male frogs; veterinary care is provided by the Queens College Vivarium Facility, under the directorship of Marie Birne. The colony is inspected and cared for on a daily basis. To generate embryos, female frogs were injected with 600-800 units of human chorionic gonadotropin (HCG). 12 hours after injection, female frogs were gently squeezed to express eggs from the cloaca. The females were allowed to rest for 3-4 months between each spawning. Eggs were fertilized in vitro; to obtain sperm, male frogs were anesthetized with 3-aminobenzoic acid ethyl ester. A midline incision was made and testes were removed. While under anesthetic the male was sacrificed by decapitation, double pithed, and disposed of according to the guidelines of the American Veterinary Medical Association. The testes are viable for approximately two weeks at 4˚C; thus, testes from a single male were enough to fertilize eggs from about 20 females.
The research described here focuses on the study of early Xenopus laevis embryos, which were derived via in vitro fertilization and examined up to early tadpole stages (within approximately 48 hours after fertilization). To induce egg-laying, female frogs were primed by subcutaneous injection of 800uL recombinant human chorionic gonadotropin. 12 hours after injection, eggs were extruded by gently squeezing the females (this procedure is painless, and mimics the action of the male of the species during mating). Eggs were harvested in glass Petri dishes and fertilized by application of sperm from minced testes of Xenopus males-synchronous fertilization, required for these studies, can only be achieved via in vitro fertilization, necessitating the killing of males for testes removal. Males were euthanized prior to testes extraction; euthanasia was performed by decapitation following anesthesia, via tricaine methanesulfonate (TMS) immersion. For these studies, we sacrificed approximately 1 male frog every other week (sperm/testis culture can last up to 14 days at 4 degrees Celsius). Embryos were euthanized by immersion in an anesthetic overdose of TMS.
2. Justification for use of animals. We utilized biochemical and molecular biological ("in vitro") approaches in these studies; however, elucidation of the dynamic processes underlying early development also require an understanding of gene and cell regulation in both space and time, a resolution not generally provided by tissue culture experiments. We therefore must also study gene and protein function in the context of the intact, developing organism.
Embryos of the frog Xenopus laevis have a number of distinct advantages over other vertebrate developmental model systems. First, the source of biological material is abundant, since it is possible to generate a large number of embryos (thousands per day) that can grow in a simple buffered solution. Second, fertilization and embryonic growth in Xenopus is external, greatly facilitating the analysis of early developmental events. Third, early stage Xenopus embryos are large, and thus well-suited for microsurgical and microinjection techniques. Finally, and perhaps most important, a large body of experimental embryological data has been gathered from this model organism; in fact, much of our current understanding of early vertebrate embryology comes from the study of the amphibian embryo.
3. Procedures to limit discomfort, distress, pain, and injury. Some of the embryological studies described here involve eggs and embryos at blastula and gastrula stages of development, prior to formation of the nervous system; thus, these embryos are not in discomfort, stress, or pain. For those experiments that required the use of older embryos, animals were anesthetized in a solution of 3-aminobenzoic acid ethyl ester prior to fixation. Generating eggs is not a painful process for the female: the gentle squeezing required to release the eggs actually mimics the arm movements of the male during natural mating. The males required for this study were anesthetized with 3-aminobenzoic acid ethyl ester prior to sacrifice. Minimizing discomfort and stress for the colony is a requirement for generating healthy embryos; our frog facility has state of the art equipment for regulating water quality, pH, aeration, light, and temperature, in order to maintain the colony under optimal conditions. 4. Method of euthanasia. The only adult animals sacrificed for this research were approximately one male frog per week. Additionally, tadpole stage embryos were sacrificed for some of these studies. As described above, sacrifices were performed under complete anesthetic conditions so as to minimize pain, using methods consistent with the recommendations of the American Veterinary Medical Association (AVMA) Guidelines for the Euthanasia of Animals.

Cloning and isolation of Xenopus apoCI
Xenopus Apolipoprotein CI was isolated as a gene upregulated in ectodermal cells expressing an Engrailed repressor-HNF3β fusion protein in stage 11 Xenopus embryos. Library construction and screening was performed using a modified protocol originally described in [17]. Full-length Apolipoprotein CI was subsequently cloned from a λZapII Stage 28 head library [18] using the following primers: 3-2U: 5'-GATACAAAGTGACTCATC; 3-2D: 5'-GGCTCACTGTTGTGCAAA.

RNA preparation and RT-PCR
Xenopus laevis embryos were staged according to [19] and harvested at appropriate stages according to morphological criteria. Total RNA was prepared using RNA Bee RNA isolation reagent from Tel-Test Inc. RT-PCR was performed as described [20]. Primers used in this study are as follows: ODC

In situ hybridization
Whole mount in situ hybridization was carried out using standard protocols [21]. For preparation of antisense apoCI RNA probe, pBluescript SK+ 3-2 plasmid vector was linearized with Not I and transcribed in vitro with T7 RNA polymerase. For double in situ hybridization, probes were synthesized using digoxygenin or fluorescein RNA labeling mix (Roche) and detected with corresponding antibodies (Roche). First antibody reactions were quenched using 0.1M glycine-HCl, pH2.2 [22]. Magenta phosphate (Sigma), BCIP (Roche) and BM Purple (Roche) were used for chromogenic reactions. Other probes used in this study were Otx-A, Pax-6 and Slug [23,24]. 40 μM vibratome sections were obtained by embedding 4% paraformaldehyde-fixed embryos in 20% type B Bovine Gelatin (Sigma).

Results
Differential screening was used to identify transcripts enriched in dorsal tissues of gastrula stage Xenopus laevis embryos. One transcript isolated from this screen, clone 3-2, shares highest homology with members of the Apolipoprotein family from higher vertebrates. The predicted translation product of this cDNA (Fig 1A) is most closely related to human Apolipoprotein CI (hAPOCI; 38%). 3-2 shares significantly less sequence identity with human APOC2 (20%), APOC3 (17%) or other human apolipoproteins (Fig 1B). Mammalian Apolipoprotein CI is a secreted molecule that contains a twenty-six amino acid N-terminal signal sequence that is co-translationally cleaved to yield a mature peptide [2]; similarly, the first twenty residues of clone 3-2 are predicted to encode a signal peptide. The gene encoding the 3-2 transcript maps to Xenopus laevis chromosome 7L (XLA7L) (X. laevis 9.2 on Jbrowse); there is no apoCI-like gene on XLA7S, suggesting that there is no second identifiable apoCIlike allogene in the Xenopus laevis genome. Taken together, our data suggest that 3-2 encodes the Xenopus homolog of ApoCI. The temporal expression of Xenopus apoC1 was analyzed by reverse transcription polymerase chain reaction (RT-PCR) analysis of RNA extracted at various embryonic stages, from cleavage through early tadpole (Fig 2). apoCI expression is not observed at cleavage stage 4, suggesting that there is no maternal expression of this gene. apoCI expression is first detected at low levels at embryonic stage 8, which coincides with the midblastula transition, the start of zygotic gene expression [25]. Higher levels of apoCI expression are seen at stage 10, the onset of gastrulation, increasing progressively through stage 32.
Whole mount in situ hybridization analysis with an antisense probe for apoCI revealed an interesting spatially-restricted expression pattern in Xenopus embryos beginning at early gastrula stages. apoCI expression becomes apparent at stage 10, when it is expressed in an area bordering the dorsal lip of the involuting marginal zone that extends animally to include the dorsal ectoderm, a region that contributes to the central nervous system later in development (Fig 3A and 3A') [26,27]. The dorsal localization of apoCI becomes more evident with increased expression levels as gastrulation progresses (Fig 3B and 3B').
At stage 13, apoCI expression is seen in a crescent-like pattern in the anterior neural plate, in the region of the prospective lens and trigeminal placodes (Fig 3C and 3C') [28]. As neurulation proceeds, expression also becomes visible in the dorsal midline (Fig 3D and 3D'). Anterior apoCI expression then expands laterally to include expression in the neural crest (Fig 3E). In  situ hybridization using probes of both apoCI (Fig 3F) and the neural crest marker slug (Fig 3G) reveals overlap of apoCI with the anteriormost region of slug expression marking the cranial neural crest (Fig 3H). Anterior expression of apoCI at neurula stage flanks the expression of lens marker pax6 (see Fig 3F, 3I and 3J) but shows overlap with the more posterior expression domain of forebrain marker otxA (see Fig 3F, 3K and 3L). After neural tube closure, expression of apoCI is seen in anterior neural tissue and in the cranial neural crest, including the region surrounding the developing eye; expression of apoCI is excluded from both the cement gland and the adenohypophyseal stomodeal anlage (Fig 3M) [29]. Transverse sections of early neurula embryos reveal that transient midline expression of apoCI is localized to the dorsal neural tube (Fig 3N and 3O). At the late tailbud stage, expression is visible surrounding the lens and in the olfactory placode, the trigeminal nerve, and the tailfin (Fig 3P). Diffuse apoCI expression is observed throughout the anterior endoderm at tadpole stages, with strong expression in the liver primordium (Fig 3Q). Dorsal and ventral tailfin expression increases markedly at tadpole stages: apoCI expression is observed at the distal margin of the fin, and at the border between the inner fin and outer fin where the latter structure is present; apoCI expression is also observed in scattered cells in both the dorsal and ventral outer fin (Fig 3Q and data not shown).

Discussion
This work describes the identification and spatiotemporal expression of apoCI during early vertebrate embryogenesis. At gastrula stages, Xenopus apoCI expression is enriched in the dorsal ectoderm. During neurulation, apoCI is strongly expressed in the placodal primordium, the cranial neural crest and also the developing spinal cord. At tadpole stages expression is also seen in the anterior ventral endoderm, in the region of the presumptive liver, consistent with the strong expression of ApoCI in human and mouse liver.
The highly localized expression of Xenopus apoCI suggests a novel role for this factor in early neuroectodermal and/or cranial development, perhaps independent of its known role in lipid homeostasis. APOCI expression has been detected in human adult brain tissue, but analysis of localized expression is unavailable. In the Orange-spotted grouper, apoCI is also prominently expressed in the brain 7 days after hatching [30]. These studies suggest that ApoCI also functions during teleost and/or mammalian neurodevelopment.
Our studies do not directly address the mechanisms by which ApoCI may regulate steps in early development. The known functions of ApoCI in the transport of lipoproteins and modulation of binding to lipoprotein receptors raises the possibility that ApoCI may be involved in the modulation of lipid bound morphogens or lipoprotein receptors that are involved in early developmental signaling pathways, including those that mediate differentiation and/or patterning of the neurectoderm and neural border fates [4].