Identification, Characterization, and Tissue Expression Pattern of Alternatively Spliced Transcript Variants of Mouse RARRES2 Gene


 Chemerin is a chemoattractant protein with adipokine and antimicrobial properties encoded by the retinoic acid receptor responder 2 (RARRES2) gene. Chemerin bioactivity is largely dependent on carboxyl-terminal proteolytic processing that generates chemerin isoforms with the different chemotactive, regulatory and antimicrobial potential.While these mechanisms are relatively well known, a role of alternative splicing in generating isoform diversity remains obscure. Using rapid amplification of cDNA ends (RACE) PCR, we have identified novel transcript variant 4 of mouse RARRES2 encoding mChem153K. Moreover, RT-QPCR results and analysis of publicly available RNA-seq datasets showed that different alternatively spliced variants of mouse RARRES2 are present in mouse tissues, and their expression pattern was not affected by inflammatory nor infectious stimuli. Finally, we demonstratedthat chemerin isoform mChem157S exhibits higher bactericidal but not chemotactic activity compared to mChem156S. Together, our findings highlight the importance of an alternative splicing in generation of chemerin isoforms diversity and activity.


Introduction
Protein isoforms can play important roles in various biological processes, like growth, differentiation or signal transduction. They can originate from separate genes, or single gene can code for multiple proteins due to the process called alternative mRNA splicing. Alternative polyadenylation, RNA editing, as well as post-translational modi cation may also generate the number of functionally distinct proteins.
However, the alternative splicing of transcripts is one of the main sources of proteomic diversity in eukaryotes. Despite sharing a high degree of amino acid sequence homology, each isoform can have various, even opposite, biological roles. [1][2][3] Therefore, discovery of novel mRNA transcripts and protein isoforms may uncover new biological roles and functions of genes. 4 Chemerin is a multifunctional chemoattractant, adipokine and antimicrobial agent capable of regulating different biological processes, including immune cell migration, adipogenesis, osteoblastogenesis, angiogenesis, glucose homeostasis and microbial growth. 5,6 The gene encoding chemerin is known as retinoic acid receptor responder 2 (RARRES2), or as tazarotene-induced gene 2 (TIG2). Liver and adipose tissue are reported to be the major sites of chemerin production; nonetheless, RARRES2 mRNA is detectable in many other tissues, including the adrenal glands, ovaries, pancreas, lungs, kidney, and skin. 7,8 Chemerin-induced signalling is mediated predominantly through chemokine-like receptor 1 (CMKLR1), which is expressed by many cells including hepatocytes, adipocytes, keratinocytes, plasmacytoid dendritic cells (pDCs), or macrophages. 7,9−13 Chemerin is secreted as a pro-chemerin, functionally inert precursor proteinreferred to as hChem163S (human) or mChem162K (mouse), with number and capital letter referring to the terminal amino acid position and single amino acid code, respectively. 14 Pro-chemerin can be converted to chemotactically active isoforms through posttranslational carboxyl-terminal processingby proteases belonging to the coagulation, brinolytic, andin ammatory cascades. The most active form of human chemerin, hChem157S, can be produced by direct cleavage of six C-terminal amino acids by neutrophil elastase, or cathepsin G. 15 Various proteolytic activities can generate other isoforms, including 152G, 153Q, 154F, 155A, 156F, and 158K,with low or no activity. [15][16][17][18] Several murinechemerin isoforms have been characterizedin a mouse model of obesity, mChem156S and mChem155F showed the highest biological activity. 19 Mouse chemerin undergoes tissue-speci c proteolytic cleavage similar to human chemerin. 19 While mechanisms of proteolytic processing in generation of chemerin isoforms are relatively well described, a role of alternative splicing remains obscure. Mouse RARRES2 gene is comprised of 6 exons and 5 introns. So far, three known (NM_001347168.1, NM_027852.3, NM_001347167.1) and one predicted (XM_011241467.3) protein-coding transcripts, encoding 162 and 163 aa protein, have been described. 20 The mChem162K is the major chemerin form in plasma. 19 However, tissue expression pro le and physiological role of mChem163K remains to be determined.
Generation of multiple chemerin isoforms is a key to control local, and context-speci c bioactivity of this protein. Therefore, better understanding of mechanisms underlying diversity of chemerin isoforms is of particular importance. Here, we show that alternatively spliced variants of mouse RARRES2 are present across different tissues and organs. Moreover, in addition to the variants encoding mChem163K and mChem162K, we have identi ed new transcript variant 4 encoding mChem153K. We demonstrate that in ammatory, and infectious conditions do not affect expression pattern of RARRES2 splice variants. Furthermore, we determined chemotactic and antimicrobial activity of chemerin isoforms generated by an alternative splicing, and show that mChem157S exhibitshigher bactericidal but not chemotactic activity compared to mChem156S. As such, we provide novel insights into the mechanisms that may contribute to the chemerin isoforms diversity and activity.

Results
Characterization of alternatively spliced transcript variants of RARRES2.
To identify transcript variants of mouse chemerin present in adult mouse tissues including liver and white adipose tissue (WAT), 3' and 5' RACE PCR were performed. We detected three transcript variants that have been already described. Mouse RARRES2 variant 1 represents the longest transcript and encodes the longer isoform 1 (mChem163K) (Fig.1 A-B). RARRES2 variant 2 uses an alternate in-frame splice site in the 3' coding region, compared to variant 1. This results in a shorter protein (isoform 2, mChem162K), compared to isoform 1. RARRES2 variant 3 differs in the 5' UTR and uses an alternate in-frame splice site in the 3' coding region, compared to variant 1. Therefore, variants 2 and 3 encode mChem162K. In addition to the previously reported variants 1, 2 and 3, we have identi ed the novel variant 4, generated by an alternate in-frame splice site in the 3' coding region (Fig. 1A-B). This variant 4 is composed of exons 1 to 6 (Fig. 1A). However, exon 5 lacks 27 bp fragment. The transcript variant 4 of mouse RARRES2 was not predicted nor annotated by Ensembl 21 and RefSeq 22 .
The chemerin protein isoforms, encoded by transcript variants 1 to 4, were aligned (Fig.1C). RARRES2 transcript variant 2 and 3 encode mChem162K which is the major chemerin form in plasma. 19 Splice variant 1 codes for chemerin mChem163K that has one extra amino acid, glutamine, at position 128. Interestingly, newly discovered mChem153K, encoded by the transcript variant 4, is devoid of 10 amino acids 128-137 compared to mChem163K. All the changes in amino-acid sequence of mouse prochemerin are linked to exon 5.
Expression pattern of RARRES2 splice variants across different tissues and experimental conditions. Given the multiple existing alternatively spliced transcript variants, we next assessed in which tissues they are present and whether they were differentially expressed. To do so, we analyzed publicly available results of RNA-seq experiments or performed standard RT-QPCR.
Using VastDB 23 , an atlas of alternative splicing pro les and functional associations in vertebrate cell and tissue types, we quanti ed transcript variants encoding for mChem162K and mChem163K but not mChem153K since transcript variant 4 is not included in the VastDB. The analysis revealed dominance of transcript variant 2 and 3 (mChem162K) in all investigated tissues with an average percent plicedin score (PSI) at around 68,5 ( Fig. 2A). However, transcript encoding mChem163K accounted for up to 42% in cerebellum or pancreas. These results were corroborated by the RNA-Seq ndings (Fig. 2B). In contrast, RARRES2 transcript 4, encoding mChem153K, is rare and showed the PSI score up to 1.5 ( Fig. 2A). The expression pattern of RARRES2 splice variants was not changed by high-fat diet, viral, bacterial nor parasite infections. There were no statistically important differences between control and treatment groups. However,the levels of newly discovered RARRES2 variant 4 tend to increase in kidney and skin after high-fat diet and S. aureus infection, respectively.
In line with the analysis of publicly available RNA-seq databases, RT-QPCR experiments con rmed that RARRES2 variant 4 is rare with the highest levels found in heart (Fig. 3A).The relative incidence values (RIV) 24 of the variant 4 may vary from ~0,31% in liver up to ~3,4% in kidneys and seminal vesicles (Fig.  3B).
We have shown previously that acute-phase cytokines, interleukin 1b (IL-1β) and oncostatin M (OSM), regulate chemerin expression in mouse adipocytes and human 3D skin cultures. 7,25 Therefore, we next analyzed if these cytokines can affect the balance between newly discovered RARRES2 transcript variant 4 and others splice variants in mouse tissues. The transcript ratio remained stable in all investigated mouse tissues, and there were no statistically signi cant differences between PBS and cytokine treated animals (Fig. 3C).
Expression of RARRES2 protein isoforms in E. coli and HEK293 cells.
To determine the physiological role of mouse chemerin isoforms we set up production of full-length protein variants (mChem163K, mChem162K, mChem153K), and variants lacking 6 aa at C-terminus (mChem157S, mChem156S, mChem147S). To do so, we employed E.coli and previously established protocols used to produce human recombinant chemerin. 26 We were able to produce and purify mChem163K and mChem162K and their derivatives, but we could not purify mChem153K. This include puri cation on NI-sepharose, ion exchange chromatography puri cation (MonoSP and MonoQ sepharose), Ni-sepharose based puri cation in denaturing or semi-denaturing conditions, construction of expression vector coding for mChem153K containing two His-tag sequences at the N and C terminus, various protein expression strains of E.coli (BL21, ArcticExpress, Rosetta). However, western blotting analysis of a supernatant fraction from E. coli culture using an anti-His tag antibody revealed the presence of mChem153K band (Fig. 4A). To prove that RARRES2 splice variant 4 can be translated into a protein in eucaryotic cells we transfected HEK293 cells with different chemerin constructs. Indeed, we detected mChem153K in cell lysates ( Fig. 4B) but not supernatants (data not shown). Interestingly, the chemerin band was only present for protein with signal peptide and His-tag located at the C-terminus.
Schematic representation of chemerin isoforms used in this study is shown on Fig. 4C.
Chemerin isoform mChem157S shows higher bactericidal but not chemotactic activity compared with mChem156S.
We hypothesized that changes in amino acid sequence of chemerin may affect its biological functions including chemotactic and bactericidal activity. Therefore, we performed transwell cell migration assay and antimicrobial microdilution assay (MDA) using chemerin responsive CMKLR1 + cells and E. coli HB101, respectively, and bioactive chemerin isoforms lacking six terminal aa, rmChemHis-157S and rmChemHis-156S. Bioactive chemerin isoform mChemHis-157S showed higher potency against E. coli HB101 by MDA assay compared with mChem156S ( Fig. 5A). PBS and human (hu) or mouse (m) chemerin derived peptide p4 were used as a negative and positive controls, respectively;In contrast, there were no statistically signi cant differences in chemerin isoform-mediated chemotactic activity by transwell migration assay (Fig. 5B). Experiments were performed using at least two different chemerin production batches.

Discussion
Our knowledge of the post-translational modi cations of chemerin that generate a variety of protein isoforms has increased signi cantly in the last two decades. However, these studies focused mainly on the proteolytic processing of human (hChem163S) or mouse pro-chemerin (mChem162K) by extracellular proteases. 15,16,19,27,28 ; Alternative splicing is a key factor increasing cellular and functional complexity.
Nevertheless, it remains obscure how an alternative splicing of RARRES2 contribute to the isoforms diversity despite identi cation of at least three transcripts variants of murine chemerin encoding two different protein isoforms. 22 In this study, we have described, for the rst time, transcript variant 4 of mouse RARRES2 coding for 153 aa chemerin isoform 3 (mChem153K). mChem153K lacks 10 aa at the position 128-137 compared to isoform 1 (mChem163K). This modi cation may signi cantly affect protein structure since it removes cysteine residue involved in the formation of one out of three intrachain disul de bonds. 14 Therefore, conformational changes could be the reason why all the Chem153K puri cation procedures have failed.
However, we have shown that mChem153K can be translated into protein since it was detectable in a supernatant fraction from E. coli culture and HEK293 lysates; Our in silico and in vivo studies have revealed that RARRES2 transcript variant 4 accounts for a small fraction of other splice variants under physiological conditions. The average percentage for all investigated mouse tissues was 0,55 % and 1,31 % for RNA-seq and real time RT-PCR studies, respectively.
One common outcome of alternative splicing is downregulationof the function of a gene by the production ofnon-functional isoforms of the active gene product. Thiscan be achieved by the alteration of functional domains of the protein. 29 Nonetheless, all the detected RARRES2 transcript variants are generated by an alternate in-frame splice site in the 3' coding region of exon 5 or differs in the 5' UTR (variant 3). These modi cations do not affect the C-terminal region of chemerin which is crucial for its bioactivity. 5 Alternative transcripts are very often differentially expressed between cells or tissues and display different functions. [30][31][32] Moreover, changes in alternative splicing events can be associated with exposure to different stimuli. 33 Altered chemerin expression may be of relevance in the context of pathological conditions like obesity, cancer, and in ammation. 10,27,34−36 Chemerin expression is regulated by a variety of in ammatory and metabolic mediators in a manner dependent on cell type. 25,37 We previously showed that IL-1β and OSM upregulated chemerin expression in human skin cultures 7 , and mouse adipocytes 25 . Moreover, bacteria, like S. aureus, upregulate chemerin levels in models of human epidermis and mouse skin. 7 Skin transcriptome analyses of antimicrobial peptides differentially regulated following skin infection with C. acnes or Leishmania braziliensis, revealed elevated levels of RARRES2 transcript. 38 Here, our study showed that splicing pattern of RARRES2 mRNA was not altered by high-fat diet, bacterial, viral or parasite infection nor cytokine treatment in different mouse organs. Therefore, these factors are not the major determinants for splice site selection.
We also asked if there were any differences in antimicrobial or chemotactic activity between biologically active chemerin isoformsm Chem157S and mChem156S, since they differ only by a single amino acid, glutamine, at position 128. This change does not affect directly antimicrobial region (p4) of chemerin, localized in the middle of the protein sequence (position 66-85 or 68-87 for human and mouse chemerin, respectively). 6 Chemerin isoform mChem157S exhibited increased antibacterial activity compared to mChem156S. In contrast, we did not nd any changes in chemotactic activity. It is not clear how single amino acid can affect chemerin structure and functions. As is the case for chemoattractant activity, the inhibitory C-terminal peptide present in the pro-chemerin must be removed for full antibacterial effects. It was hypothesized that removal of inhibitory peptide enables structural accessibility of chemerin antimicrobial domain, and/or a release of internal antimicrobial peptide. 26 While four transcript variants of mouse RARRES2 encoding three protein isoforms are known, there is only one con rmed splice variant of human RARRES2 (NM_002889.4). This variant is translated into hChem163S precursor protein. 20 Interestingly, one predicted transcript (XM_017012491.1), encoding 132 aa peptide (isoform X1), is annotatedin the NCBI database. Amino acid sequence of the C-terminal region of chemerin differs signi cantly between hChem163S and isoform X1. However, to date, such transcript variant has not been detected yet in human tissues.
In summary, our studies reveal novel insights into the mechanisms accounting for chemerin isoforms diversity. For the rst time, we report the identi cation of rare transcript variant 4 of mouse RARRES2, encoding mChem153K (isoform 3). RARRES2 transcript variants from 1 to 4 were present in all investigated mouse tissues, and variants encoding chemerin isoform mChem162K are the most abundant. Our research showed that splicing pattern of RARRES2 mRNA was not altered by high-fat diet, bacterial, viral or parasite infection nor pro-in ammatory cytokine treatment. Chemerin isoform mChem157S showed higher antimicrobial but not chemotactic activity compared to mChem156S. These ndings provide a basis for further investigations of the role of alternative splicing on chemerin functions.

Materials
If not stated differently, all chemicals were purchased from Sigma-Aldrich (St. Louis, MO, USA).

Animal studies
Male eight-to 12-week-old C57BL/6 mice were used for these investigations. The mice were maintained under speci c pathogen-free conditions at the Faculty of Biochemistry, Biophysics, and Biotechnology of Jagiellonian University animal care facility. IL-1β and OSM were injected intraperitoneally at doses of 10 μg/kg BW and 160 μg/kg BW, respectively as previously described. 25 After 48 h, different tissues were isolated and subjected to RT-QPCR analysis. All experimental procedures were approved by the First Local Ethical Committee on Animal Testing at the Jagiellonian University in Krakow, Poland (permit no. 41/2014), in accordance with the ARRIVE guidelines and the Guidelines for Animal Care and Treatment of the European Community. The mice were sacri ced by an overdose of anesthesia (a mixture of ketamine and xylasine), followed by cervical dislocation.

Rapid ampli cation of cDNA ends (RACE)
Total RNA was extracted as described by Chomczynski  . Expression stabilities of commonly used reference genes were analyzed as described previously. 25 Relative gene expression normalized to the geometric mean of these housekeeping genes was calculated using the 2 −ΔΔCT method. 40 Relative Incidence of Variant (RIV) was obtained with the method described by Londoño et al. 24 , PCR e ciencies of primer sets were calculated with CFX Maestro Software (Bio-Rad) using pcDNA3.1 plasmids encoding mChem162K and mChem153K as a template Alternative splicing analyses from RNA-Seq datasets Information about RARRES2 expression level in distinct tissues and cell lines as well as isoform quanti cation were acquired from VastDB. 23 To assess isoform ratios in publicly available RNA-Seq datasets we calculated Percent Spliced In (PSI) scores with vast-tools. 23 Notably, we analyzed NCBI:GEO datasets which investigate the molecular effects of high-fat diet: GSE76133, GSE75984 and GSE117249 as well as records related to transcriptional changes upon distinct infections: Staphylococcus aureus (GSE108718), Toxoplasma gondii (GSE119855) and in uenza virus (GSE114232).Differential splicing analyses were carried out with vast-tool's module diff.
Then, pcDNA3vectors were used as a template to generate pNIC28-Bsa4 plasmids encoding different isoforms of chemerin (mChem163K, mChem162K, mChem153K, mChem157S, mChem156S,  mChem147S). The following starters: 5'-GCACCATCATCATCATCATTCTTCTGGTGAGCCCGAACTCAGCGAGACC, 5'-CACAATTCAGAAAATATCATAATATCTCATTTCACTATTTGGTTCTCAGGGCCCTGGAGAAG were designed to be complementary to a speci c site of the target expression vector pNIC28-Bsa4. The PCR products were cloned into pNIC28-Bsa4 expression vector at a site preceded by the sequence coding for hexahistidine tag, using the overlap-extension PCR method. All constructs lacked the native chemerin signal peptide. The identity of the created pNIC28-Bsa4 constructs was veri ed by sequencing (Genomed, Warsaw, Poland).

Expression of mouse chemerins in HEK293 cells and western blot
HEK293 cells were grown in DMEM medium supplemented with 10% FBS and gentamycin (50 µg/mL).
Cells were seeded at a density of 1.5 × 10 4 cells in a 24-well culture plate and, 24 h later.Then, the cells were transfected with the pcDNA plasmids described above using ViaFect transfection reagent (Promega Corporation, Madison, WI, USA) in accordance with the producers instructions. Transiently transfected cells were RIPA-lysed after 48h and used for Western Blot analysis. Proteins were electrophoretically separated on a SDS-polyacrylamide gel and wet-transferred onto PVDF membrane (Bio-Rad), followed by blocking with 5% skim milk (Merck). His-tagged chemerin isoforms were detected using anti-Histag antibodies(ab15149, Abcam, Cambridge, MA). Chemiluminescent detection was carried out using WesternBright ECL (Advansta) and ChemiDoc MP imaging system (Bio-Rad).
Production and puri cation of mouse chemerin isoforms in E. coli Chemerin isoforms were expressed using E. coli strain NiCo21(DE3) (New England Biolabs, MA, USA) transformed with plasmids described above. Bacteria were precultured in 37°C in LB medium until culture density reached OD600 value between 0,6-0,8. Protein expression was induced by addition of IPTG to a nal concentration of 1mM and carried out overnight in 18°C. After centrifugation the bacterial pellet was dissolved in PBS with 1mM EDTA and cOmplete protease inhibitor cocktail (Roche), and sonicated. After sonication samples were centrifuged (40000g, 20min, 4°C) and pellets were resuspended in denaturing buffer (6M GuHCl, 50mM NaCl, 50mM TRIS, pH8). After centrifugation (12000g, 12min, 4°C) supernatants were 100-fold diluted in renaturation buffer (0,5M GuHCl, 0,4M Sucrose, 0,1M TRIS, 1mM GSH, 0,1mM GSSG, pH8). Any precipitate was removed by centrifugation, and the protein solutions were concentrated using Amicon Ultra Centrifugal Filters (Merck). Concentrated solutions were 10-fold diluted in dilution buffer (0,1M TRIS, 0,1M Sucrose, 1mM GSH, 0,1mM GSSG, pH 8). Any precipitate was removed from solution by centrifugation. Proteins were puri ed from solution by incubation with Ni-Sepharose 6 Fast Flow (GE Healthcare, Uppsala, Sweden), washed on column in wash buffer (0,1M TRIS, 1mM GSH, 0,1mM GSSG, pH 8), then eluted with 500 mM imidazole in wash buffer. mChem163K and mChem162K protein samples were dialyzed against buffer A (25mM TRIS, 25mM NaCl, pH 7,6). Any precipitates were removed by centrifugation. Protein samples were loaded on Q-Sepharose Fast Flow columns (GE Healthcare, Uppsala, Sweden). Elution fractions containing increasing concentrations of NaCl were collected and analysed by SDS-PAGE electrophoresis and coomassie blue staining. Fractions of low NaCl concentration were pooled, concentrated on Amicon Ultra Centrifugal Filters (Merck), and then dialyzed against PBS. Protein samples were routinely >90% pure as assessed by SDS-PAGE and Coomassie Blue staining. The concentration of chemerin was determined by measuring the absorbance at 280 nm using NanoDrop ND-1000 spectrophotometer (ThermoFisherScienti c, USA), and bicinchoninic acid (BCA) assay (ThermoFisherScienti c, USA). Chemerin activity pro ling between batches was evaluated by in vitro transwell assay using CMKLR1 expressing L1.2 cells as described below. At least two different batches of chemerin isoforms were used for experiments.

Antimicrobial microdilution assay (MDA)
For antimicrobial experiments E.coli HB101 were grown in brain heart infusion (BHI) broth at 37°C. To determine the antimicrobial activity of the chemerin isoforms, bacteria in mid-logarithmic phase were harvested, washed three times with PBS and diluted to 4 x 10 5 CFU/ml with PBS. Then bacteria were incubated with either chemerin isoforms (3 μM) or PBS (control) for 2 h. The number of viable bacteria were enumerated by CFU counting.

Chemotaxis assay
Puri ed mouse chemerin isoforms were tested for the ability to stimulate migration of the murine pre-B lymphoma cell line L1.2 stably transfected with mouse CMKLR1 (L1.2-CMKRL1). L1.2-CMKRL1+ cell were provided by Dr. Brian Zabel and Dr. Eugene C. Butcher (Stanford University School of Medicine and Veterans Affairs Palo Alto Health Care System). A total of 100 μl cells (2 × 10 5 cells/well) was added tothe top well of 5-μm pore Corning Costar Transwell inserts (Corning, USA). Chemotaxis assay was performed in chemotaxis media (RPMI 1640 with 10% FBS) containing 1 nM of chemerin, added to the bottom well in a 600-μl volume. Migration was assayed for 2 h at 37°C. The inserts were then removed, and cells that had migrated through the lter to the lower chamber were collected and counted by ow cytometry (LSRII; BD Biosciences, USA). The results are presented as percentage of input migration.

Statistical analysis
Differential splicing quanti cation from RNA-Seq experiments was performed using vast-tools (-r 0.95, and -m 0.1). Other data were analyzed using STATISTICA 13 (StatSoft, Tulsa, OK, USA). The results were visualized with Prism (GraphPad Software, San Diego, CA, USA), and presented as mean ± standard deviation (SD). The Student's t-test was used for comparison between two groups. For multiple comparisons, analysis of variance (ANOVA) with Tukey's post-hoc test was used. Differences were considered statistically signi cant for p-values of less than 0.05.

Figure 2
Analysis of RNA-seq experiments and VastDB database reveals tissue-wide expression of RARRES2 splice variants. Alternative spicing events ofRARRES2 in distinct mouse tissues was acquired from VastDB database (A). The effect of high-fat diet, S. aureus, T. gondii, in uenza virus or lymphocytic choriomeningitis virus infection onRARRES2 splicing pattern was assessed using publicly available RNA-Seq datasets (B). The results are expressed as percent spliced-in (PSI) values. Differential splicing analyses using vast-tools did not reveal any statistically signi cant changes in splicing pattern between control and treatment groups.  HEK293 cells were transfected with the pcDNA plasmids coding for mChem153K with different localizations of signal peptide (Sig) and His-tag (His). Cell lysates were subjected to western blot analysis as described above (B). Commercially available human (rhChem157S) and mouse (rmChem156S) recombinant chemerin lacking His-tag were used as negative control. His-tagged recombinant mouse chemerin (rmChem162K, 20 ng) was used as positive control. Schematic representation of chemerin isoforms used in the study is shown(C).

Figure 5
Chemerin isoform mChem157Sexhibits higher bactericidal but not chemotactic activity compared to mChem156S. Bacteria were incubated with chemerin isoforms (3 µM) lacking six terminal aa, PBS (negative control), human or mouse peptide p4 (positive control) for 2 h. Cell viability, shown as the percentage of a control cells, was analyzed by MDA assay (A). Chemotactic bioactivity of chemerin isoforms (1nM) was evaluated by in vitro transwell assay using CMKLR1 expressing L1.2 cells. Migration to bioactive recombinant rhChem157S and rmChem156S at 1nM, and chemotaxis medium is shown as a positive and negative control, respectively (B). Results are expressed as the mean ± SD of at least three independent experiments using two different chemerin production batches. *** p<0.001, ** p<0.01, * p<0.05 by one-way ANOVA with post-hoc: Tukey's multiple comparisons test.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.