Expression of a long variant of CRACR2A that belongs to the Rab GTPase protein family in endothelial cells

Highlights • Knockdown of CRACR2A lacks effect on CRAC channels in endothelial cells.• Knockdown of CRACR2A depletes a protein twice the mass of CRACR2A.• A long variant of CRACR2A, CRACR2A-L, occurs in endothelial cells.• CRACR2A-L is a previously unrecognised EF-hand-containing Rab GTPase.• CRACR2A-L has a positive role in endothelial tube formation.

Orai1 and STIM1 proteins are widely expressed and Ca 2+ entry mediated by CRAC channels is suggested to be functionally important in many cell types. In endothelial cells, a positive role has been suggested in cell migration induced by the key mediator vascular endothelial growth factor [4]. Here we sought the relevance of CRACR2A to endothelial cells and CRAC channels of these cells.

Cell culture
Human umbilical vein endothelial cells (HUVECs) (Lonza) were maintained in endothelial growth medium (EGM-2) supplemented with 2% foetal calf serum (FCS) and growth factor bullet kit (Lonza). Experiments were performed on passage 2-8 cells. Cells were maintained at 37°C in a 5% CO 2 incubator. Other endothelial cells were from PromoCell and maintained similarly to HUVECs.

Western blotting
Transfected HUVECs were harvested in PBS and lysed in buffer containing 0.5% NP40, 10 ml Tris (pH 8), 150 mM NaCl, 0.5 mM EDTA, protease inhibitor cocktail (Fermentas) and 1 mM phenylmethanesulfonylfluoride (PMSF). Equal protein amounts were loaded on 10% gels and resolved by electrophoresis. Products were transferred onto PVDF membranes and incubated overnight with primary anti-CRACR2A antibody and anti-b-actin antibody used at 1:800 and 1:2000 dilution respectively. Anti-CRACR2A was visualised using horse radish peroxidase conjugated goat anti-rabbit secondary antibody and anti-actin was visualised with anti-mouse secondary antibody and SuperSignal Femto detection reagent (Per-Bio Science). ImageJ was used to quantify protein band densities.

RT-PCR
RNA was extracted using TRI reagent (Sigma). Reverse transcription using Reverse Transcriptase (Applied Biosystems), followed by real-time PCR using SYBR green was performed using a LightCycler (Roche). The forward primer was (5 0 -3 0 ) CGTCATGTA CGATCTCAC and the reverse primer GTGACCAGAGTAGGCG.

Cloning
Reverse transcription of HUVEC total RNA was performed using Superscript II Reverse Transcriptase (Invitrogen) and a CRAC2RA-L gene specific primer (5 0 GGTATGGCTGACCTTGTAGGACCCTATCA). CRACR2A-L sequence was amplified from the enriched cDNA using GoTaq Green (Promega) and cloning primers (Forward: 5 0 GAT-CCGCTAGCGCTACACCATGGCTGCCCCTGACGGG, reverse: 5 0 GGTAT GGCTGACCTTGTAGGACCCTATCA) designed to be compatible with the GFP vector using Infusion-HD enzyme (Clontech). PCR reaction conditions were: 95°C for 2 min (initial denaturation stage) followed by 40 cycles of: 95°C for 30 s, 53°C for 30 s, 73°C for 2.5 min followed by a final extension at 73°C for 5 min. CRAC2RA-L sequence was further amplified by a second round of PCR using the high fidelity DNA polymerase Phusion (Thermoscientific). Constructs were generated using the kanamycin resistant eGFP-C1 plasmid (Clontech). Constructs were verified by sequencing (Beckman Coulter).

Co-culture tube formation assay
Normal Human Dermal Fibroblasts were seeded into a 96 well plate (Greiner) at a density of 2 Â 10 3 cells/well and incubated at 37°C in a 5% CO 2 incubator for 72 h. The cell culture medium was removed and HUVECs at a density of 5.4 Â 10 4 cells/well were seeded on top of the fibroblasts in endothelial cell growth medium containing 3 ng/mL vascular endothelial growth factor and 2% FBS. Co-cultures were incubated for 7 days then fixed with ice-cold methanol for 10 min. Nonspecific binding sites were blocked with 5% donkey serum (in PBS) for 30 min at 37°C. HUVEC tubes were visualised by staining with a primary monoclonal mouse anti-human CD31 antibody (1:500 in 1% donkey serum for 1 h at room temperature). After 3 washes in PBS a secondary Alexa Fluor 488 antibody was added for 1 h at room temperature. Plates were imaged using an IncuCyte (Essen Bioscience) and tube formation was quantified using the IncuCyte angiogenesis v2.0 software.

Data analysis
Data were analysed and figures prepared using Origin 7.5 software (OriginLab Corporation). For FlexStation experiments data are presented as n/N, which indicates the number of independent experiments (n) and the number of individual wells (replicates) in the 96-well plate. For Western blot data n is the number of independent experiments. Paired data sets were compared using twotailed Student's t-tests and expressed as mean ± standard error of the mean (s.e.m.). P < 0.05 indicated statistically significant difference (*).

Results
Intracellular Ca 2+ was recorded from human umbilical vein endothelial cells (HUVECs) to observe Ca 2+ release evoked by thapsigargin (TG) in the absence of extracellular Ca 2+ and then CRAC channel-mediated Ca 2+ entry as extracellular Ca 2+ was added back (Fig. 1A). Unexpectedly, transfection with short interfering RNA (siRNA) targeted to CRACR2A failed to affect Ca 2+ release or Ca 2+ entry ( Fig. 1A and B). In contrast, siRNA targeted to Orai1 suppressed Ca 2+ entry but not Ca 2+ release, consistent with Orai1-dependent CRAC channels mediating Ca 2+ entry ( Fig. 1C and D). The data suggest that CRACR2A is unimportant for CRAC channels in these cells.
An explanation for the above data could be that CRACR2A is not expressed. Indeed, anti-CRACR2A antibody failed to detect protein of $45 kDa ( Fig. 2A), which is the expected mass of CRACR2A [3]. However, a doublet around 95 kDa was labelled by the antibody and the lower of these two bands was depleted by two different siRNAs targeted to CRACR2A ( Fig. 2A and B). The upper band may reflect an unrelated protein, labelled non-specifically by the antibody. The data suggest that CRACR2A is expressed in endothelial cells but that it occurs at about twice the expected molecular mass.
To obtain an explanation for the larger mass we searched genome sequence databases and found that the gene encoding CRACR2A, EFCAB4B, is predicted to be alternatively spliced to give two variants. Alignment of the predicted amino acid sequences of the short (CRACR2A) and long (CRACR2A-L) variants revealed that the N-terminal components are identical but that CRACR2A-L contains a distinct long C-terminus bearing little or no similarity to CRACR2A (Fig. S1). The analysis suggests that the $95 kDa protein might be CRACR2A-L.
To determine if endothelial cells express transcript encoding CRACR2A-L we performed full-length cloning based on the predicted sequence and using HUVEC mRNA. Sequencing showed the clone to be identical to the sequence predicted to encode CRACR2A-L except for one codon for an additional amino acid between serine 424 and glutamine 425 (Fig. S1). The extra codon was also detected in human dermal microvascular endothelial cells and RT-PCR detected mRNA encoding CRACR2A-L in endothelial cells from 6 additional sources (Fig. 2C). The data suggest that the long variant, CRACR2A-L, is the variant of endothelial cells. Analysis of the distinct long C-terminus of CRACR2A-L indicated G boxes and other sequence similarities characteristic of Rab GTPases (Fig. 3). Construction of a dendrogram placed CRACR2A-L as a previously unrecognised member of the large Rab protein family (Fig. S2). It is relatively unusual amongst the Rab proteins in containing EF hands (Fig. 3), Ca 2+ -binding motifs that have been shown to be functional in CRACR2A [3].
To determine if CRACR2A-L has functional significance in endothelial cells we performed an assay in which HUVECs form tube, capillary-like, structures on a bed of fibroblasts, mimicking aspects of angiogenesis (Fig. 4A). Depletion of CRACR2A-L did not prevent tube formation but reduced the overall length of tubes and the number of branch points ( Fig. 4B and C). The data suggest that CRACR2A-L is a functional protein of endothelial cells.

Discussion
The results suggest a previously unrecognized protein of the Rab GTPase family. It is encoded by the same gene as CRACR2A, a protein expressed in T cells and regulating CRAC channels, but alternative splicing enables a long C-terminal extension and total protein of about twice the size of CRACR2A. We refer to it as CRACR2A-L. The N-terminus is identical to CRACR2A. It is the additional C-terminal sequence that makes it a member of the Rab GTPase family. CRACR2A-L is expressed in endothelial cells, whereas CRACR2A is not. CRACR2A-L has no effect on CRAC channels in these cells and so we assume that the C-terminal sequence directs the protein away from these channels.
Rab proteins comprise the largest sub-group of the RAS superfamily and are best known for roles in intracellular trafficking [5,6]. Most Rab proteins are 20-25 kDa [7] and thus smaller than CRACR2A-L. Rab44 and Rab45 are, by contrast, similar in size to CRACR2A-L at 108 and 83 kDa [5,8]. As in CRACR2A-L the extra length appears to exist for incorporation of EF-hands and thus Ca 2+ regulation. We speculate therefore that CRACR2A-L might be a mechanism for coupling cytosolic Ca 2+ elevation to protein trafficking which in turn impacts on endothelial tube formation and potentially other as yet unidentified downstream processes. It would be challenging to test this idea through studies of tube formation because the impact of CRACR2A-L on this process was relatively small. Therefore, it will be important to identify specific downstream effectors and other downstream implications of CRACR2A-L in order to understand its suggested integrative role.
It is intriguing why the gene encoding this CRACR2A-L protein would also be capable of generating the shorter CRACR2A protein which regulates the Ca 2+ -permeable CRAC channels [3]. Although there was only CRACR2A-L in endothelial cells, regulation of alternative splicing may lead to co-expression of CRACR2A and CRACR2A-L in some contexts. In such contexts, CRACR2A would be expected to have a positive effect on CRAC channels, elevating the cytosolic Ca 2+ concentration and potentially modulating the function of CRACR2A-L, integrating Ca 2+ -signalling and protein trafficking via a single gene.
We observed mild but significant positive impact of CRACR2A-L on endothelial cell remodelling. Additional functions of CRACR2A-L can be expected to emerge, however, and it would be premature to suggest that regulation of endothelial cell remodelling is its primary function.