- Research article
- Open Access
Relationship of porcine IGF2 imprinting status to DNA methylation at the H19 DMD and the IGF2 DMRs 1 and 2
© Braunschweig et al; licensee BioMed Central Ltd. 2011
- Received: 10 February 2011
- Accepted: 17 May 2011
- Published: 17 May 2011
Porcine IGF2 and the H19 genes are imprinted. The IGF2 is paternally expressed, while the H19 gene is maternally expressed. Extensive studies in mice established a boundary model indicating that the H19 differentially methylated domain (DMD) controls, upon binding with the CTCF protein, reciprocal imprinting of the IGF2 and the H19 genes. IGF2 transcription is tissue and development specific involving the use of 4 promoters. In the liver of adult Large White boars IGF2 is expressed from both parental alleles, whereas in skeletal muscle and kidney tissues we observed variable relaxation of IGF2 imprinting. We hypothesized that IGF2 expression from both paternal alleles and relaxation of IGF2 imprinting is reflected in differences in DNA methylation patterns at the H19 DMD and IGF2 differentially methylated regions 1 and 2 (DMR1 and DMR2).
Bisulfite sequencing analysis did not show any differences in DNA methylation at the three porcine CTCF binding sites in the H19 DMD between liver, muscle and kidney tissues of adult pigs. A DNA methylation analysis using methyl-sensitive restriction endonuclease Sac II and 'hot-stop' PCR gave consistent results with those from the bisulfite sequencing analysis. We found that porcine H19 DMD is distinctly differentially methylated, at least for the region formally confirmed by two SNPs, in liver, skeletal muscle and kidney of foetal, newborn and adult pigs, independent of the combined imprinting status of all IGF2 expressed transcripts. DNA methylation at CpG sites in DMR1 of foetal liver was significantly lower than in the adult liver due to the presence of hypomethylated molecules. An allele specific analysis was performed for IGF2 DMR2 using a SNP in the IGF2 3'-UTR. The maternal IGF2 DMR2 of foetal and newborn liver revealed a higher DNA methylation content compared to the respective paternal allele.
Our results indicate that the IGF2 imprinting status is transcript-specific. Biallelic IGF2 expression in adult porcine liver and relaxation of IGF2 imprinting in porcine muscle were a common feature. These results were consistent with the IGF2 promoter P1 usage in adult liver and IGF2 promoter P2, P3 and P4 usages in muscle. The results showed further that bialellic IGF2 expression in liver and relaxation of imprinting in muscle and kidney were not associated with DNA methylation variation at and around at least one CTCF binding site in H19 DMD. The imprinting status in adult liver, muscle and kidney tissues were also not reflected in the methylation patterns of IGF2 DMRs 1 and 2.
- Paternal Allele
- CTCF Binding Site
- Bisulfite Sequencing Analysis
- Adult Boar
- Maternal Hypomethylation
Porcine insulin-like growth factor 2 (IGF2) and H19 genes are reciprocally imprinted in most tissues. In mice, these two genes share common endodermal and mesodermal enhancers and the mouse Igf2 gene is also paternally expressed in most tissues whereas the H19 gene is maternally transcribed [1–3]. Mice lacking the Igf2 gene weighed about 40% less than their litter mates . The H19 gene expresses a non-protein-coding RNA and is located 88.1 kb downstream of IGF2 [5, 6]. Recently, it was found that H19 transcripts can function as microRNA precursors .
The pig INS-IGF2-H19 imprinting cluster is highly homologous to the corresponding human gene cluster and is thus a good model to study epigenetic mechanisms . Recently, a quantitative trait nucleotide (QTN) at position IGF2-intron3-3072 was identified and various antisense transcripts originate from the paternal allele demonstrating the complex transcription from this gene [8, 9].
An extensive number of studies have been conducted to elucidate the epigenetic mechanisms of IGF2 and H19 which are thought to be co-ordinately regulated, both in terms of their expression patterns and their reciprocal imprinting (for review see ). It was shown by deletions in mice that a region of paternal-specific DNA methylation (differentially methylated domain, DMD) upstream of H19 is an epigenetic mark required for imprinting of IGF2 and H19 [10, 11]. Bell and Felsenfeld  reported that activity of H19 DMD depends upon the vertebrate eleven-zinc finger protein CTCF that binds to this DMD and mediates the function of the boundary/insulator element. They also found that methylated CpG sites at the CTCF binding site abolished binding in vitro. Based on this finding Bell and Felsenfeld  developed a model explaining the reciprocal imprinting. On the maternal allele the enhancer downstream of H19 has no access to the IGF2 promoters due to the boundary function of CTCF proteins bound to the unmethylated DMD whereas the H19 gene can still be transcribed. On the paternal allele DNA methylation at the H19 DMD eradicates the boundary function which leads to IGF2 gene transcription and silencing of the H19 gene. These findings were made simultaneously using transgenic mice and cell culture and contributed to establish the boundary model . More recently it was demonstrated that differentially methylated regions in the mouse Igf2 and H19 genes interact in an epigenetically regulated manner that partition maternal and paternal alleles into distinct loops. The maternal allele H19 DMD interacts with Igf2 DMR1 allowing maternal H19 to be expressed while the paternal H19 DMD interacts with Igf2 DMR2, allowing Igf2 to be expressed and leaves the H19 gene silent. This model was named the chromatin loop model .
In an ongoing study we investigated the combined imprinting status of all IGF2 expressed transcripts in liver, skeletal muscle and kidney tissues of adult boars. We hypothesized that an alteration in IGF2 imprinting status might be reflected in DNA methylation variations at differentially methylated regions as suggested by the boundary and chromatin loop models. To test this hypothesis we studied the association between the IGF2 imprinting status in three different tissue samples of six boars and their DNA methylation of H19 DMD, IGF2 DMR1 and DMR2 (Figure 1). Furthermore, we included samples from two foetus and two newborn pigs in order to examine IGF2 imprinting and DNA methylation at these differentially methylated regions during development. We were curious to see whether the imprinting status of IGF2 in liver, skeletal muscle and kidney at different developmental stages is also reflected in the DNA methylation patterns of H19 DMD and in particular at CTCF binding sites as well as in IGF2 DMR1 and IGF2 DMR2.
Imprinting status of IGF2
Imprinting status in different pig tissues and for three development stages expressed as ratio of paternal to maternal IGF2 gene expression.
Ratio paternal to maternal gene expression (± SD)
Adult liver (N = 6)
4.2 (± 6.2)
Newborn liver (N = 2)
21.3 (± 0.01)
Foetal liver (N = 2)
Adult muscle (N = 6)
9.2 (± 6.2)
Newborn muscle (N = 2)
19.9 (± 6.0)
Foetal muscle (N = 2)
Paternal expression (N = 1) and 14 (N = 1))
Adult kidney (N = 6)
22.9 (± 10.2)
Newborn kidney (N = 1)
Foetal kidney (N = 2)
The observed biallelic IGF2 expression in these adult boars' liver is in agreement with previous reports showing biallelic expression of IGF2 in the livers of humans and pigs [15–19]. RT-PCR analysis of adult liver cDNA indicated that all 4 IGF2 promoters were used whereas cDNA from muscle and kidney revealed very low level of products after 40 PCR cycles for transcripts from promoter P1 and high level of products for transcripts from promoter P2, P3, and P4 (data not shown). This result is in agreement with previous findings from Amarger et al. and Li et al. [5, 16], however, from previous Northern blot analysis it is known that promoter P1 is predominantly used in adult pig liver and transcripts from promoter P2, P3 and P4 were not detected. The RT-PCR analysis of microsatellite SWC9 is a semi-quantitative approach to investigate IGF2 imprinting status and the results conclusively indicate that IGF2 imprinting is reversed similarly to biallelic expression in adult liver and it is relaxed to different degrees in muscle and kidney tissues during development.
Bisulfite sequencing analysis of H19 DMD
DNA methylation analysis using methyl-sensitive restriction endonuclease
Bisulfite sequencing analysis of IGF2 DMR1
DNA polymorphisms that allowed the deduction of the parental origin of alleles in the IGF2 DMR1 were not found. Nevertheless we used bisulfite sequencing to analyse a considerable number of clones to search for specific DNA methylation patterns both between the three tissue samples and during development, which are supposed to be associated with the IGF2 imprinting status. We compared DNA from between 7 and 37 single clones per tissue and developmental stage and could not find any significant difference between DMR1 methylation in muscle and kidney tissues within foetal, newborn and adult individuals (Wilcoxon two-sample test, two-sided).
Bisulfite sequencing analysis of IGF2 DMR2
Van Laere et al.  and Wrzeska et al.  showed relaxation of imprinting in skeletal muscle tissues of 4 month old pigs and exclusive paternal IGF2 expression in the tissues of adult pigs' skeletal muscle, respectively. It is important to point out that insufficient PCR cycles, or template, may lead to the product arising from the maternal allele to go undetected, especially in the muscle and kidney, where relaxation of imprinting seems to increase with aging (Figure 2). This might be a reason for the conflicting data found by Van Laere et al.  and Wrzeska et al. . In a comprehensive study the imprinting status of IGF2 and H19 were determined in 13 tissue samples of week-old piglets . Li et al. (2008)  found biallelic IGF2 expression from promoter P1 in heart, liver, brain, lung, kidney, stomach, pancreas, thymus, tongue, muscle, bladder, spleen, and placenta tissues of week-old pigs. Their RT-PCR analysis of microsatellite SWC9 in these 13 different tissues revealed, however, exclusive or nearly exclusive paternal IGF2 expression. Furthermore, their real-time PCR analysis of IGF2 exon 2 originating from promoter P1 and of IGF2 exon 9, that is common for all IGF2 transcripts, resulted in roughly 33% and 10% promoter P1 IGF2 transcription relative to total IGF2 transcription in brain and placenta, respectively. These results are in agreement with the data presented for IGF2 imprinting in liver and suggest that transcription from the IGF2 promoter P1 may be regulated by other mechanisms than that from promoters P2, P3 and P4. Together these data emphasize the promoter-specific IGF2 imprinting status in different tissues and during development . Biallelic expression was also observed in the liver and brain of 6-month-old lambs but not in their kidneys .
H19 DMD was paternally hypermethylated and maternally hypomethylated in liver, muscle and kidney of all three developmental stages independent of the combined imprinting status of all IGF2 expressed transcripts. This finding challenges the boundary model [12, 13] postulating that the vertebrate eleven-zinc finger protein CTCF binds the maternal unmethylated H19 DMD insulating the upstream IGF2 promoters from enhancers downstream of H19. On the paternal allele the methylated DMD abolishes CTCF binding and enhancers 3' of H19 have access to IGF2 promoters. Our results demonstrate that IGF2 is expressed from both alleles, mainly in adult liver and, to a much lesser extent in skeletal muscle and kidney, although H19 DMD is indeed differentially methylated. Histone modifications might still cause these effects but if so, they would be independent of DNA methylation. Investigations of mouse Igf2 DMR 1 and 2 led to a model of parent-specific chromatin loops that regulate Igf2 imprinting . DNA methylation at DMR 1 and 2 in our samples does not support a parent-specific chromatin loop model regulating IGF2 imprinting in pig. It remains inconclusive if subtle differences in DNA methylation in DMR1 between foetal and adult liver and that between the parental alleles in DMR2 of foetal and newborn liver are involved in the control of the IGF2 imprinting status.
From our imprinting and DNA methylation analyses we conclude, firstly, that IGF2 expression from both parental alleles in adult porcine liver and relaxation of IGF2 imprinting in adult porcine muscle are not associated with DNA methylation variation at and around at least one CTCF binding site in H19 DMD. Secondly, the lower DNA methylation content in DMR1 in foetal liver, as compared to adult liver, should be evaluated on molecules from which the parental origin could be established. Thirdly, similar to DMR1 porcine DMR2 is hypermethylated on both parental alleles rather than differentially methylated, as observed for H19 DMD. Furthermore, the maternal DMR2 allele was more methylated in foetal and newborn animals when compared to the respective paternal allele. Finally, the transition of IGF2 imprinting in foetal liver to IGF2 expression from both alleles in adult liver may inherent new mechanisms involved in IGF2 imprinting regulation and provides a promising subject for further study.
From a collection of Swiss Large White pigs 2 male foetus, 1 female and 1 male newborn and 6 adult boars were selected based on their heterozygosity for the microsatellite marker SWC9 located in the 3'-UTR of the IGF2 gene.
DNA and RNA isolation and first strand cDNA synthesis
DNA was isolated from liver, skeletal muscle and kidney tissues using the DNeasy Blood & Tissue Kit from Qiagen (Hombrechtikon, Switzerland). RNA extraction was performed with Trizol® Reagent according to the manufacturer's protocol (Invitrogen, Lucerne, Switzerland). Total RNA was digested with RNase-Free DNAse I according to the supplier's instructions (Ambion, Rotkreuz, Switzerland). DNA free RNA was reverse transcribed using a First-Strand cDNA Synthesis Kit (GE Healthcare, Glattbrug, Zurich) and products were subsequently purified with QIAquick columns (Qiagen).
Imprinting status was investigated by means of the SWC9 microsatellite marker located in the 3'-UTR of the IGF2 gene (Figure 1). The SWC9 microsatellite marker was amplified using Qiagen's Multiplex PCR Master Mix including a FAM labeled forward primer. PCR was performed with an initial step at 94°C for 15 minutes and 34 cycles of a denaturation step at 94°C for 30 seconds, an annealing step at 60°C for 1 and a half minutes and an elongation step at 72°C for 1 minute, and a final elongation for 10 minutes. All samples including the standard samples were analysed in triplicates and subjected to the same PCR run in a 96 well plate. Sequences of the PCR primers are shown in Supplementary Table 1. Following PCR 1 μl of each reaction was combined with 10 μl of genotyping mix (980 μl of HiDi formamide and 20 μl of GeneScanTM-500 LIZTM Size Standard (Applied Biosystems, Rotkreuz, Switzerland). The mixture was denatured for 2 min., chilled on ice and loaded on an ABI 3730 capillary sequencer (Applied Biosystems). Data was analyzed using GeneMapper software version 4.0 (Applied Biosystems). A standard dilution series was established based on the peak areas of mixed DNA samples from two homozygous individuals for the respective SWC9 236 and SWC9 247 alleles. We used ratios of 32:1, 16:1, 8:1, 4:1, 2:1, 1:1, 1:2, 1:4, 1:8, 1:16, and 1:32 of the respective SWC9 236 and SWC9 247 homozygous DNA samples. The ratio of the peak areas was used to calculate the ratio between the SWC9 alleles 236 and 247 according to the formula y = 0.9519x + 0.014 Figure 2 and.
DNA was converted with the EpiTect Bisulfite kit according to the supplier's manual (Qiagen). Bisulfite-conversion-based methylation PCR primers were designed with the program Methprimer http://www.urogene.org/methprimer/index.html and in the case of IGF2 _DMR2 with Methyl Primer Express (ABI). Primer sequences and product sizes of the 4 fragments covering the H19 DMD (H19 _DMD_1, H19 _DMD_2, H19 _DMD_3, H19 _DMD_4), the IGF2 DMR1 (IGF2 _DMR1) and the IGF2 DMR2 (IGF2 _DMR2) is shown in Additional file 1 Table S1. PCR was performed with the Multiplex PCR Master Mix and products from liver, skeletal muscle and kidney were cloned (TOPO TA Cloning Kit (Invitrogen). White colonies were picked diluted in 50 μl water and amplified with the illustra™ TempliPhi amplification kit (GE Healthcare) and sequenced on an ABI 3730 capillary sequencer (Applied Biosystems). Bisulfite sequencing analysis was performed with the programs BiQ Analyzer http://biq-analyzer.bioinf.mpi-sb.mpg.de/ and MethTools http://genome.fli-leibniz.de/methtools/.
DNA methylation analysis using methyl-sensitive restriction endonuclease Sac II and 'hot-stop' PCR
About 100 ng genomic DNA was digested over night with the methyl-sensitive Sac II endonuclease (New England BioLabs, Allschwil, Switzerland). CpG methylation of the recognition site CCGCGG inhibits digestion. A Sac II recognition site is present in a CTCF binding site 5' upstream of H19 and is referred to pig repeat P2 . We previously re-sequenced the pig imprinting control region (ICR) containing the three pig CTCF binding sites upstream of H19. By this means we identified two SNPs, one at position AY044827.1:g.32530C>T and the other at position AY044827.1:g.32619G>A. PCR primers ('hot-stop') were designed which encompass the CTCF binding site in P2 and the two SNPs (Supplemented Table 1). The forward primer is tailed with a M13 forward sequence. An aliquot of the Sac II digested DNA was PCR amplified for each sample using these primers by a 'hot-stop' PCR procedure for linear quantification of allele ratios . The 'hot-stop' PCR was performed with Multiplex PCR Master Mix (Qiagen) for 35 cycles with an initial denaturation step at 94°C for 15 min followed by denaturation at 94°C for 30 sec, an annealing temperature of 55°C for 30 sec, an annealing temperature of 68°C for 30 sec and an elongation step at 72°C for 30 sec. The PCR was then paused after these 34 cycles at 72°C and 0.2 μM IRDye™700 labeled M13 primers was added to the reaction. The PCR was then resumed for an additional cycle and a final elongation step at 72°C for 10 min.
The PCR product contains a common 5'...CGCG...3' recognition site for BstU I (New England BioLabs) and a second that includes the AY044827.1:g.32530C>T SNP which was used to discriminate the parental origin of the alleles in this fragment of H19 DMD. An aliquot of the 'hot-stop' PCR was subsequently digested over night with BstU I according to the supplier's recommendation. The digested products were separated on a 1.5% agarose gel and scanned on an Odyssey Infrared Imaging System according to LI-COR's instruction (LI-COR Biosciences, Bad Homburg, Germany). Bands were visualized using the LI-COR Odyssey software.
Parent of origin determination of alleles at the IGF2 DMR2
To determine the allele origin of IGF2 DMR2 we used a single nucleotide polymorphism (SNPs) in the 3'-UTR at position IGF2-exon9-612A>T. Based on this SNP we analyzed parental DNA methylation patterns at the IGF2 DMR2.
We thank Leeson J. Alexander for careful reading the manuscript and critical comments. This research project has been co-financed by the European Commission, within the 6th Framework Programme, contract no. FOOD-CT-2006-016 250. The text represents the authors' views and does not necessarily represent a position of the Commission, who will not be liable for the use made of such information.
- DeChiara TM, Robertson EJ, Efstratiadis A: Parental imprinting of the mouse insulin-like growth factor II gene. Cell. 1991, 64: 849-859. 10.1016/0092-8674(91)90513-X.View ArticlePubMedGoogle Scholar
- Bartolomei M, Zemel S, Tilghman SM: Parental imprinting of the mouse H19 gene. Nature. 1991, 351: 153-155. 10.1038/351153a0.View ArticlePubMedGoogle Scholar
- Arney KL: H19 and Igf2--enhancing the confusion?. Trends Genet. 2003, 19: 17-23. 10.1016/S0168-9525(02)00004-5.View ArticlePubMedGoogle Scholar
- DeChiara TM, Efstratiadis A, Robertson EJ: A growth-deficiency phenotype in heterozygous mice carrying an insulin-like growth factor II gene disrupted by targeting. Nature. 1990, 345: 78-80. 10.1038/345078a0.View ArticlePubMedGoogle Scholar
- Amarger V, Nguyen M, Van Laere A-S, Braunschweig MC, Nezer C, Georges M, Andersson L: Comparative sequence analysis of the INS-IGF2-H19 gene cluster in pigs. Mamm Genome. 2002, 13: 388-398. 10.1007/s00335-001-3059-x.View ArticlePubMedGoogle Scholar
- Bartolomei MS, Vigneau S, O'Neill MJ: H19 in the pouch. Nat Gene. 2008, 40: 932-933. 10.1038/ng0808-932.View ArticleGoogle Scholar
- Cai X, Cullen BR: The imprinted H19 noncoding RNA is a primary microRNA precursor. RNA. 2007, 13: 313-316. 10.1261/rna.351707.PubMed CentralView ArticlePubMedGoogle Scholar
- Van Laere A-S, Nguyen M, Braunschweig M, Nezer C, Collette C, Moreau L, Archibald AL, Haley CS, Buys N, Tally M, Andersson G, Georges M, Andersson L: A regulatory mutation in IGF2 causes a major QTL effect on skeletal muscle growth in the pig. Nature. 2003, 425: 832-836. 10.1038/nature02064.View ArticlePubMedGoogle Scholar
- Braunschweig MH, Van Laere A-S, Buys N, Andersson L, Andersson G: IGF2 antisense transcript expression in porcine postnatal skeletal muscle is affected by a quantitative trait nucleotide in intron 3. Genomics. 2004, 84: 1021-1029. 10.1016/j.ygeno.2004.09.006.View ArticlePubMedGoogle Scholar
- Tremblay KD, Saam JR, Ingram RS, Tilghman SM, Bartolomei MS: A paternal-specific methylation imprint marks the alleles of the mouse H19 gene. Nature Genet. 1995, 9: 407-413. 10.1038/ng0495-407.View ArticlePubMedGoogle Scholar
- Thorvaldsen JL, Duran KL, Bartolomei MS: Deletion of the H19 differentially methylated domain results in loss of imprinted expression of H19 and Igf2. Genes Dev. 1998, 12: 3693-3702. 10.1101/gad.12.23.3693.PubMed CentralView ArticlePubMedGoogle Scholar
- Bell AC, Felsenfeld G: Methylation of a CTCF-dependent boundary controls imprinted expression of the Igf2 gene. Nature. 2000, 405: 482-485. 10.1038/35013100.View ArticlePubMedGoogle Scholar
- Hark AT, Schoenherr CJ, Katz DJ, Ingram RS, Levorse JM, Tilghman SM: CTCF mediates methylation-sensitive enhancer-blocking activity at the H19/Igf2 locus. Nature. 2000, 405: 486-489. 10.1038/35013106.View ArticlePubMedGoogle Scholar
- Murrell A, Heeson S, Reik W: Interaction between differentially methylated regions partitions the imprinted genes Igf2 and H19 into parent-specific chromatin loops. Nat Genet. 2004, 36: 889-893. 10.1038/ng1402.View ArticlePubMedGoogle Scholar
- Wrzeska M, Żyga A, Rejduch B, Słota E: A note on biallelic expression of the IGF2 gene in the liver and brain of adult pigs. J Anim Feed Sci. 2006, 15: 57-60.Google Scholar
- Li C, Bin Y, Curchoe C, Yang L, Feng D, Jiang Q, O'Neill M, Tian XC, Zhang S: Genetic imprinting of H19 and IGF2 in domestic pigs (Sus scrofa). Anim Biotechnol. 2008, 19: 22-27. 10.1080/10495390701758563.View ArticlePubMedGoogle Scholar
- Ekström TJ, Cui H, Li X, Ohlsson R: Promoter-specific IGF2 imprinting status and its plasticity during human liver development. Development. 1995, 121: 309-316.PubMedGoogle Scholar
- Li X, Cui H, Sandstedt B, Nordlinder H, Larsson E, Ekström TJ: Expression levels of the insulin-like growth factor-II gene (IGF2) in the human liver: developmental relationships of the four promoters. J Endocrinol. 1996, 149: 117-24. 10.1677/joe.0.1490117.View ArticlePubMedGoogle Scholar
- Wu J, Qin Y, Li B, He WZ, Sun ZL: Hypomethylated and hypermethylated profiles of H19DMR are associated with the aberrant imprinting of IGF2 and H19 in human hepatocellular carcinoma. Genomics. 2008, 91: 443-450. 10.1016/j.ygeno.2008.01.007.View ArticlePubMedGoogle Scholar
- Lo HS, Wang Z, Hu Y, Yang HH, Gere S, Buetow KH, Lee MP: Allelic variation in gene expression is common in the human genome. Genome Res. 2003, 13: 1855-1862.PubMed CentralView ArticlePubMedGoogle Scholar
- Sandovici I, Leppert M, Hawk PR, Suarez A, Linares Y, Sapienza C: Familial aggregation of abnormal methylation of parental alleles at the IGF2/H19 and IGF2R differentially methylated regions. Hum Mol Genet. 2003, 12: 1569-1578. 10.1093/hmg/ddg167. Erratum in Hum Mol Genet 2004, 13:781View ArticlePubMedGoogle Scholar
- McLaren RJ, Montgomery GW: Genomic imprinting of the insulin-like growth factor 2 gene in sheep. Mamm Genome. 1999, 10: 588-591. 10.1007/s003359901050.View ArticlePubMedGoogle Scholar
- Uejima H, Lee MP, Cui H, Feinberg AP: Hot-stop PCR: a simple and general assay for linear quantitation of allele ratios. Nat Genet. 2000, 25: 375-376. 10.1038/78040.View ArticlePubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.