Leiden BioMedical Journal

Leiden BioMedical Journal 5e editie, juni 2014

Leids Universitair Medisch Centrum

Inhoud Voorwoord M.F.L.S.

3

Voorwoord Decaan

3

Overzicht onderwijsprijzen

4

Introductie genomineerden BW

6

Evaluation of biomarkers for CLL progression in the p53 pathway: a genetic approach - Elleke Peterse

8

Protection of proliferation of growth plate chondrocytes results in a decrease in damage to the growth plate following radiotherapy - Iris Pelsma

14

Cannabinoid type 1 receptor blockade diminishes dyslipidemia and obesity via peripheral activation of brown adipose tissue - Leonard Pelgrom

20

Introductie genomineerden GNK

26

Radiocarbon analysis of human tooth enamel to estimate the year of birth of unknown decedents- Larissa Brezden

28

Stromal BMP-2 expression is associated with poor survival in SMAD4 negative colorectal cancers- Thijs de Vos

34

Metabolomics analysis of pharmacotherapies for sensorineural hearing loss- Tjeerd Muurling

40

Oud-winnaars

46

LAG

47 Colofon 5e editie, juni 2014

Het Leiden BioMedical Journal (LBMJ) is een jaarlijkse uitgave van de Medische Faculteit der Leidse Studenten (M.F.L.S.) Artikelen gepubliceerd in het LBMJ zijn geen officiĂŤle publicaties.

Oplage: 500 ŠCopyright 2014. Niets uit deze uitgave mag worden verveelvoudigd en/of openbaar gemaakt worden zonder schriftelijke toestemming van de M.F.L.S.

2|Leiden BioMedical Journal |#5 | June 2014

Redactie: Wietske Boer - Lid Onderwijs Geneeskunde Merlin Weeda- Lid Onderwijs Biomedische wetenschappen Lay-out Eline Ruigrok- Lid Extern M.F.L.S., K1-R Postbus 9600, 2300 RC Leiden Telefoon 071-5264484 e-mail lbmj@mfls.nl

Voorwoord M.F.L.S. Beste lezer, Het Leiden BioMedical Journal (LBMJ) is een wetenschappelijk tijdschrift voor (bio)medische studenten van de Universiteit Leiden. Het is een platform voor studenten om de kennis te delen die zij middels de wetenschap hebben gevonden. Onderzoek doen gaat hand in hand met het academische leven en elke student draagt bij aan de wetenschap tijdens zijn of haar studententijd.

Op deze manier wordt de patiëntenzorg continu verbeterd en wordt nieuwe informatie bemachtigd.

Deze editie van het LMBJ bevat contributies van de genomineerden van de Student Research Awards 2013-2014. Het Leids Universitair Medisch Centrum (LUMC) organiseert deze prijzen om studenten te belonen die excellent (bio)medisch onderzoek hebben gedaan. Deze uitgave van het LBMJ geeft inzicht in de ontdekkingen van de studenten en geeft de lezer een kijkje in het leven van de auteur doormiddel van een achtergrondinterview.

Ieder jaar organiseert de Medische Faculteit der Leidse Studenten (M.F.L.S.) de Onderwijsprijzen in samenwerking met het LUMC. Tijdens de ceremonie worden prijzen uitgereikt om onderwijsprestaties te belonen en betrokkenen de aandacht te geven die zij verdienen. Tijdens de Onderwijsprijzen wordt het LMBJ feestelijk uitgereikt!

Ik hoop dat het LMBJ studenten verder motiveert om zich in te zetten voor het wetenschappelijke avontuur.

Ernest le Roy | Voorzitter M.F.L.S.

Voorwoord Decaan Het Leids Universitair Medisch Centrum is sterk wetenschappelijk georiënteerd.

Onderzoek staat hoog in het vaandel en is geïntegreerd in het onderwijs van zowel Geneeskunde als Biomedische wetenschappen. Als student krijg je tijdens de studie uitgebreid de kans om je verder te ontwikkelen op wetenschappelijk gebied. Het LUMC blinkt uit in een aantal onderzoeksgebieden; de profileringsgebieden. Dit zijn zorgvuldig uitgekozen profielen die klinisch en fundamenteel onderzoek zo goed mogelijk combineren.

De Onderwijsprijzen worden georganiseerd om bijzondere onderwijsprestaties in de Geneeskunde en Biomedische wetenschappen door studenten of docenten te belonen. Het onderwijs is één van de kerntaken van het LUMC en de prijzen geven het onderwijs de aandacht die het verdient.

Deze editie van het LBMJ bevat een aantal voorbeelden van hoogstaand wetenschappelijk werk verricht door onze studenten. De Raad van Bestuur van het LUMC feliciteert deze studenten met hun nominatie voor de Student Research Award 2013-2014. We hopen dat alle lezers geïnspireerd worden door ons wetenschappelijk karakter en nodigen jullie uit om hieraan deel te nemen.

Prof. Dr. Pancras C.W. Hogendoorn, Decaan LUMC

June 2014|#5|Leiden BioMedical Journal |3

M.F.L.S.-LUMC Onderwijsprijzen Wat zijn de M.F.L.S.-LUMC Onderwijsprijzen?

Je kent ze wel; docenten of studenten die net een stapje harder lopen dan de rest. Die ene docent, die vol passie kan vertellen over een bepaald onderdeel van de stof waardoor je het opeens helemaal begrijpt. Een blok waarin op allerlei creatieve manieren onderwijs wordt gegeven en blokcoördinatoren die klaar staan voor hun studenten. Maar ook studenten die een uitdagende onderzoekstage hebben gevonden en vele uren maken in het lab om de wetenschap een beetje verder te helpen. Al deze bijzondere talenten willen we wat extra in het zonnetje zetten. Met dit idee in het achterhoofd zijn ooit de M.F.L.S.-LUMC Onderwijsprijzen in het leven geroepen. Jaarlijks worden de prijzen uitgereikt tijdens een feestelijke bijeenkomst. Maar wat zijn precies de prijzen die beschikbaar zijn? En hoe worden de genomineerden bepaald? Om het voor jullie wat duidelijker op een rijtje te krijgen volgt hieronder een overzicht.

Prof. Dr. G.J. Tammeling prijs

De Prof. Dr. G.J. Tammelingprijs is de prijs voor de beste docent aan het LUMC van het betreffende academisch jaar. Er wordt geen onderscheid gemaakt tussen Biomedische wetenschappen en Geneeskunde. De prijs is ooit in het leven geroepen door Prof. Dr. G.J. Tammeling, voormalig decaan van de Raad van Bestuur van het LUMC. Vanaf 1987 wordt de prijs voor beste docent van het LUMC uitgereikt en sinds 1997 draagt de prijs ook de naam van de ouddecaan en dit eerste erelid van de M.F.L.S. In april wordt door alle Jaarvertegenwoordigingen met input van hun jaar één docent aangedragen. Uit alle jaren Biomedische wetenschappen en Geneeskunde komen dus 8 docenten naar voren.

De Studenten Advies Commissie bepaald uit deze 8 docenten de 3 genomineerde docenten en ook de uiteindelijke winnaar. De winnaar ontvangt de beroemde wisselbeker waarop ook alle namen van vorige winnaars een plaatsje hebben gekregen. Ook dit jaar zal de familie van dhr. Tammeling weer aanwezig zijn om de prijs uit te reiken.

prof. dr. G.J. Tammeling

Beste Blok Geneeskunde en Biomedische wetenschappen

Van alle blokken die in een jaar worden gegeven, springen er vaak een paar uit. Om de blokcoördinatoren en alle meewerkende docenten hiervoor extra te belonen wordt voor iedere studie een prijs voor het beste blok uitgereikt. Ieder cohort nomineert een blok en de Studenten Advies Commissie kies hieruit de drie genomineerden en het uiteindelijke winnende blok. De prijs die wordt uitgereikt is de Bronzen Veer.

Beste Co-schap

Naast een prijs voor het beste blok wordt er ook ieder jaar een beste co-schap aangewezen. Iedere co-assistent krijgt de mogelijkheid om het voor hem of haar beste co-schap te nomineren door een enquête. Deze enquêtes worden bekeken door de Leidse Co-Raad (LCR) en zij filteren hieruit de uiteindelijke winnaar. De winnaar ontvangt eveneens de Bronzen Veer.

4|Leiden BioMedical Journal |#5 | June 2014

Studenten Advies Commissie De Studenten Advies Commissie (SAC) bestaat uit één afgevaardigd studentlid van elke onderwijscommissie van het LUMC. De commissie wordt voorgezeten door de studentassessor van het LUMC. Met elkaar bespreekt de commissie actuele zaken op onderwijsgebied. De SAC formuleert één studentenmening en geeft aan de hand hiervan mogelijk een advies aan de verschillende onderwijscommissies. Een van de taken van de SAC is het bepalen van de winnaars van M.F.L.S.LUMC Onderwijsprijzen.

LUMC Student Research Award Geneeskunde en Biomedische wetenschappen

Ook voor studenten is er tijdens de Onderwijsprijzen aandacht. Iedere student doet in zijn studie in ieder geval één groot onderzoeksproject. Voor Biomedische wetenschappen is dit het Junior Research Project die de studenten schrijven in hun master. Voor Geneeskunde doet iedere student een wetenschapsstage, waarvan het verslag kan worden ingeleverd voor deze prijs. Een strenge doch rechtvaardige jury onder leiding van prof. dr. Frank Willem Jansen bepaalt de genomineerden en ook de uiteindelijke winnaars. Deze winnaars ontvangen ook de welbekende Bronzen Veer en daarnaast ook een geldbedrag van €2.500,- wat door de Raad van Bestuur ter beschikking wordt gesteld. Dit mooie bedrag mag de winnaar besteden aan een studiegerelateerd doeleinde. Je kunt bijvoorbeeld denken aan een onderzoek in het buitenland, een congresbezoek of een cursus.

Deze editie van het LBMJ staat volledig in het teken van de genomineerden voor de student research award.

De ingekorte versies van de artikelen laten zien wat een prachtig werk deze studenten hebben geleverd. Dit blad wordt uitgereikt tijdens de M.F.L.S.-LUMC Onderwijsprijzen op woensdag 11 juni 2014 om 16:30 uur in HePatho. Dan zal bekend worden gemaakt wie de winnaars zijn van de Student Reseach Award 20132014. Tevens krijgen de winnaars van de andere Onderwijsprijzen hun welverdiende prijs. Het belooft een spannende maar onvergetelijke avond te worden.

De Bronzen Veer De Bronzen Veer is ontworpen door kunstenaar een gynaecoloog prof.

Frank Willem Jansen. In de prijs, die voor het eerst werd uitgereikt in 2009, is een aantal symbolen verwerkt. Zo staat de bronskleurige veer symbool voor een pennenveer die in de oudheid de wetenschap vooruit bracht en daarnaast voor een pluim die de winnaar op een desgewenste plek kan stoppen. De vreugde die iemand hiermee toekomt is weergegeven in de zwierigheid waarmee het beeld op haar basis staat. Dat het beeld voorzien is van forse heupen is het teken van vruchtbaarheid. Het moge duidelijk zijn: de gecreëerde wetenschap is uit vruchtbare bodem voortgekomen. Echter het ontvangen van een prijs zal in ootmoed moeten geschieden, de gebogenheid van het hoofd is daar symbool voor.

Voor een overzicht van de winnaars van de Onderwijsprijzen van de afgelopen 5 jaar, zie pagina 46.

June 2014|#5|Leiden BioMedical Journal |5

Introductie genomineerden BW Naam: Elleke Peterse Leeftijd: 23 Startjaar studie: 2009 Titel verslag: Evaluation of biomarkers for CLL progression in the p53 pathway: a genetic approach Begeleider: dr. Gareth Bond, Ludwig Institute for Cancer Research, University of Oxford.

Kun je in het kort vertellen waar je onderzoek naar hebt gedaan?

Mijn project ging over chronische lymfatische B-cel leukemie (B-CLL). Men is nog niet goed in staat om bij de diagnose te kunnen voorspellen hoe het ziekteverloop zal zijn, sommige patiënten overlijden binnen enkele maanden terwijl andere nooit symptomen ontwikkelen. Daarom ging ik op zoek naar genetische variaties (SNPs) die het ziekteverloop van deze patiënten kunnen voorspellen en begon ik met bestuderen wat de cellulaire consequenties van deze SNPs zijn.

Wat vond je het leukste aan het doen van onderzoek?

Het leukste aan het onderzoek vond ik de samenwerking met vele internationale (PhD) studenten. Al deze studenten waren ontzettend ambitieus en gemotiveerd. Daarnaast waren er veel interessante lezingen van vooraanstaande sprekers. Zo heb ik een lezing van James Watson (van het onderzoeksduo Watson & Crick) bijgewoond, erg inspirerend.

Wat is de opvallendste bevinding die je hebt gedaan tijdens je onderzoek?

De opvallendste bevinding die ik heb gedaan is dat ik SNPs geïdentificeerd heb ik die mogelijk correleren met de prognose van B-CLL. Dit onderzoek is natuurlijk nog niet afgerond, daar is vijf maanden veel te kort voor.

Stel je wint de prijs, waar zou je het prijzengeld voor willen gebruiken? (Let op: je moet het geld wel besteden aan een studiedoel) Ik zou het geld willen gebruiken voor een Summer course, die ik dan in 2015 zou willen volgen. Mijn voorkeur gaat hierbij uit naar de Harvard University. Als dit te ambitieus blijkt te zijn zou ik ook nog graag terug gaan naar Oxford, voor een Summer course of een congres.

Welke richting wil je later op?

Per 1 oktober ga ik beginnen met mijn promotieonderzoek op de afdeling Pathologie van het LUMC. Mijn onderzoek focust zich op de rol van the insulin-like growth factor (IGF) pathway in Osteosarcoma. Wat ik daarna wil gaan doen weet ik nog niet.

Naam: Iris Pelsma Leeftijd: 24 Startjaar studie: 2008 Titel verslag: Protection of proliferation of growth plate chondrocytes results in a decrease in damage to the growth plate following radiotherapy Begeleider: Dr. J. Baron, NICHD, National Institutes of Health Kun je in het kort vertellen waar je onderzoek naar hebt gedaan?

Er zijn veel soorten kanker die bij kinderen voorkomen en die gevoelig zijn voor radiotherapie. Daarom worden ze nog steeds met radiotherapie behandeld, ondanks alle bijwerkingen. De bijwerking waar ik me op gericht heb is ‘short stature’, wat betekent dat de lengte van een kind voor zijn of haar leeftijd meer dan 2 standaarddeviaties onder het gemiddelde is. Als kinderen worden bestraald en in het gebied ligt een groeischijf, wat bij ‘total body 6|Leiden Biomedical Journal |#5 | June 2014

irradiation’ altijd voorkomt, zullen de cellen van de groeischijf voor een deel afsterven en zal de uiteindelijke lengte van het kind worden beïnvloed.

Ik heb geprobeerd om een oplossing te vinden voor dit probleem, door specifiek de groeischijf te beïnvloeden en te beschermen tegen bestraling.

Wat is de opvallendste bevinding die je hebt gedaan tijdens je onderzoek?

De meest opvallende bevinding was dat het medicijn dat we voor ogen hadden al direct beschermend leek te werken in muizen tijdens de eerste trials. Dat hadden we geen van allen zo snel verwacht, ook al is het laatste woord hierover natuurlijk nog niet gezegd. Doorgaans zijn meerdere versies van een experiment nodig om duidelijk resultaat te zien.

Stel je wint de prijs, waar zou je het prijzengeld voor willen gebruiken? (Let op: je moet het geld wel besteden aan een studiedoel) Ik heb het onderzoek gedaan als eerste masterstage van mijn studie Biomedische wetenschappen. Om af te studeren moet ik nog een tweede onderzoeksstage doen. Deze zou ik graag weer in Amerika doen en daarvoor kan ik zeker het prijzengeld goed gebruiken.

Welke richting wil je later op?

Ik studeer naast BW ook sinds 2011 Geneeskunde via het Dubbeltraject en ik wil later graag kinderarts worden, het liefst op een Intensive Care. Daarnaast zou ik ook graag onderzoek willen blijven doen en/of coördineren.

Het meest ideale zou zijn als ik mijn tijd 50/50 zou kunnen verdelen tussen arts en onderzoeker en ik hoop dat dat me ooit gaan lukken.

Naam: Leonard Pelgrom Leeftijd: 25 jaar Startjaar studie: 2006 Titel verslag: The CB1R inverse agonist Rimonabant diminishes dyslipidemia and obesity via peripheral activation of brown adipose tissue Begeleider: Mariëtte Boon& Patrick Rensen

Kun je in het kort vertellen waar je onderzoek naar hebt gedaan?

Ik heb onderzoek gedaan naar het mechanisme waarmee Rimonabant, een therapeutisch middel met antiobesitas eigenschappen, stoornissen in de vetstofwisseling vermindert. Een afname in de hoeveelheid vetten in het bloed bleek voornamelijk het gevolg te zijn van opname door bruin vet. In tegenstelling tot wit vet, dat overtollige energie opslaat, verbrandt bruin vet de energiereserves voor de productie van warmte.

Waarom denk je dat je genomineerd bent voor de Student Research Award?

Ik hoop dat de nomatie het gevolg is van een prikkelend en vooral lekker weg te lezen verslag. Wetenschap bedrijf je met elkaar en daarom is communicatie, waarvan je bevindingen presenteren in de vorm van een verslag/artikel één onderdeel is, zo belangrijk.

Stel je wint de prijs, waar zou je het prijzengeld voor willen gebruiken?

Ik zou de prijs gebruiken om een extra stage te doen in het buitenland voordat ik aan een PhD-traject begin.

Ik heb namelijk al mijn stages gedaan in het Leids Universitair Medisch Centrum en ik zou de buitenlandse ervaring goed kunnen gebruiken om een juiste keuze te maken voor mijn loopbaan na het afstuderen.

Welke richting wil je later op?

Ik wil verder het onderzoek in om expertise op te bouwen in het gebied van stofwisselingziekten. Op lange termijn wil ik mij bezig houden met beleid en preventie van overgewicht en gerelateerde aandoeningen.

Heb je nog tips voor studenten die onderzoek willen gaan doen?

Probeer een begeleider een een groep te zoeken waarbij je gelijk een klik hebt. Wetenschap gaat niet altijd over rozen en plezier hebben in je werk helpt om eventuele hobbeltjes in de weg te nemen.

June 2014|#5|Leiden Biomedical Journal |7

Elleke Peterse

Evaluation of biomarkers for CLL progression in the p53 pathway: a genetic approach Elisabeth F.P. Peterse, Jorge Zeron-Medina, Anna M. Grawenda, Emmanouela Repapi, Juliet Hewitt, Elisabeth E Bond, Gareth L. Bond Abstract B-cell chronic lymphocytic leukemia (B-CLL) is characterized by a varied symptoms course, and genetic markers that could help detect patients at high risk of faster progression and poorer survival could help manage treatment options. An important signaling cascade in B-CLL is the p53 stress response, whose genes are frequently somatically mutated resulting in altered progression. Besides somatic mutations, inherited genetic variations, such as single nucleotide polymorphisms (SNPs) can also attenuate p53 signaling. Therefore, a systematic screening of known functional SNPs in the p53 pathway was performed in the hopes of identifying potential genetic biomarkers of B-CLL progression. We identified 22 functional p53 pathway SNPs, 15 of which we successfully genotyped and examined in a patient cohort (n=481). Two SNPs, rs11968166 and rs11090865, showed significant correlations with both treatment free survival (log-rank: p=0.011 and p=0.007, respectively) and overall survival (log-rank: p=0.002 and p=10-6, respectively). Interestingly, one of the SNPs, SNP rs11968166, is closely linked to two other SNPs (rs3218086 and rs3218092), all of which reside in non-coding regions that harbor signatures of transcriptional enhancer elements as defined by ENCODE. Moreover, rs11968166 and rs3218092 are predicted to alter transcription factor binding, suggesting these SNPs could alter gene expression resulting in the observed differences in B-CLL progression. Indeed, we observed differential binding of protein nuclear extracts to DNA oligonucleotides containing either alleles of rs11968166 and rs3218092. Future research will identify if these differences in protein binding translate into allelic differences in transcriptional regulation.

Introduction B-cell chronic lymphocytic leukemia (B-CLL) is the most common type of leukemia in Western countries with an increased incidence in elderly individuals.

Interestingly, there is great heterogeneity in patients with B-CLL in terms of disease progression, time to first treatment and survival. Some patients already have symptoms at diagnoses and die within a few years while others can live for decades and die of unrelated causes (1). Staging alone cannot predict the course of the disease in these patients. For this reason it is important to identify genetic markers that can help detect patients at high risk of faster progression and poorer survival so that their treatment can be handled accordingly.

One important genetic marker that has been incorporated into risk-stratification to determine firstline treatment approaches for patients with B-CLL is the deletion of the p53 tumor suppressor gene (2). The loss of p53 or mutations in crucial nucleotides has been shown to significantly accelerate B-CLL progression, hamper response to chemotherapy and shorten overall survival (3). This is not specific for B-CLL, since it is estimated that 50% of all human cancers mutate the p53 gene to ablate p53 function. However, a large subset of the remaining cancers inhibit p53 function 8|Leiden BioMedical Journal |#5 | June 2014

by altering other central genes in the p53 pathway (4). One gene that is often overexpressed in cancer is MDM2, which targets p53 for degradation. Not only somatic mutations can results in higher MDM2 levels, as demonstrated by the discovery of an inherited single nucleotide polymorphism (SNP) in the MDM2 promoter (MDM2 SNP309 T/G) (5).

SNPs are the most common type of inherited genetic variation with a minimum frequency of 1% in a given population. The vast majority of SNPs are thought to have little or no effects on human phenotypic variation. However, functional SNPs could contribute to wide spectrum of differences observed between individuals including susceptibility to cancer and differential response to treatments, as shown by the MDM2 SNP309 T/G. This genetic variant results in an increased expression of MDM2, resulting in decreased p53 activity and an increased risk and poor therapeutic response of multiple types of cancer, including B-CLL (6-10). This suggest that other functional high frequency inherited polymorphisms in genes in the p53 signaling pathway could also impact B-CLL, and therefore further identify those patients with attenuated p53 signaling. The identification of such genetic variants could also further increase our

Nominatie Biomedische wetenschappen understanding of the cellular processes involved in the variable course of this disease, thereby exposing potential nodes of therapeutic intervention. Thus, in this study, a systematic screening of known functional SNPs in the p53 pathway was performed.

Results Identification of Functional p53 Pathway SNPs Sixty-nine genes in the p53 pathway were identified using the KEGG database, which records networks of molecular interactions in the cells. SNPs in these genes that are associated with human traits in previously published Genome-Wide Association Studies (GWAS) were identified by searching for these genes in three databases: The National Human Genome Institute (NHCRI) catalogue (www.genome. gov/gwasstudies), the GWAS central database (www.gwascentral.org) and the National Centre for Biotechnology Information (NCBI) database (www. ncbi.nlm.nih.gov). In these genes, 22 SNPs in 13 genes were identified that are associated with human traits.

We were able to obtain genotypes for 15 SNPs of these 22 SNPs (or linked SNPs with an r2 > 0.8) in data derived from an Affymatrix Genome-wide human SNP array 6.0 generated in a subset of the patients from the Discovery Cohort (n=185) (11).

Interestingly, four SNPs showed either significant or borderline significant associations with Treatment Free Survival (TFS) or Overall Survival (OS). These four SNPs are rs1801516, rs11970772, rs11968166 and rs11090865 and we genotyped these in the remaining patients of the Discovery Cohort (n=296) (Figure 1). This genotyping was performed by PCR amplification of genomic DNA and subsequent allelic discrimination using SNP genotyping assays from Applied Biosystems as previously described (12).

Clinical correlation After our genotyping and together with the genotypes obtained from the Array data, we were able to explore potential associations in a cohort of 481 patients with B-cell chronic lymphocytic leukemia and complete data for OS and TFS (for clinical characteristics, see table 1). No significant allelic differences were noted for rs1801516 in either OS or TFS in our genotyped cohort, tested by Log-ranks and Cox multivariate regression models (Factors included can be seen in table 1). In contrast, for rs11970772, significant allelic differences were noted in OS (p=0.048 Log-rank, p=0.025 Cox). Specifically, patients with at least one T-allele harbor an increased hazard ratio (HR) of 1.5 (95% CI: 1.099, 2.039) compared to patients homozygous for the X-allele. Rs11968166 correlates significantly with both TFS (p=0.011) and OS (p=0.002, Figure 2).

Combined =

SNP array

Our genotyping

481

185

296

323 (67) 152 (32) 6 (1)

138 (75) 47 (25) -

185 (63) 105 (35) 6 (2)

Stage MBL A B C Missing

48 (10) 217 (45) 123 (26) 82 (17) 11(2)

80 (43) 63 (34) 38 (21) 4 (2)

48 (16) 137 (46) 60 (20) 44 (15) 7 (3)

Deletion TP53 Yes No Missing

21 (4) 423 (88) 37 (8)

21 (4) 423 (88) 37 (8)

9 (3) 263 (89) 24 (8)

Deletion ATM Yes No Missing

81 (17) 369 (77) 31 (6)

50 (27) 124 (67) 11 (6)

31 (10) 245 (83) 20 (7)

IgVH Mutated Unmutated Missing

193 (40) 276 (57) 12 (3)

73 (39) 100 (54) 12 (7)

120 (41) 176 (59) -

Total amount Gender

Male Female Missing

Table 1 - Clinical characteristics of the discovery cohort.

Figure 1 - Methodology followed for the selection of the SNPs for genotyping in the remaining patients of the discovery cohort.

The associations become slightly less significant after correcting for known prognostic factors, were the association with TFS is borderline significant (p=0.059) while the OS is still significant (p=0.023).

Furthermore, the Cox multivariate regression models reveal a HR of 1.40 (95% CI: 1.05, 1.87) for TFS in patients carrying one A allele and a HR of 1.34 (95% CI: 0.772, 2.32) for patients carrying two A alleles.

The HR is 1.473 (95% CI: 1.068, 2.03) and 1.731 (95% CI: 0.945, 3.170) for OS for patients carrying one or June 2014|#5|Leiden BioMedical Journal |9

Elleke Peterse two A alleles, respectively. Rs11090865 demonstrates very significant allelic differences in TFS and OS with p-values of 0.007 and 10-6 respectively in our genotyped cohort. The Cox multivariate regression model for TFS has a P-value of 0.017 with HR of 4.17 (95% CI: 1.23, 14.08) for TT vs GT carriers and 2.55 (95% CI: 0.779-8.33) for TT vs GG carriers.

For OS, the P-value is 0.008 in the Cox multivariate regression model, with Hazard ratios of 13.0 (95% CI: 3.75, 45.5) and 10.9 (95% CI: 3.25, 37.0) for respectively TT vs GT and TT vs GG carriers.

In summary, of the four SNPs tested, rs11090865 and rs11968166 associated with significant allelic differences in both TFS and OS. We therefore decided to focus on these two SNPs in our further analyses.

Potential Molecular Function of SNPs In order to begin to explore the possibility that rs11090865 and rs11968166 could alter the function or levels of p53 pathway gene(s), we determined which other SNPs are closely linked to these SNPs in European populations using the SNP annotation and proxy search (SNAP) from the 1000 genomes project (13, 14). Surprisingly, we found that 112 additional SNPs are in linkage disequilibrium with rs11090865 (with an r2 > 0.8 in CEU). In contrast, only 13 SNPs were closely linked to rs11968166. Given the time frame of this study, we decided to focus our subsequent analysis and experiments on rs11968166, which resides in the cyclin D3 gene, and its linked variants in order to explore the possibility that one of these SNPs could alter the function or levels of this p53 pathway gene. All 14 SNPs reside in non-coding regions either

in or surrounding the Cyclin D3 (CCND3) Gene. In order to identify which SNPs could reside in potential regulatory regions we took advantage of data derived from the Encyclopaedia of DNA Elements Project (ENCODE)(15). Specifically, we scanned the 14 SNPs for residing in regions that contained epigenetic signatures that characterize potential transcriptional enhancer regions, namely histone post-translational modifications such as the mono-methylation of histone 3 in lysine 4 and the acetylation of histone 3 in lysine 27. Interestingly, three SNPs, namely rs3218086, rs3218092 and rs11968166 itself did.

These data suggest that one or all of these three SNPs could lie in transcriptional enhancer regions.

To further explore this hypothesis, the possibility that these SNPs could demonstrate allelic differences in transcription factor binding was explored. To do this, we used the Explain 3.0 algorithm with the TRANSFAC 2009.3 database (16) and noted that rs11968166 (A/G) is predicted to affect binding of the transcription factors AHR, ARNT and HIF1 (Figure 3A). Moreover, rs3218092 (C/T) is predicted to affect binding of the transcription factor E2F (Figure 3B).

No allelic difference in transcription factor binding for rs3218086 was predicted by the algorithm.

In order to begin to further test these predictions, we preformed, a protein/DNA binding assay, the electro-mobility shift assay (EMSA). For this assay, oligonucleotide probes (25bp) sets for rs11968166 (A/G), rs3218086 (G/A) and rs3218092 (T/C) were obtained. Nuclear protein extracts were derived from 2 different cell lines: a melanoma cell line, 501, which was cultured under hypoxic conditions for twelve

Figure 2 - Kaplan-Meiers of rs11968166. TFS=Treatment Free Survival, OS=Overall Survival

10|Leiden BioMedical Journal |#5 | June 2014

Nominatie Biomedische wetenschappen

Figure 3 - The result of the transcription factor binding computational algorithm for rs11968166 (A) and rs3218092 (B). Rs11968166 is predicted to bind differentially to AHR, ARNT and HIF1 and rs3218092 is predicted to bind differentially to E2F.

hours and therefore had high levels of HIF1, one of the predicted transcription factors, and renal cancer cell line, RCC4, which has high HIF1 levels due to VHL loss (17). The results are depicted in Figure 4.

Strikingly, significantly more protein/DNA complexes can be observed for rs11968166 G compared to rs11968166 A, as predicted. No clear differences were observed between rs3218086 G and rs3218092 A, also as predicted. Also as predicted, for rs3218092, the C allele showed more protein/DNA complexes than the T allele. Together, these results indicate that both rs11968166 and rs3218092 could affect transcriptional regulation and underlie the noted associations between rs11968166 and B-CLL prognoses.

Figure 4 - Autoradiograph of an EMSA with nuclear extract derived from hypoxic 501 and RCC4 cells. Differential binding is observed for rs11968166 and rs3218092 in both extracts.

Discussion In this study, we set out to explore if functional SNPs in p53 pathway genes could serve as biomarkers for B-CLL progression. We were able to screen 15 SNPs and identified two SNPs, rs11968166 and rs11090865 that were significantly associated with B-CLL treatment free survival and overall survival in a patient cohort (n=481). These SNPs reside in the p53 pathway genes, Cyclin D3 and G2 and S phase-expressed protein 1, both of which have been shown to be important in regulating p53 responses to cellular stresses.

During our explorations into the abilities of these two SNPs or their linked SNPs to alter either the function or levels of the p53 pathway genes, we were able to make substantial progress, specifically for rs11968166.

Rs11968166 is located in an intron of the Cyclin D3 gene (CCND3). CCND3 is a member of the Cyclin D gene family, which is involved in the regulation of cell cycle progression (18). A total of 13 SNPs were identified that are closely linked to rs11968166. Based on the presence of histone modifications that serve as markers for regions with regulatory potential (19), we identified three of these SNPs that lie in potential regulatory regions. Specifically, these SNPs, rs11968166, rs3218086 and rs3218092, were positive for enhancer markers (H3K4Me1 and/or H3K27Ac) in K562, which is chronic myeloid leukemia cell line (20). Two of these SNPs also were predicted to harbor allelic differences in transcription factor binding. Specifically, rs11968166 is predicted to affect binding of three related transcription factors: Hypoxia-inducible factor (HIF1), Aryl Hydrocarbon Receptor (AHR) and aryl hydrocarbon receptor nuclear translocator (ARNT) (21, 22). Strikingly, HIF1-Îą, which is the oxygen regulated subunit of HIF1, is expressed even in normaxia in CLL cells (16, 23). The function of HIF1 in CLL is mainly attributed to its function to induce vascular endothelial growth factor (VEGF) transcription (24), which induces tissue neovascularisation in the bone marrow (25) and lymph nodes (26) of CLL patients. In addition, HIF1-Îą is also involved in stabilizing p53 (27, 28). P53 induces the expression of p21, which is an inhibitor of the cyclin D/CDK4-6 complex. Therefore, CCND3 is located downstream of HIF1 in the p53 pathway.

Our analysis also predicted allelic differences in transcription factor binding for rs3218092 to E2F, where the minor allele binds to E2F and the major allele does not. E2F is a transcription factor that, when dissociated from pRB under the influence of Cyclin D/ CDK4 complex, can induce the transcription of genes June 2014|#5|Leiden BioMedical Journal |11

Elleke Peterse that are necessary for G1/S transition. It is therefore located downstream of cyclin D3 in the p53 pathway.

As HIF1 was one of the transcription factors predicted to bind to rs11968166 in an allele-specific manner, the protein/DNA binding assay was performed using nuclear extracts from VHL- cell line RCC4 and a melanoma cell line 50 which was cultured in hypoxic conditions. The results clearly demonstrate that the rs11968166 G allele had stronger binding affinities to a protein than the A allele. The computational analysis predicts this protein to be HIF1, ARNT or AHR. A difference in the binding of the DNA fragments of rs3218092 was also observed, were the C allele has a stronger signal than the A allele for a certain protein.

We hypothesize, based on the results from the computational algorithm used to predict binding to transcription factors, that this factor could be E2F. In conclusion, during the time course of this study, two inherited genetic biomarkers for B-CLL progression were identified that merit further study. Moreover, we identified a potential molecular underpinning of these associations for one of the SNPs. Specifically, rs11968166 and one of its linked SNPs, rs3218092, reside in transcriptional regulatory regions, are predicted to alter transcription factor binding, and indeed associate with differential protein/DNA complexes. Together these data suggest that these SNPs could alter the transcriptional regulation if the p53 pathway gene, Cyclin D3, and therefore underlie the associations with altered B-CLL progression.

12|Leiden BioMedical Journal |#5 | June 2014

References 1.

Rozman C, Montserrat E. Chronic lymphocytic leukemia. N Engl J Med. 1995;333(16):1052-7.

2.

Zenz T, Dohner H, Stilgenbauer S. Genetics and risk-stratified approach to therapy in chronic lymphocytic leukemia. Best practice & research Clinical haematology. 2007;20(3):439-53.

3.

Badoux XC, Keating MJ, Wierda WG. What is the best frontline therapy for patients with CLL and 17p deletion? Current hematologic malignancy reports. 2011;6(1):36-46.

4.

Vogelstein B, Lane D, Levine AJ. Surfing the p53 network. Nature. 2000;408(6810):307-10.

5.

Bond GL, Hu W, Bond EE, Robins H, Lutzker SG, Arva NC, et al. A single nucleotide polymorphism in the MDM2 promoter attenuates the p53 tumor suppressor pathway and accelerates tumor formation in humans. Cell. 2004;119(5):591-602.

6.

Post SM, Pant V, Abbas H, Quintas-Cardama A. Prognostic impact of the MDM2SNP309 allele in leukemia and lymphoma. Oncotarget. 2010;1(3):16874.

7.

Gryshchenko I, Hofbauer S, Stoecher M, Daniel PT, Steurer M, Gaiger A, et al. MDM2 SNP309 is associated with poor outcome in B-cell chronic lymphocytic leukemia. J Clin Oncol.

2008;26(14):2252-7.

8. Willander K, Ungerback J, Karlsson K, Fredrikson M, Soderkvist P, Linderholm M. MDM2 SNP309 promoter polymorphism, an independent prognostic factor in chronic lymphocytic leukemia.

Eur J Haematol. 2010;85(3):251-6.

9.

Asslaber D, Pinon JD, Seyfried I, Desch P, Stocher M, Tinhofer I, et al. microRNA-34a expression correlates with MDM2 SNP309 polymorphism and treatment-free survival in chronic lymphocytic leukemia. Blood. 2010;115(21):4191-7.

10. Dong HJ, Fang C, Fan L, Zhu DX, Wang DM, Zhu HY, et al. MDM2 promoter SNP309 is associated with an increased susceptibility to chronic lymphocytic leukemia and correlates with MDM2 mRNA expression in Chinese patients with CLL. Int J Cancer. 2012;130(9):2054-61.

11.

Parker H, Rose-Zerilli MJ, Parker A, Chaplin T, Wade R, Gardiner A, et al. 13q deletion anatomy and disease progression in patients with chronic lymphocytic leukemia. Leukemia. 2011;25(3):489-97.

12.

Vazquez A, Grochola LF, Bond EE, Levine AJ, Taubert H, Muller TH, et al. Chemosensitivity profiles identify polymorphisms in the p53 network genes 143-3tau and CD44 that affect sarcoma incidence and survival. Cancer Res. 2010;70(1):172-80.

Nominatie Biomedische wetenschappen 13.

Johnson AD, Handsaker RE, Pulit SL, Nizzari MM, O’Donnell CJ, de Bakker PI. SNAP: a web-based tool for identification and annotation of proxy SNPs using HapMap. Bioinformatics. 2008;24(24):2938-9.

14. Broad Institute SNP Annotation and Proxy Search. .Available from www.broadinstitute.org/mpg/ snap/Idsearch.php2013.

15. Birney E, Stamatoyannopoulos JA, Dutta A, Guigo R, Gingeras TR, Margulies EH, et al.

Identification and analysis of functional elements in 1% of the human genome by the ENCODE pilot project. Nature. 2007;447(7146):799-816.

16. Shachar I, Cohen S, Marom A, BeckerHerman S. Regulation of CLL survival by hypoxiainducible factor and its target genes. FEBS Lett.

2012;586(18):2906-10.

17. Maxwell PH, Wiesener MS, Chang GW, Clifford SC, Vaux EC, Cockman ME, et al. The tumour suppressor protein VHL targets hypoxia-inducible factors for oxygen-dependent proteolysis. Nature.

1999;399(6733):271-5.

18.

Musgrove EA, Caldon CE, Barraclough J, Stone A, Sutherland RL. Cyclin D as a therapeutic target in cancer. Nature reviews Cancer. 2011;11(8):558-72.

19.

Maher B. ENCODE: The human encyclopaedia.

Nature. 2012;489(7414):46-8.

20. Lozzio CB, Lozzio BB. Human chronic myelogenous leukemia cell-line with positive Philadelphia chromosome. Blood. 1975;45(3):321-34.

21. Wang GL, Jiang BH, Rue EA, Semenza GL. Hypoxia-inducible factor 1 is a basic-helix-loop-helixPAS heterodimer regulated by cellular O2 tension.

Proceedings of the National Academy of Sciences of the United States of America. 1995;92(12):5510-4.

22. Lindebro MC, Poellinger L, Whitelaw ML. Protein-protein interaction via PAS domains: role of the PAS domain in positive and negative regulation of the bHLH/PAS dioxin receptor-Arnt transcription factor complex. The EMBO journal. 1995;14(14):352839.

23.

Ghosh AK, Shanafelt TD, Cimmino A, Taccioli C, Volinia S, Liu CG, et al. Aberrant regulation of pVHL levels by microRNA promotes the HIF/VEGF axis in CLL B cells. Blood. 2009;113(22):5568-74.

24.

Skinner HD, Zheng JZ, Fang J, Agani F, Jiang BH. Vascular endothelial growth factor transcriptional activation is mediated by hypoxia-inducible factor 1alpha, HDM2, and p70S6K1 in response to phosphatidylinositol 3-kinase/AKT signaling. The Journal of biological chemistry. 2004;279(44):4564351.

25.

Kini AR, Kay NE, Peterson LC. Increased bone marrow angiogenesis in B cell chronic lymphocytic leukemia. Leukemia. 2000;14(8):1414-8.

26.

Chen H, Treweeke AT, West DC, Till KJ, Cawley JC, Zuzel M, et al. In vitro and in vivo production of vascular endothelial growth factor by chronic lymphocytic leukemia cells. Blood. 2000;96(9):31817.

27. An WG, Kanekal M, Simon MC, Maltepe E, Blagosklonny MV, Neckers LM. Stabilization of wildtype p53 by hypoxia-inducible factor 1alpha. Nature.

1998;392(6674):405-8.

28.

Chen D, Li M, Luo J, Gu W. Direct interactions between HIF-1 alpha and Mdm2 modulate p53 function. The Journal of biological chemistry.

2003;278(16):13595-8.

June 2014|#5|Leiden BioMedical Journal |13

Iris Pelsma

Protection of proliferation of growth plate chondrocytes results in a decrease in damage to the growth plate following radiotherapy ICM Pelsma, JM Wit, J Baron Abstract Irradiation as a treatment modality, either given in a single dose or fractioned, has been shown to result in impaired linear growth and reduced final height. We hypothesized that administration of glucocorticoid drugs would decrease chondrocyte proliferation and thus protect growth plate from damaging radiotherapy. We used both an in vitro and an in vivo model to test this hypothesis. In rat metatarsal bone culture, administration of high-dose dexamethasone significantly inhibited growth rate and chondrocyte proliferation (45 – 66 % decrease versus controls). Moreover, we created an in vivo model using 3-week-old C57Bl/6 mice treated with high-dose dexamethasone prior to 15 - 17.5 Gy radiation to the knee joint. We found that 17.5 Gy irradiation to the hind leg appeared to be unsalvageable by dexamethasone pre-treatment. However, animals that were radiated with 15 Gy and treated with dexamethasone appeared to have improved growth 1 and 2 weeks after radiation, but this finding requires confirmation. Future experiments will be done to optimize the conditions to determine whether dexamethasone could be used as a protective drug prior to radiotherapy.

Introduction Multimodality therapy that includes surgery, chemotherapy and radiation has become the standard of care for numerous childhood malignancies. In contrast to the therapeutic damage to the cancerous cells, radiation also leads to acute and late damage to healthy tissues (1). The more severe side effects that have been described involve endocrine dysfunction and disturbances including growth failure, gonadal failure and primary hypothyroidism (2, 3). Radiation exposure can cause shorter stature by injuring the growth plate, which is responsible for longitudinal bone growth and therefore determines a child’s height (4). Impaired linear growth and reduced final height have been shown to occur in individuals treated with either single dose or fractionated total body irradiation (TBI) (2, 5-12).

Longitudinal bone growth involves the production of cartilage in the growth plate, which will be first converted into woven spongy bone and then further modeled and remodeled into lamellar trabecular bone in the metaphysis (13), a process which is referred to as endochondral ossification.

replenish the pool of highly replicative proliferative chondrocytes (15). After multiple divisions, the proliferative chondrocytes stop dividing and terminally differentiate to hypertrophic chondrocytes (14). After the chondrocyte population has undergone apoptosis, the invading bone cells remain to form the initial structure of bone.

The endocrine regulation of the growth plate and its chondrocytes is very complex. Multiple hormones and derivatives have been described to increase or decrease the proliferative and survival capacity of chondrocytes (16). Glucocorticoids are widely used anti-inflammatory and immunosuppressive drugs in the pediatric clinics. In contrast to their short term positive effects, long term treatment with glucocorticoids can lead to growth failure and numerous other side effects. This growth inhibition mostly affects longitudinal bone growth and is partly due to a direct effect on chondrocytes in the epiphysis. Glucocorticoid administration to cultured chondrocytes inhibits their proliferation (17), but also conserves the number of resting zone chondrocytes (18). Furthermore, systemic and local administration To give a short overview, the growth plate, consisting of glucocorticoids inhibits longitudinal bone growth solely of chondrocytes, comprises three principal in vivo (19, 20). layers: the resting zone, proliferative zone and It has been previously described that treatment with hypertrophic zone. The resting zone chondrocytes dexamethasone inhibits growth in chondrocyte replicate slowly (14) and act as progenitor cells that culture and in a culture model for postnatal rat 14|Leiden BioMedical Journal |#5 | June 2014

Nominatie Biomedische wetenschappen metatarsal bones. Moreover, the effects of either fractionated or non-fractionated radiotherapy have been described in the metatarsal culture model. In this study we would like to combine the knowledge on glucocorticoids and their actions related to the epiphysis with the negative effects of radiation on the growth plate. We hypothesize that the administration of glucocorticoids decreases proliferation which could lead to protection from damaging radiotherapy. To test our hypothesis, we have used two different research models. The rat metatarsal culture system was used to test the protective potential of dexamethasone in vitro.

In addition, we have translated our findings from the in vitro model to a mouse model to mirror the effects of both glucocorticoid treatment and radiation in vivo.

Results Dexamethasone treatment increases proliferation rate after 2 days and 12 Gy of radiation Metatarsal bones were dissected and randomized in 4 treatment groups: control with and without 12 Gy of radiation and 1 µM of dexamethasone with and without 12 Gy of radiation after 2 days of treatment. When looking at the proliferation rate 2 days after radiation, the bones that were treated with dexamethasone and radiated with 12 Gy showed a 25 % higher proliferation rate than the control bones.

However, in the bones that did not receive radiation, dexamethasone-treated bones showed a 50% higher proliferation rate than control bones (Figure 1).

400

CTL 1 uM DEX

**

BrdU + cells

300

200

***

***

100

0 0

12

Dose of radiation

Figure 1 -Withdrawal from high dose dexamethasone significantly increase chondrocyte proliferation in metatarsal bones Metatarsal bones were dissected from 4 day-old rats and cultured ex vivo in DMEM organ culture medium with 1 µM dexamethasone or control medium for two days. Half of the bones were radiated with 12 Gy on day 2 of culture. After radiation the bones were cultured in normal organ culture medium. The bones were harvested 2 days after radiation. Chondrocyte proliferation was analyzed based on BrdU incorporation during 24 hours. The number of BrdU positive cells was counted manually. Data represents mean ± SEM. ** P<0.01 *** P<0.001 by one-way ANOVA.

Linear growth and proliferation of chondrocytes is stimulated after withdrawal of dexamethasone Metatarsal bones were randomized into three treatment groups: control, 1 µM dexamethasone (high-dose) and 1 pM dexamethasone (low-dose).

The bones were treated for 2 days and followed for 2 more days. After two days the bones were washed and changed into either control medium for the control group or 1 pM dexamethasone medium for the dexamethasone treated groups. On day 1 of culture, the effect of low versus high-dose dexamethasone on the metatarsal growth rate is the most prominent, as was seen in the previous experiments (Figure 2B). The growth rate was increased 45 % for bones treated with 1 pM dexamethasone and decreased 44 % for bones treated with 1 µM dexamethasone in comparison to controls. On day 2 of culture, the effect of dexamethasone on the growth rate seems to be no longer present. After washing, the bones previously treated with high-dose dexamethasone display an impressive increase in growth rate. In terms of cumulative growth, the low-dose dexamethasone bones reach the greatest average length, indicating that treatment with low-dose dexamethasone really is stimulating linear growth (Figure 2A). The bones treated with high-dose dexamethasone displayed inhibition of growth and this inhibition was released after the washing. In addition to the growth rate data, the proliferation rate of the 1 µM dexamethasone treated bones is a significant two-and-a-half fold higher than the control bones or the ones treated 1 pM dexamethasone (data not shown).

Dexamethasone injections prior to irradiation protects growth rate in vivo Twenty 3-week old C57Bl/6 mice were randomized in four treatment groups: one group receiving a single dose of 15 Gy to the right knee joint and one group receiving 17.5 Gy to the right knee joint. Mice were treated with 40 µg dexamethasone daily, which significantly reduced chondrocyte proliferation in vivo (data not shown), or saline as control for 7 days prior to radiation exposure. The animals were followed for 3 weeks. After one week of dexamethasone treatment, there were differences in both the femur and tibia length between the treated and non-treated animals. One week post radiation, the control and dexamethasone treated bones that were not radiated grew at a similar speed. The control bones radiated with a single dose of 15 Gy grew significantly less than the control bones, with a 57.4 % decrease in growth rate (data not shown). The bones that were previously June 2014|#5|Leiden BioMedical Journal |15

Iris Pelsma

A

*

0.8

10.5

**

0.6

** 0.4

CTL 1 pM DEX 1 uM DEX

9.5

1

2

3

1

4

0.6

B **

2

3

1.0

0.8

Growth rate (mm.day)

0.5

CTL DEX CTL 17.5 Gy DEX 17.5 Gy

8.5 8.0

0

Growth rate (mm/day)

10.0

9.0 0.2

0.0

B

11.5 11.0

Length (mm)

Cumulative growth (mm)

A

0.4

0.3

***

0.2

**

0.1

0.0

0.6

0.4

0.2

0.0 1

2

3

4

Days of culture

2

3

Weeks of follow-up

Figure 2 -High dose dexamethasone inhibits the growth rate of cultured metatarsal bones Metatarsal bones were dissected from 4 day-old rats and cultured ex vivo in DMEM organ culture medium with or without the indicated concentrations of dexamethasone (DEX). DEX was added to the culture medium the first 2 days of culture. Thereafter, the bones were washed 6 times in 1 mL PBS and 1 time with 0.8 mL medium for one hour.

Afterwards, the high dose DEX metatarsal bones were transferred to low dose DEX, the other groups remained the same. The cumulative growth (A) and growth rates (B) were calculated daily from bone length measurements. Data represents mean Âą SEM. *** P<0.001 ** P<0.01 by one-way ANOVA.

Figure 3 - Dexamethasone induces catch-up growth after 7 days of treatment in both radiated and non-radiated bones Three-week-old C57BL/6 mice were treated with 40 ug dexamethasone or vehicle daily for 7 days. On day 7, the right hind leg received a single dose of 17.5 Gy. Weekly, the animals were radiographed to assess femur length (A).Growth rates (B) of the femur were calculated from weekly bone length measurements. Data represents mean Âą SEM.

treated with dexamethasone and received the 15 Gy radiation dose had a higher growth rate than the bones that were not treated (43 % decrease in growth rate versus controls). Two weeks post radiation, for both the radiated and non-radiation bones, the bones previously treated with dexamethasone had higher growth rates. In contrast to what was found for the 15 Gy animal groups, the animals that were about to receive 17.5 Gy were more sensitive to dexamethasone treatment, as was displayed by the difference in bone length at the time of radiation (Figure 3A). The same pattern of growth was observed for the 17.5 Gy treated animals and the 15 Gy treated animals. There is an 11.4 % and 47 % increase in growth rate at one week and two weeks post radiation respectively (Figure 3B).

Discussion In this study, we present different feasible models for the study of linear growth in the long bones, both in vitro and in vivo. The aim of the study was to investigate whether treatment with glucocorticoid drugs could protect the growth plate from the damaging effects of radiation. We hypothesized that the temporary administration of glucocorticoid drugs would decrease chondrocyte proliferation which could lead to protection of the growth plate from damaging radiotherapy. To test our hypothesis, we used two different research models. The first research model used was the well described rat metatarsal bone culture system and the second model used was an in vivo mouse model.

16|Leiden BioMedical Journal |#5 | June 2014

Moreover, there seems to be a small protective effect of dexamethasone treatment two weeks after radiation, with a two-and-a-half fold increase in growth rate for the treated animals versus the controls.

Nominatie Biomedische wetenschappen Preliminary results from the in vitro model indicated that treatment with dexamethasone decreased longitudinal growth rate and chondrocyte proliferation significantly. Continuous administration of 1 ÂľM of dexamethasone in the culture medium significantly decreased bone growth. Increasing the dose did not improve the inhibitory effect on growth or proliferation of the metatarsal. In addition to the positive preliminary findings on dexamethasone, irradiating the metatarsal bones while in culture proved to diminish growth rate and proliferation to a certain extent. A single dose of 12 Gy was sufficient to significantly decrease growth rate and proliferation rate. However, the amount of radiation used did not seem to completely obliterate growth. These results confirmed the choice of the metatarsal bone culture system as a feasible system to study growth and proliferation of the long bones in vitro.

was 40 Âľg dexamethasone. Chondrocyte proliferation was halved after 5 days of dexamethasone injections.

Preliminary results from radiation exposure show that irradiating the hind leg with a single dose of 17.5 Gy diminished growth potential in both the dexamethasone treated and non-treated animals one week after radiation exposure. Two weeks after radiation exposure, the dexamethasone treated animals show a slightly increased growth rate in both the radiated and non-radiated animals. However, there does not seem to be fast catch-up growth, indicating that 17.5 Gy might be too high a dose for the growth plate to recover. Moreover, treatment with dexamethasone prior to a single dose of 15 Gy to the hind leg increased the growth rate and therefore proliferative potential slightly in the animals treated with dexamethasone. However, this increase in growth rate was insufficient to reach similar length as control animals. In the non-radiated animals, the growth rate One problem with the in vitro model is that the of the dexamethasone treated animals remained lower withdrawal of dexamethasone from the culture than controls. . This might indicate there is a small medium led to a marked rebound in growth, even in protective effect of dexamethasone, but this effect the absence of radiation, which complicated the data not strong enough to reverse the damaging effects of interpretation. This phenomenon, which is called radiation. catch-up growth, has been described in many different research models after glucocorticoid withdrawal To enhance the protective effects of dexamethasone, (21-23). We first hypothesized that removing the a shorter time of exposure to a higher dose of medium from the culture plate did not remove all the dexamethasone might decrease proliferative capacity dexamethasone present, so that adding regular culture to a higher extent and therefore might offer more medium would dilute the high-dose dexamethasone protection against irradiation damage. Moreover, the to a low-dose of residual dexamethasone. It has been time of follow-up seems marginal, since mice can shown that in contrast to the inhibitory effect of live much longer than the 8 weeks described to reach high-dose dexamethasone, low-dose dexamethasone adulthood. Following growth over a longer period increases growth rates in metatarsal bone culture. of time might increase our knowledge on the long Preliminary results on this hypothesis show that term effects of dexamethasone on radiation damage. even after a longer culture time the metatarsal bones The doses of total body irradiation used for leukemia previously treated with high-dose dexamethasone or hematopoietic stem cell transplantation protocol have an increased growth rate for several days after usually range from 5 to 12 Gy in total. The doses used the withdrawal. in our experiments greatly exceed those doses. A higher dose and longer follow-up time, using a lower Because of the limitations of the in vitro model, dose of radiation might also increase the protective we decided to test our primary hypothesis on the effect of dexamethasone. protective effect of dexamethasone in an in vivo model. Using mice shortly after the wean date allowed Previous studies have tried other approaches to for follow-up from a young to an adult age. At the protect the growth plate from radiation Chemical start of dexamethasone treatment, the mice were radioprotectants such as pentoxifylline,amifostine and approximately 21 days old. At the time of radiation, misoprostol prevent growth failure to a variable extent the juvenile mice were 4 weeks old and were followed (24, 25). However, these drugs did not ameliorate for 4 more weeks until they reached the adult age of 8 all the damaging effects of radiation. Furthermore, weeks. The dose of dexamethasone that had a similar radioprotectants protect both malignant cells and effect on the proliferation rate in vivo as the effect of 1 non-malignant cells. These results combined might ÂľM of dexamethasone in the metatarsal culture model imply that there is protective potential of all of these

June 2014|#5|Leiden BioMedical Journal |17

Iris Pelsma drugs, but that the exact mechanism and dose needed dexamethasone could be used as a radioprotective to fully protect the growth plate from radiation is not drug. known.

The use of growth hormone antagonists might be similarly useful. By antagonizing the effects of growth hormone, it is believed that antagonists like pegvisomant inhibit proliferation in both the resting and proliferative zone of the growth plate.

Therefore, future research regarding protection against radiotherapy should include the use of growth hormone antagonists.

Moreover, radiation is not the only method to destroy active proliferating cells. Chemotherapeutic drugs which disrupt normal cell division may manifest their effects on the growth plate as either a reduction in cell number or the loss of functional integrity of extracellular matrix (26). Recent studies have shown the similar effects of glucocorticoids and their protection against chemotherapy induced growth failure (23).

If future results from the animal model or metatarsal culture show that treatment with dexamethasone or another protective agent show that there is a significant benefit of using those prior to radiation, it might change the way radiation is performed in the pediatric clinic.

In summary, radiotherapy given to pediatric patients can severely impact multiple aspects of their health.

Side effects include significant growth deformities if the active growth plate is included in the field of radiation.

These severe side effects can be diminished in several ways. Here we present a model to study the protective effect of pre-treatment with glucocorticoids prior to radiation. This study shows that the growth plate related actions of glucocorticoids like dexamethasone enable it to temporarily decrease linear growth and chondrocyte proliferation both in vitro and in vivo.

This decrease in growth rate and proliferation is reversed when dexamethasone treatment is ended.

Unfortunately, interpretation of the in vitro data is complicated by the presence of catch-up growth after dexamethasone treatment, even in the absence of radiation. The in vivo model appears more promising.

Our in vivo data suggest that there may be a small protective effect of dexamethasone treatment prior to high-dose radiation, although the results are not complete. In future experiments, we will attempt to optimize the conditions to determine whether 18|Leiden BioMedical Journal |#5 | June 2014

References (1) Pawlik TM, Keyomarsi K. Role of cell cycle in mediating sensitivity to radiotherapy. Int J Radiat Oncol Biol Phys 2004;59:928-42.

(2) Sklar C, Boulad F, Small T, Kernan N. Endocrine complications of pediatric stem cell transplantation. Front Biosci 2001;6:G17-G22.

(3) Armenian SH, Sun CL, Kawashima T, Arora M, Leisenring W, Sklar CA, et al. Long-term healthrelated outcomes in survivors of childhood cancer treated with HSCT versus conventional therapy: a report from the Bone Marrow Transplant Survivor Study (BMTSS) and Childhood Cancer Survivor Study (CCSS). Blood 2011;118:1413-20.

(4) Fletcher BD, Crom DB, Krance RA, Kun LE. Radiation-induced bone abnormalities after bone marrow transplantation for childhood leukemia.

Radiology 1994;191:231-5.

(5) Brauner R, Fontoura M, Zucker JM, Devergie A, Souberbielle JC, Prevot-Saucet C, et al. Growth and growth hormone secretion after bone marrow transplantation. Arch Dis Child 1993;68:458-63.

(6) Chow EJ, Friedman DL, Yasui Y, Whitton JA, Stovall M, Robison LL, et al. Decreased adult height in survivors of childhood acute lymphoblastic leukemia: a report from the Childhood Cancer Survivor Study. J Pediatr 2007;150:370-5, 375.

(7) Cohen A, Rovelli A, Bakker B, Uderzo C, van Lint MT, Esperou H, et al. Final height of patients who underwent bone marrow transplantation for hematological disorders during childhood: a study by the Working Party for Late Effects-EBMT. Blood 1999;93:4109-15.

(8) Hovi L, Saarinen-Pihkala UM, Vettenranta K, Lipsanen M, Tapanainen P. Growth in children with poor-risk neuroblastoma after regimens with or without total body irradiation in preparation for autologous bone marrow transplantation. Bone Marrow Transplant 1999;24:1131-6.

(9) Huma Z, Boulad F, Black P, Heller G, Sklar C. Growth in children after bone marrow transplantation for acute leukemia. Blood 1995;86:819-24.

(10) Robison LL, Nesbit ME, Jr., Sather HN, Meadows AT, Ortega JA, Hammond GD. Height of children successfully treated for acute lymphoblastic leukemia: a report from the Late Effects Study Committee of Childrens Cancer Study Group. Med Pediatr Oncol 1985;13:14-21.

Nominatie Biomedische wetenschappen (11) Shalet SM, Brennan BM. Growth and growth hormone status following treatment for childhood leukaemia. Horm Res 1998;50:1-10.

(12) Vilela MI, Viana MB. Longitudinal growth and risk factors for growth deficiency in children treated for acute lymphoblastic leukemia. Pediatr Blood Cancer 2007;48:86-92.

(13) Spadaro JA, Horton JA, Margulies BS, Luther J, Strauss JA, Farnum CE, et al. Radioprotectant combinations spare radiation-induced damage to the physis more than fractionation alone. Int J Radiat Biol 2005;81:759-65.

(14) Kember NF. Cell population kinetics of bone growth: the first ten years of autoradiographic studies with tritiated thymidine. Clin Orthop Relat Res 1971;76:213-30.

(15) Abad V, Meyers JL, Weise M, Gafni RI, Barnes KM, Nilsson O, et al. The role of the resting zone in growth plate chondrogenesis. Endocrinology 2002;143:1851-7.

(16) Nilsson O, Marino R, De LF, Phillip M, Baron J. Endocrine regulation of the growth plate. Horm Res 2005;64:157-65.

(17) Jux C, Leiber K, Hugel U, Blum W, Ohlsson C, Klaus G, et al. Dexamethasone impairs growth hormone (GH)-stimulated growth by suppression of local insulin-like growth factor (IGF)-I production and expression of GH- and IGF-I-receptor in cultured rat chondrocytes. Endocrinology 1998;139:3296-305.

(18) Schrier L, Ferns SP, Barnes KM, Emons JA, Newman EI, Nilsson O, et al. Depletion of resting zone chondrocytes during growth plate senescence. J Endocrinol 2006;189:27-36.

(19) Baron J, Huang Z, Oerter KE, Bacher JD, Cutler GB, Jr. Dexamethasone acts locally to inhibit longitudinal bone growth in rabbits. Am J Physiol 1992;263:E489-E492.

(20) Gafni RI, Weise M, Robrecht DT, Meyers JL, Barnes KM, De-Levi S, et al. Catch-up growth is associated with delayed senescence of the growth plate in rabbits. Pediatr Res 2001;50:618-23.

(21) Chagin AS, Karimian E, Sundstrom K, Eriksson E, Savendahl L. Catch-up growth after dexamethasone withdrawal occurs in cultured postnatal rat metatarsal bones. J Endocrinol 2010;204:21-9.

(22) Boersma B, Wit JM. Restricted catch-up growth after cessation of steroid treatment in a growthhormone-deficient child. Acta Paediatr 1996;85:9981002.

(23) Robson H, Anderson E, Eden O, Isaksson O, Shalet S. Glucocorticoid pretreatment reduces the cytotoxic effects of a variety of DNA-damaging agents on rat tibial growth-plate chondrocytes in vitro.

Cancer Chemother Pharmacol 1998;42:171-6.

(24) Spadaro JA, Baesl MT, Conta AC, Margulies BM, Damron TA. Effects of irradiation on the appositional and longitudinal growth of the tibia and fibula of the rat with and without radioprotectant. J Pediatr Orthop 2003;23:35-40.

(25) Spadaro JA, Horton JA, Margulies BS, Luther J, Strauss JA, Farnum CE, et al. Radioprotectant combinations spare radiation-induced damage to the physis more than fractionation alone. Int J Radiat Biol 2005;81:759-65.

(26) Siebler T, Shalet SM, Robson H. Effects of chemotherapy on bone metabolism and skeletal growth. Horm Res 2002;58 Suppl 1:80-5.

June 2014|#5|Leiden BioMedical Journal |19

Leonard Pelgrom

Cannabinoid type 1 receptor blockade diminishes dyslipidemia and obesity via peripheral activation of brown adipose tissue Leonard R. Pelgrom1,2, MariĂŤtte R. Boon1,2, Patrick C.N. Rensen1,2 Department of Endocrinology and Metabolic Diseases and 2Einthoven Laboratory for Experimental Vascular Medicine, Leiden University Medical Center, Leiden, The Netherlands 1

Abstract The endocannabinoid system is an important player in energy balance by regulating hunger, reward and energy metabolism. Chronic cannabinoid type 1 receptor (CB1R) blockade with the inverse agonist Rimonabant leads to reduction of lipid levels and long-term maintained weight loss in obese rodents and humans. However, the exact mechanism has not yet been elucidated. This study shows that strictly peripheral CB1R blockade in a mouse model for human-like lipoprotein metabolism reversed dyslipidemia and diet-induced obesity by increasing clearance of very low-density lipoprotein (VLDL)-derived triglycerides (TG) from the blood by brown adipose tissue (BAT), which dissolves large amounts of TG by combustion into heat. Our data suggests that blockade of peripheral CB1R in BAT can be a very promising therapy to combat obesity and cardiovascular risk.

Introduction The endocannabinoid sytem (ECS) is an important regulator of energy balance and consists of G-protein coupled cannabinoid receptors, endogenous lipid ligands (endocannabinoids) and the enzymes involved in the production and breakdown of these ligands.

The cannabinoid type 1 receptor (CB1R) is highly expressed throughout brain regions that control hunger and reward, as well as in peripheral organs that are critical for energy metabolism. In contrast to the CB1R, the cannabinoid type 2 receptor (CB2R) is expressed almost exclusively by immune cells and is involved in immune function1.

CB1R knockout mice are lean and resistant to dietinduced obesity (DIO)2, while the ECS is overactivated in obese humans3. Chronic CB1R blockade with the inverse agonist Rimonabant leads to long-term maintained weight loss and reduction of lipid levels in obese rodents4 and humans5. Rimonabant was considered one of the most promising therapeutic drugs to treat obesity. However, the appearance of psychiatric side effects, which may result from its action in the brain, resulted in its removal from the market. Nevertheless, several lines of evidence indicate that psychiatric side effects can be avoided, while retaining the beneficial anti-obesity and lipid 20|Leiden BioMedical Journal |#5 | June 2014

lowering effects of CB1R blockade. Functional CB1R receptors are present on liver cells6, skeletal muscle cells7 and fat cells in culture4. More specifically, Tam et al.8 recently showed that the strictly peripheral CB1R antagonist AM6545 induces weight loss and diminishes liver fattening in a mouse model.

Despite clear evidence that pharmacological CB1R antagonism improves dyslipidemia, the exact mechanisms and the peripheral tissues involved have not yet been elucidated. Recently, brown adipose tissue (BAT) has been shown to control the clearance of triglycerides (TG) from TG-rich lipoproteins in the blood9. In contrast to white adipose tissue (WAT), which stores excess TG as fat, BAT dissipates energy into heat, a process mediated by the mitochondrial uncoupling protein-1 (UCP-1)10. Metabolically active BAT stores exist in adult humans, and BAT volume and activity are lower in obese subjects11. In addition, BAT volume and activity are lower in South Asians, a population that is disposed to the development of type 2 diabetes mellitus and cardiovascular diseases12.

Together, these findings have increased interest in the therapeutic potential of BAT to combat obesity and related disorders, such as dyslipidemia.

Nominatie Biomedische wetenschappen The aim of this study was to elucidate the mechanism by which peripheral CB1R blockade attenuates dyslipidemia in diet-induced obesity by using a mouse model for human-like lipoprotein metabolism.

Methods Mice and experimental design Homozygous human cholesteryl ester transfer protein (CETP) transgenic mice were crossbred with hemizygous APOE*3-Leiden (E3L) mice to obtain E3L.CETP mice, as described previously13. Ten weekold E3L.CETP male mice were housed under standard conditions with ad libitum access to water and highfat diet (HFD; Research Diets, 60% lard fat) for 12 weeks to induce obesity. From the 7th week onwards, the water was supplied with 10% w/v fructose. After 12 weeks of HFD, diet-induced obese (DIO) mice received 60% HFD with or without 10 mg/kg body weight/day (0.00885%, w/w) AM6545 (SigmaAldrich, St. Louis, USA) for 4 weeks.

All mouse experiments were performed in accordance with the Institute for Laboratory Animal Research Guide for the Care and Use of Laboratory Animals and have received approval from the Departmental Ethical Review Board (Dept. Endocrinology, Leiden University Medical Center, Leiden, The Netherlands).

In vivo clearance of labelled VLDL-like emulsion particles VLDL-like emulsion particles (80 nm) labeled with glycerol tri[3H]oleate [triolein (TO)] were prepared and characterized as described previously14. The mice were fasted for 4 h (from 9.00 to 13.00 h) and injected (t = 0) via the tail vein with 200 μL of emulsion particles (1 mg TG per mouse). Blood samples were drawn from the tail vein at 2, 10, 20 and 30 min after injection to determine serum decay of [3H]TO as described previously15. After taking the last blood sample, mice were terminated by cervical dislocation.

Subsequently, the organs were collected, dissolved overnight at 60°C in Tissue Solubilizer (Amersham Biosciences, Rosendaal, The Netherlands) and uptake of [3H]TO-derived activity was quantified and expressed per gram wet tissue weight.

RNA isolation and qPCR RNA was isolated from snap-frozen tissue samples using the Nucleospin® RNA II kit from MachereyNagel (Bioké, Leiden, The Netherlands). RNA was reverse transcribed with the iScript™ cDNA Synthesis Kit (Bio-Rad, Veenendaal, The Netherlands) and purified using the Nucleospin® Extract II kit (Macherey-Nagel). qPCR was performed with the SensiMix SYBR Hi-Rox Kit (GC Biotech, Alphen aan de Rijn, The Netherlands) on an iQ5 Single-Color real-time PCR detection system (Bio-Rad). mRNA expression was normalized for 36b4 and beta-2 microglobulin (B2M) mRNA content and expressed as fold change compared to control mice using the ∆∆CT method.

Body weight and body composition Body weight was measured twice a week. After 4 weeks of treatment, body composition was measured using the Norland pDEXA Sabre X-ray Bone Densitometer.

Mice were anaesthetized intraperitoneally (ip) with a combination of 6.25 mg⁄kg acepromazine (Alfasan), 6.25 mg⁄kg midazolam (Roche) and 0.31 mg⁄kg fentanyl (Janssen-Cilag). The heads were excluded Statistical analysis from the analyses.

All data are expressed as mean ± SEM. Statistical analysis using the two-tailed unpaired Student’s test Lipoprotein profiles and plasma parameters was performed with the SPSS 20.0 software package Blood was collected from the tail vein of 4-6 hour fasted for Windows (SPSS, Chicago, United States). A mice into chilled capillaries that were coated with P-value <0.05 was considered statistically significant. paraoxon (Sigma-Aldrich). The distribution of lipids over plasma lipoprotein fractions was determined Results using FPLC. Plasma was pooled per group and 50 To investigate the effect of peripheral CB1R blockade on µL of each pool was injected onto an ÄKTA systems energy and TG metabolism, diet-induced obese E3L. Superose 6 column (Amersham Pharmacia Biotech, CETP mice were treated with the strictly peripheral Piscataway, NJ, USA) and eluted at a constant flow CB1R antagonist AM6545 for 4 weeks. AM6545 rate of 50 µL/min with PBS, pH 7.4. TC and TG were markedly reduced body weight (-19%, p<0.001; Figure analyzed using commercially available enzymatic kits 1A) and fat mass (-23%, p<0.001; Figure 1B), without from Roche Diagnostics (Mannheim, Germany). affecting lean mass. AM6545 did not affect total caloric intake or locomotor activity (data not shown). Instead, AM6545 increased total energy expenditure (data not shown), which was the result of both increased

June 2014|#5|Leiden BioMedical Journal |21

Leonard Pelgrom fat oxidation and carbohydrate oxidation (data not shown). Furthermore, AM6545 significantly lowered plasma TG (-49%, p<0.05; Figure 1C) and TC (-35%, p<0.05; data not shown). This lipid-lowering effect of AM6545 mostly resulted from a reduction in TG-rich very low-density lipoprotein (VLDL) particles (AUC -73%; Figure 1D).

To determine which organs had contributed to the lipid lowering effect of AM6545, mice were injected via the tail vein with VLDL-like emulsion particles labeled with [3H]TO. AM6545 accelerated the clearance of VLDL-like emulsion particles from the plasma (Figure 2A) by greatly increasing TG-derived fatty acid uptake by BAT (+235%, p<0.01; Figure 2B). Increased expression of lipoprotein lipase (Lpl;

Figure 1- The effect of AM6545 on body weight and plasma TG levels in diet-induced obese E3L.CETP transgenic mice.

Mice received AM6545 (at 10 mg/kg/day orally) for 4 weeks. Body weight was measured twice a week using a digital scale (A), fat and lean mass were measured at the end of the treatment period using a DEXA scanner (B). Plasma triglyceride (TG) levels were analyzed using a commercially available enzymatic kit (C) and the distribution of TG over plasma lipoprotein fractions was determined using FPLC (D). Data represent mean Âą SEM from 8 mice per group. *p < 0.05 and ***p < 0.001 vs. controls.

22|Leiden BioMedical Journal |#5 | June 2014

Nominatie Biomedische wetenschappen

Figure 2 - The effect of AM6545 on the kinetics of glycerol tri[3H]oleate-labeled VLDL-like particles and gene expression in BAT of diet-induced obese E3L. CETP transgenic mice.

Mice received AM6545 (at 10 mg/kg/day orally) for 4 weeks. At the end of the treatment period, the mice were fasted for 4 hours and then injected with VLDL-like TG-rich particles labeled with glycerol tri[3H]oleate. Plasma clearance rate of 3H-activity was measured by extrapolating 3H-activity from blood samples that were drawn at t = 2, 10, 20 and 30 min after injection (A). The uptake of 3H-activity by organs was quantified by dissolving the organs overnight after the last blood sample was taken (B). Expression of Ucp1, Lpl and Cd36 in snap-frozen BAT were measured by qPCR (C). Data represent mean Âą SEM from 8 mice per group. *p < 0.05 and ***p < 0.01 vs. controls.

+67%, p<0.05), fatty acid transporter (Fat/Cd36; +59%, p<0.05) and uncoupling protein-1 (Ucp1; +32%, p<0.05; Figure 2C) suggested an increase in BAT activity. Of note, TG-derived fatty acid uptake by WAT was decreased (-51%, P<0.01; Figure 2B).

Discussion In this study, we demonstrate that the strictly peripheral CB1R antagonist AM6545 reverses the increase in body weight, fat mass and plasma VLDLTG levels that are associated with HFD-induced obesity. AM6545 selectively increased the uptake of VLDL-TG-derived fatty acids by BAT, which was accompanied by an increased expression of Lpl, Cd36, and Ucp1 in BAT.

whole-body CB1 receptor antagonist Rimonabant (Boon et al., unpublished). The somewhat greater efficiency of Rimonabant over AM6545 in reducing body weight is likely explained by the fact that AM6545, in contrast to Rimonabant, did not affect total caloric intake over the treatment period, as has been shown before8.

Previous studies suggested CB1R blockade in the brain as the most important mechanism in the induction of BAT activation by whole-body CB1R blockade with Rimonabant, due to increased sympathetic outflow from the brain towards BAT16, 17. Bajzer and colleagues17 reported that denervation of BAT in mice blunted the effect of Rimonabant on insulin-mediated glucose uptake by BAT. Nevertheless, denervation did not blunt the increase in energy expenditure, Interestingly, the effects of strictly peripheral CB1 and decrease in body weight and fat mass induced receptor antagonism on lipid metabolism were almost by Rimonabant, suggesting that a peripheral CB1R as pronounced as we have previously shown with the blockade is at least in part involved in the beneficial effects.

June 2014|#5|Leiden BioMedical Journal |23

Leonard Pelgrom Our study shows that CB1R blockade diminishes dyslipidemia by inducing BAT-mediated VLDLTG uptake and BAT thermogenesis via a peripheral mode of action. Our data suggest that blockade of peripheral CB1R in BAT can be a very promising therapy to combat obesity and cardiovascular risk without inducing psychiatric side effects.

24|Leiden BioMedical Journal |#5 | June 2014

Nominatie Biomedische wetenschappen References 1.

Pagotto U, Marsicano G, Cota D, Lutz B, Pasquali R. The emerging role of the endocannabinoid system in endocrine regulation and energy balance.

Endocr Rev 2006;27(1):73-100.

2.

Ravinet TC, Delgorge C, Menet C, Arnone M, Soubrie P. CB1 cannabinoid receptor knockout in mice leads to leanness, resistance to diet-induced obesity and enhanced leptin sensitivity. Int J Obes Relat Metab Disord 2004;28(4):640-648.

3.

Cote M, Matias I, Lemieux I et al. Circulating endocannabinoid levels, abdominal adiposity and related cardiometabolic risk factors in obese men. Int J Obes (Lond) 2007;31(4):692-699.

4.

Cota D, Marsicano G, Tschop M et al. The endogenous cannabinoid system affects energy balance via central orexigenic drive and peripheral lipogenesis. J Clin Invest 2003;112(3):423-431.

5. Van Gaal LF, Rissanen AM, Scheen AJ, Ziegler O, Rossner S. Effects of the cannabinoid-1 receptor blocker rimonabant on weight reduction and cardiovascular risk factors in overweight patients: 1-year experience from the RIO-Europe study. Lancet 2005;365(9468):1389-1397.

6.

Osei-Hyiaman D, DePetrillo M, Pacher P et al.

Endocannabinoid activation at hepatic CB1 receptors stimulates fatty acid synthesis and contributes to dietinduced obesity. J Clin Invest 2005;115(5):1298-1305.

7.

Eckardt K, Sell H, Taube A et al. Cannabinoid type 1 receptors in human skeletal muscle cells participate in the negative crosstalk between fat and muscle. Diabetologia 2009;52(4):664-674.

8.

Tam J, Vemuri VK, Liu J et al. Peripheral CB1 cannabinoid receptor blockade improves cardiometabolic risk in mouse models of obesity. J Clin Invest 2010;120(8):2953-2966.

9.

Bartelt A, Bruns OT, Reimer R et al. Brown adipose tissue activity controls triglyceride clearance.

Nat Med 2011;17(2):200-205.

10. Cannon B, Nedergaard J. Brown adipose tissue: function and physiological significance. Physiol Rev 2004;84(1):277-359.

11. van Marken Lichtenbelt WD, Vanhommerig JW, Smulders NM et al. Cold-activated brown adipose tissue in healthy men. N Engl J Med 2009;360(15):15001508.

12. Bakker LE, Boon MR, van der Linden RA et al. Brown adipose tissue volume in healthy lean south Asian adults compared with white Caucasians: a prospective, case-controlled observational study.

Lancet Diabetes Endocrinol 2014;2(3):210-217.

13. Westerterp M, van der Hoogt CC, de HW et al. Cholesteryl ester transfer protein decreases high-density lipoprotein and severely aggravates atherosclerosis in APOE*3-Leiden mice. Arterioscler Thromb Vasc Biol 2006;26(11):2552-2559.

14. Rensen PC, van Dijk MC, Havenaar EC, Bijsterbosch MK, Kruijt JK, van Berkel TJ. Selective liver targeting of antivirals by recombinant chylomicrons-a new therapeutic approach to hepatitis B. Nat Med 1995;1(3):221-225.

15. Jong MC, Rensen PC, Dahlmans VE, van der Boom H, van Berkel TJ, Havekes LM. Apolipoprotein C-III deficiency accelerates triglyceride hydrolysis by lipoprotein lipase in wild-type and apoE knockout mice. J Lipid Res 2001;42(10):1578-1585.

16. Verty AN, Allen AM, Oldfield BJ. The effects of rimonabant on brown adipose tissue in rat: implications for energy expenditure. Obesity (Silver Spring) 2009;17(2):254-261.

17. Bajzer M, Olivieri M, Haas MK et al.

Cannabinoid receptor 1 (CB1) antagonism enhances glucose utilisation and activates brown adipose tissue in diet-induced obese mice. Diabetologia 2011;54(12):3121-3131.

June 2014|#5|Leiden BioMedical Journal |25

Introductie genomineerden GNK Naam: Larissa Brezden Leeftijd: 34 Startjaar studie: 2007 Titel verslag: Radiocarbon analysis of human tooth enamel to estimate the year of birth of unknown decedents Begeleider: drs. Reza Gerretsen, M.D., forensisch antropoloog, Nederlands Forensisch Instituut Kun je in het kort vertellen waar je onderzoek naar hebt gedaan?

Medische forensische antropologen bepalen de identiteit van slachtoffers veelal door middel van een morfologische analyse van restmateriaal , bijvoorbeeld skeletdelen of tanden. Hiermee worden vooral leeftijd en geslacht van het slachtoffer geschat. Deze methode is echter verre van nauwkeurig, er moet rekening gehouden worden met marges van ± 10 jaar. Wij onderzochten een nieuwe methode om het geboortejaar van slachtoffers te bepalen gebaseerd op koolstof-14 isotoop-analyse van tandglazuur.

Wat vond je het leukste aan het doen van onderzoek?

Het samenspel tussen zelfstandig en creatief denken en het samenwerken met anderen: het leren van en werken met experts van allerlei disciplines - medici, statistici, internationale collega’s, technici - in alle stappen van het onderzoek, van literatuuronderzoek tot resultaatverwerking.

Waarom denk je dat je genomineerd bent voor de Student Research Award?

Ik werd gaandeweg zo gegrepen door het onderzoek dat ik er uiteindelijk veel meer tijd en energie in gestoken heb dan in eerste opzet gepland was. Ik denk dat deze passie en hard werk tot uiting zijn gekomen in het uiteindelijke resultaat.

Stel je wint de prijs, waar zou je het prijzengeld voor willen gebruiken?

Graag zou ik eens willen deelnemen aan het American Academy of Forensic Sciences (AAFS) congres.

Dit multidisciplinair congres vindt 1 keer per jaar plaats in de V.S. en sluit heel goed aan mijn toekomstige carrièredoel.

Welke richting wil je later op?

Ik begin met de huisartsopleiding in september a.s., maar wellicht in de toekomst wil ik mijn carrière uitbreiden naar de forensisch geneeskunde.

Naam: Thijs de Vos Leeftijd: 24 jaar Startjaar studie: 2008 Titel verslag: Fibroblasts promote epithelial-to-mesenchymal transition in colorectal cancer by producing BMP-2.

Begeleider:Philip Voorneveld

Kun je in het kort vertellen waar je onderzoek naar hebt gedaan?

BoneMorphogenicProteins (BMP’s) zijn eiwitten aanwezig in de darmwand en met name actief in de top van de villus waar ze apoptose induceren. Bij colorectaal carcinomen kan dit mechanisme door een mutatie ontregelt zijn waardoor BMP’s hun tumorsupressieve rol verliezen en juist metastasering bespoedigen. Waar deze BMP’s vandaan komen is nog niet precies bekend. Onze hypothese is dat BMP’s geproduceerd worden door fibroblasten rondom de tumor en deze meer BMP’s produceren onder invloed van tumorweefsel. Op deze manier kunnen fibroblasten het proces van metastasering bij colorectaal carcinomen versnellen.

26|Leiden BioMedical Journal |#5 | June 2014

Wat vond je het leukste aan het doen van onderzoek?

Het bedenken van nieuwe experimenten om je hypothese te toetsen. Op het moment dat ik de eerste resultaten van mijn eigen bedachte proef in handen kreeg werd ik extra enthousiast.

Waar liep je in je onderzoek tegenaan?

Bij het doen van laboratoriumonderzoek heb je een technisch aspect wat soms tegenvalt waardoor een experiment kan mislukken. Het is leuk om na te denken over wat er mis gegaan kan zijn tijdens de proef zodat deze volgende keer niet mislukt, maar als je proeven vaker achtereen mislukken en zonder dat je weet waarom kan dat frustrerend zijn.

Wat is de opvallendste bevinding die je hebt gedaan tijdens je onderzoek?

Bindweefselcellen maken de BMPâ&#x20AC;&#x2122;s als ze in aanraking komen met factoren die de tumor uitgescheiden heeft.

In reactie op deze BMPâ&#x20AC;&#x2122;s wordt de tumor invasief en zal sneller metastaseren.

Stel je wint de prijs, waar zou je het prijzengeld voor willen gebruiken?

Ik zou het geld willen besteden aan een buitenlandsstage, klinisch of juist een extra wetenschappelijke stage in Amerika of in Londen.

Welke richting wil je later op?

Ik vind Maag-, Darm-, Leverziekten een heel interessant vakgebied.

Heb je nog tips voor studenten die onderzoek willen gaan doen?

Denk eerder aan een wetenschapsstage in het laboratorium! Er zijn veel geneeskundestudenten die hun wetenschapsstage gebruiken om achter een computer te kruipen en veel databasewerk te verzetten, ik vond het heel leuk om zelf experimenten te kunnen doen en aan de slag te gaan in het laboratorium.

Naam: Tjeerd Muurling Leeftijd: 25 jaar Startjaar studie: 2007 Titel verslag: Metabolomic and network analysis of pharmacotherapies for sensorineural hearing loss Begeleider: Konstantina M. Stankovic, M.D., Ph.D. Assistant Professor of Otology and Laryngology, Harvard Medical School Kun je in het kort vertellen waar je onderzoek naar hebt gedaan?

Ons onderzoek richtte zich op potentieel belangrijke nieuwe targets voor de behandeling van sensorineuraal gehoorverlies (SNHL). We zijn op zoek gegaan naar interacties tussen 70 medicijnen tegen SNHL en bekende moleculaire mechanismen. Uit onze analyses kwamen enkele zeer significante netwerken en pathways. Binnen deze netwerken worden de moleculen met de meeste connecties beschouwd als een knooppunt. Het zijn deze knooppunten binnen de moleculaire mechanismen van SNHL, die belangrijk kunnen zijn bij toekomstige behandelingen van gehoorverlies.

Wat vond je het leukste aan het doen van onderzoek?

Het leukste vond ik om met iets bezig te zijn wat nog niet eerder gedaan is en daarmee relevante nieuwe informatie te verkrijgen met het oog op de behandeling van gehoorverlies. Daarnaast vond ik het absoluut een voorrecht onderzoek te kunnen doen aan het Massachusetts Eye and Ear Infirmary. De zeer motiverende omgeving en de vele leuke dingen die er in Boston te doen waren, maakten dat ik een onvergetelijke tijd heb gehad gedurende mijn wetenschapsstage.

Waarom denk je dat je genomineerd bent voor de Student Research Award?

Ik denk vanwege het vernieuwende aspect van het onderzoek, naar een veel voorkomende aandoening met nog maar weinig therapie-mogelijkheden gecombineerd met het feit dat ons onderzoek is gepubliceerd in een internationaal KNO-tijdschrift.

Heb je nog tips voor studenten die onderzoek willen gaan doen?

Voor studenten die graag in het buitenland onderzoek willen doen heb ik de tip: wees brutaal. Steek je interesse in een bepaald vakgebied niet onder stoelen of banken. Vraag bijvoorbeeld aan de professor in dat vakgebied of hij of zij wellicht nog contacten heeft in het land of de stad van voorkeur. Ze zullen bijna altijd bereid zijn je te helpen bij de eerste stappen op weg naar het realiseren van je ambitie!

June 2014|#5|Leiden BioMedical Journal |27

Larissa Brezden

Radiocarbon analysis of human tooth enamel to estimate the year of birth of unknown decedents L.M. Brezden MD, R.R.R Gerretsen MD, B. Brezden PhD, M. van Wijk MSc, Prof. Dr. J. van der Plicht PhD Netherlands Forensic Institute, The Hague, The Netherlands Abstract Biological age (date of birth) estimation of unidentified human remains, largely undertaken by forensic anthropologists, is a significant part of the forensic investigation of a crime scene or mass disaster where the identity of a victim may not be evident. Traditional morphological methods of age estimation yield a relatively wide age margin of ±10 years for adult remains. To refine the precision of this range, a novel technique based on radiocarbon (14C) analysis of human biological materials is implemented in this study. This technique exploits the phenomenon of the “bomb curve”, a double-normal atmospheric concentration of 14C due to nuclear bomb testing during the mid-20th century. This technique exploits the inherent properties of human dental enamel, which does not exhibit carbon turnover once it is formed during childhood. The radiocarbon that has become “locked” within enamel during its formation may thus be measured and compared to bomb-curve atmospheric values, producing a narrow range of carbon incorporation time (“date of birth” of the enamel). In this study, 26 isolated enamel samples from adults of known date of birth (between 1964 and 1997) were measured using 14C analysis and compared to known bomb-curve atmospheric concentrations in order to evaluate the precision of this method. Analysis of the accuracy of prediction using this method, demonstrated by the standard error of the estimate, resulted in ±1.363 years. The estimated date of birth of the individual revealed an absolute error from the actual date of birth of 1.749 ± 1.868 (mean, SD). In conclusion, radiocarbon analysis of tooth enamel delivers an accurate estimate of an individual’s date of birth, and may be a useful tool in victim identification in homicide or mass disaster cases in the forthcoming decades as bomb pulse radiocarbon continues to be measured in the atmosphere and incorporated into tooth enamel.

Introduction The identification of human remains is an essential aspect of forensic investigation in a crime scene, accident, or mass disaster where the identity of the deceased may not be evident. Victim identification often involves ascertaining the decedent’s gender and estimating the age at the time of death using current methods of analyzing age- and gender-related differences in skeletal remains. An estimation of the victim’s date of birth (DOB) may be particularly useful in matching an unidentified decedent to a list of missing persons. Current morphological age determination methods, however, typically produce estimation errors of ±10 years [1].

Teeth are of particular interest to forensic investigators in the identification of a cadaver as they are remarkably resilient to decomposition, humidity, heat, and chemical decay, and are thus the remains most commonly found in forensic contexts. Tooth enamel is the hardest substance in the body, and thus primarily accounts for the tooth’s durability. It also protects harvestable DNA, which may be extracted within the dentine, or pulp (if available), which allows for comparison with surviving family members or personal items known to belong to the victim [2]. In contrast to most other skeletal parts, tooth development 28|Leiden BioMedical Journal |#5 | June 2014

is more closely associated with chronological age, and appears to be under tighter genetic control [3].

An estimated age of an individual may be deduced from dental remains by comparing the individual’s dental developmental status with published dental development surveys, which provide information on the order of developmental events (namely deciduous and permanent tooth mineralization and eruption, which vary among different tooth elements) as well as the time frame and chronological age in which these events are believed to occur [4].

A relatively recent global phenomenon has facilitated a new method of predicting the year of birth of a victim based on accelerated mass spectrometry (AMS) analysis of radiocarbon (14C) in dental enamel. This method has demonstrated an average absolute error of 1.6±1.3 years (mean, SD) in its premier study by Spalding et. al [5], and similar results in subsequent studies by different authors [6, 17]. The premise of this method lies in the extensive detonations of above-ground nuclear weapons between the years 1955 and 1963 which have doubled the formerly relatively constant concentration of 14 C in the atmosphere. Coined the “bomb pulse,”

Nominatie Geneeskunde this increased atmospheric 14C concentration has since dispersed equally around the globe. Careful measurement of atmospheric 14C over time has been plotted onto a “bomb curve,” illustrated in Figure 1. Upon the implementation of the Partial Test Ban Treaty in 1963 (corresponding with the peak of the bomb curve), the atmospheric 14C concentration has since then gradually tapered, not due to radioactive decay, but due to diffusion and equilibration with large marine and terrestrial carbon reservoirs via the carbon cycle. Within the carbon cycle, the 14C content of new plant growth (and the animals that feed from the plants) largely matches the 14C concentration in the atmosphere. The temporal variation in the artificially high levels of radiocarbon that has been captured in organic material offers an opportunity to date the synthesis of bio-molecules, providing a unique isotopic chronometer of the last century [7].

for potential future use in forensic analysis as a standard operating procedure (SOP) in identifying unknown decedents, and to validate this method for use in the Dutch justice system for aging unknown decedents.

Materials and Methods Tooth samples Twenty-six teeth from adult individuals (aged 15-45 years) from a collection of over 200 teeth of known age (DOB) from the forensic anthropological collection at the NFI were analyzed for their carbon-14 content.

Eighteen teeth were from individuals of Dutch origin, while eight teeth had an unknown provenance. The gender proportion was fairly equal with fifteen males and eleven females. The teeth were procured from living volunteer patients of local dentists under informed consent, and have been removed for either orthodontic purposes or due to periodontal problems.

Only the type of tooth, date of birth, sex, provenance, Although most living tissues undergo continuous the procuring dentist, and date of extraction were turnover and remodeling, eye lens crystallines and noted, thus the subjects remain anonymous. Each tooth enamel in particular have the unique property tooth was given a code and stored in formalin until that their substrates (including carbon) do not express ready for use. turnover once these tissues are formed [8]. Dental enamel thus presents an ideal source of “trapped Enamel isolation carbon” and thereby an accurate representation of The crowns were horizontally separated from their the conditions under which it formed. Enamel from roots at the cervical line and mesiodistal lines using a a mature tooth contains 0.4-0.6% carbon, which diamond-blade saw. Most of the macroscopic dentine may eventually be analyzed for its radiocarbon was removed with a high-speed hand-held Dremel® concentration [5]. When this “trapped 14C” is drill and blunt dissection with a dental instrument. measured with radiocarbon analysis and compared To remove the remaining dentine, the crown was to known atmospheric levels at a given time, the date immersed in 10N NaOH in a water bath sonicator of enamel deposition (or “date of birth” of the tooth) (Branson® 2510) for 30 minutes at approximately may be deduced. Further calculation using known 69˚C. The sample was subsequently rinsed three tooth (enamel) formation times published in dental times with ddH2O, resubmerged in 10N NaOH in the development surveys [9] allows the approximation sonicator for an additional 30 min, then washed again of the date of birth of the source individual. three times with double-distilled water (ddH2O). The crown was resubmerged in 10N NaOH overnight, and The practical application of the 14C dating method is thereafter rinsed twice with ddH2O in the sonicator for additionally valuable with regards to the age group 10 minutes. This process was repeated approximately of the population being analyzed. According to the 3-5 times until all dentin, possible fillings, and soft Doe Network Database, in 2011, 885 out of 1188 cases tissues were removed from enamel. were estimated to be adults aged 20-50 at the time of death (=75%) [10]. While morphological methods Pre-treatment of enamel for AMS analysis of identification exhibit drawbacks particularly Enamel samples were pre-treated for AMS analysis within this age range, this time frame fortuitously at the Rijksuniversiteit Groningen Center for Isotope corresponds to the “bomb curve,” in that almost Research (RuG CIO). This multi-step process involves all cadavers to be analyzed in a modern context first briefly immersing the samples in 4% HCl to will have incorporated bomb-pulse radiocarbon remove surface impurities, subsequent washing in the enamel of their permanent teeth [10]. with ddH2O, followed by drying under a loose foil The goal of this study is thus to estimate the age of an tent overnight in an oven at 90˚C. The dried enamel individual via radiocarbon analysis of dental enamel samples are then placed in individual single-use

June 2014|#5|Leiden BioMedical Journal |29

Larissa Brezden reactors, and combusted in an automated CarloErba (type NC 2500) elemental analyzer (EA). Oxalic acid (Ox1) and caffeine (2009) were used as measurement standards by the EA. The EA also determined that nearly all CO2 samples contained >500μg carbon.

The evolved CO2 was purified, trapped, and reduced to graphite in the presence of an iron catalyst in the individual reactors. The graphite was then pressed into targets for subsequent analysis via the AMS. Background values were controlled by following standard laboratory practice (consistently following procedures, maintaining clean rigs and sample tubes, etc.) [11].

year given by CALIBomb is therefore considered the “estimated DOB” of the tooth. When the resultant F14C value is plotted on the y axis of the bomb curve, it will give two points, namely on the rising and falling part of the curve. This potential ambiguity is solved with logical ascertainment, as demonstrated by Spalding et. al, where two teeth with different enamel lay-down times from the same individual may be analyzed for their 14C concentration[5], or by measuring the 14C concentration of the enamel and root collagen of the same tooth[18]. A single tooth per individual in all cases was analyzed in our study for logistical purposes and for avoiding repetition of published data.

AMS analysis The graphite targets were analyzed by a fully automated, high-throughput 10 MeV HVEE 4130 14C AMS system manufactured by High Voltage Engineering Europa (HVEE), at the Centre for Isotope Research in Groningen, The Netherlands [12]. Routine precision is approximately 5‰ (‰ = per mill, equivalent to the factor 10-3 [16]) for 14C/12C and approximately 1 ‰ for 13C/12C, with routine backgrounds corresponding to the ages of 45-50 ka (ka = “thousand years,” noncalibrated 14C dates). Details on the design and protocol of the RuG AMS system may be found in the literature [12,13,14].

14 C data is reported using Fraction Modern (F14C) nomenclature, specifically developed for measuring nuclear weapons-testing (bomb pulse) 14C levels in samples, which compares the 14C content of a post-bomb sample at the time of its formation to the concentration measured in the atmosphere as a ratio of the sample activity to the standard activity measured in the same year [15]. Sample activity is compared with the activity of a reference material, or the standard activity, defined as 95% of the activity of a specific batch of Oxalic Acid nr. 1 (Ox1) [16].

The strategy for calculating an individual’s year of birth by 14C analysis of dental enamel is based on comparing the known atmospheric value of 14C on the bomb curve with the year of enamel synthesis.

The enamel formation time frames used in our study originate from reference data provided by Nolla [9], validated in the study by Alkass [6]. We used the median tooth formation time (in years) for each dental element to approximate the age of formation. This known enamel formation time (in years) is subtracted from the resultant estimated DOB of the tooth to give the estimated DOB of the individual, seen in Figure 1.

From 14C data, to estimated DOB tooth, to estimated DOB individual The accelerated mass spectrometer measures the concentration of 14C in a tooth enamel sample and reports this measurement as an F14C value. This resultant F14C value is subsequently input into the CALIBomb program, which compares analyzed data to known reference data using the Levin dataset[17].

In this analysis, the smoothing in years was set to 1.0, and a 2 sigma error was used. The analyzed (tooth) F14C value is then plotted onto the bomb curve of known atmospheric F14C values against time (in years) to determine the year of enamel synthesis. The resultant 30|Leiden BioMedical Journal |#5 | June 2014

Statistical analysis Regression analysis explores the correlation between estimated and true date of birth of the individual.

A Q-Q plot compares expected vs. observed DOBs and further determine the accuracy of the regression equation. Neurotek Analyst was implemented for statistical analyses.

Figure 1 - An arbitrary example of the analyzed tooth sample superimposed upon the bomb curve (blue line). The curved arrow represents the known enamel formation time, subtracted from the analyzed F14C value (red point), to give the estimated DOB (dotted line), which is compared with the actual DOB (solid line).

Nominatie Geneeskunde

Table 1 - 14C analysis of teeth formed during the bomb peak, 1964-1997.

a)

b)

Figure 2 - (a) Regression analysis of expected vs. observed date of birth of individuals born during the bomb peak, 1964-1997. Regression equation: f(p) = -151.026 + 1.077x. (b) Q-Q plot of residuals, demonstrating that the residuals follow a normal distribution.

Results Table 1 shows the details of the radiocarbon analysis of all cases. Although tooth enamel was analyzed, for consistency and simplicity, results are ultimately reported in DOB of the individual and not of the tooth.

The regression analysis comparing the estimated vs. actual DOB demonstrates a residual standard error of the estimate, or prediction interval, of Âą1.363 years with a 95% confidence interval (Figure 2a). An excellent correlation between the estimated and actual dates of birth was found (R2 = 0.970). The Q-Q plot shows that the residuals follow a normal distribution, indicating a reliable correlation (Figure 2b). The estimated date of birth of individuals demonstrates an absolute (average) error from the true date of birth of 1.749 Âą 1.868 (mean; SD).

Discussion For practical purposes in subjects analyzed in the future who are born between 1964-1997, a date of birth may be predicted with an accuracy of Âą1.363 years, far more accurate than traditional morphological methods of identification [2]. Whereas traditional methods of identification of an unknown decedent typically provide the age of the individual at the time of death, the date of birth is an essential piece of information in potentially matching an individual to a personal register or list of missing persons The radiocarbon dating method of tooth enamel provides this date of birth with excellent accuracy.

Figure 3 demonstrates the correlation between actual and estimated DOB of teeth formed between 19641997 superimposed upon the bomb curve. The last June 2014|#5|Leiden BioMedical Journal |31

Larissa Brezden 6 pairs of data points, namely those of individuals born after 1979, demonstrate a markedly increased deviation with an absolute average error of 4.46 ± 2.05 years, compared to the rest of the data, which demonstrates an absolute average error of 0.93 ± 0.67 years. A potential explanation for this greater spread may lie with the inherent flattening of the slope of the bomb curve, which tapers more rapidly toward zero after 1979. Buchholz supports this assertion, stating that the chronological uncertainty increases as the slope flattens to ±2-4 years in samples “born” after the year 2000[7]. The window of accuracy for radiocarbon bomb-pulse dating is thus restricted to individuals within a birth year range that remains to be further tested in order to be definitively determined.

spectrometer in reading the 14C in a given sample, may skew the results significantly. Careful measures were thus taken in every step to ensure quality of the sample and isolation of enamel.

Ultimately, the combination of the radiocarbon dating method with other available methods of age estimation may provide a (more precise) “profile” for an unknown decedent, which includes a significantly accurate estimated date of birth, and may lead to a potential profile match in a list of missing persons.

As a substantial part of missing persons in the near future will have incorporated bomb pulse 14C in their tooth enamel, radiocarbon analysis of this bomb pulse carbon may become a potentially valuable identification method for the coming decades.

The careful handling of samples in every step, from procurement to isolation of enamel to final processing for 14C analysis, is essential in maintaining the accuracy of results. Improper handling during any step of the protocol may introduce contamination which, due to the extreme sensitivity of the accelerated mass

Figure 3 - Correlated pairs of estimated and actual tooth DOBs and their corresponding F14C values along the bomb curve.

32|Leiden BioMedical Journal |#5 | June 2014

Nominatie Geneeskunde References 1. K.Alkass, B.A.Buchholz, S.Ohtani, T.Yamamoto, H.Druid, K.L.Spalding, Age estimation in forensic sciences: application of combined aspartic acid racemization and radiocarbon analysis, Mol.Cell Proteomics 9 (2010) 1022-1030 2. T. Thompson, S.Black, Forensic Human Identification: An Introduction, (2007) 3.

T.White, P.Folkens, Human Bone Manual, (2005) 4.

L.Ciapparelli, The Chronology of Dental Development and Age Assessment, (1992) 5. K.L.Spalding, B.A.Buchholz, L.E.Bergman, H.Druid, J.Frisen, Forensics: age written in teeth by nuclear bomb tests, Nature 437 (2005) 333-334 6. K.Alkass, B.A.Buchholz, H.Druid, K.L.Spalding, Analysis of 14C and 13C in teeth provides precise birth dating and clues to geographical origin, Forensic Sci Int 209 (2011) 34-41 7.

B.A.Buchholz, Carbon Bomb-Pulse Dating, (2007) 8.

N. Lynnerup, H. Kjeldsen, S. Heegard, C. Jacobsen, J. Heinemeier, Radiocarbon dating of the human eye lens crystallines reveal proteins without carbon turnover throughout life, PLoS One. 3 (2008) 9.

C.M.Nolla, The development of permanent teeth, J.Dental Child 27 (1960) 254-263

10.

K.Alkass, External & Intrinsic Signatures in Human Teeth to Assist Forensic Identification Work, (2011) 11.

A.T.Aerts-Bijma, J.van der Plicht, Automatic AMS Sample Combustion and CO2 Collection, Radiocarbon 43 (2001) 293-298 12.

A.Bayliss, G.McCormac, van der Plicht J., An illustrated guide to measuring radiocarbon from archaeological samples, Physics Education 39 (2004) 1-8 13.

A.Gottdang, P.J.van der Plicht, HVEE 14C System at Groningen, Radiocarbon 37 (1995) 649-656 14.

P.J.van der, S.Wijma, A.T.Aerts, Status report: The Groningen AMS facility, Nuclear Instruments and Methods in Physics Research 172 (2000) 58-65 15.

P.Reimer, A.Brown, R.Reimer, Discussion: Reporting and calibration of post-bomb 14C data, Radiocarbon 46 (2012) 1299-1304 16.

G.Mook, J.van der Plicht, Reporting 14C activities and concentrations, Radiocarbon 41 (1999) 227-239 17. (http://calib.qub.ac.uk/CALIBomb/frameset. html 18.

G.Cook, E.Dunbar, S.Black, S.Xu, A preliminary assessment of age at death determination using the nuclear weapons testing 14C activity of dentine and enamel, Radiocarbon 48 (2006) 305-313

June 2014|#5|Leiden BioMedical Journal |33

Thijs de Vos

Stromal BMP-2 expression is associated with poor survival in SMAD4 negative colorectal cancers Thijs de Vos, Philip Voorneveld

Abstract The tumour microenvironment plays an important role in the development and progression of cancer; the presence of tumour stroma has recently been associated with tumour aggressiveness and metastasis formation.

The mechanism behind this causality has been the subject of much debate. We show that SMAD4 negative CRC cells produce soluble factors that stimulate BMP2 expression in stromal fibroblasts and that BMP2 has prometastastic effects specifically in SMAD4 negative CRC cells. We also show that stromal BMP2 expression at the invasive front of CRC tissue is associated with a poorer patient survival only in SMAD4 negative cancers.

This would suggest that patients with SMAD4 negative CRC might benefit from inhibition of BMP signalling.

Introduction For many decades, the focus of cancer research has been on tumour cells, but recently the influence of tumour microenvironment, also known as tumour stroma, has received much attention. Colorectal cancers (CRC) surrounded by large quantities of tumour stroma (known as Stroma High cancers) have a worse prognosis compared to Stroma Low cancers.1,2 The exact mechanism of this process remains to be elucidated, but it is known is that the tumour tissue produces growth factors, which can activate the surrounding fibroblasts, inflammatory cells and endothelial cells in a paracrine way. In response, fibroblasts, inflammatory cells and endothelial cells produce proteases, growth factors and extracellular matrix components that can promote angiogenesis and malignant tumour growth.3 It has been suggested that stromal fibroblasts induce Epithelial-to-Mesenchymal Transition (EMT) in cancer cells.4 EMT is the transition of columnar shaped epithelial cells to more spindle-like cells with mesenchymal properties; this process allows tumour cells to migrate and invade the surrounding tissue.5,6Recently we found that Bone Morphogenetic Protein (BMP) signalling pathway ligands can induce EMT in SMAD4 negative CRC cells.7 We now hypothesize that stromal fibroblasts are the source of the BMP ligands and that they induce EMT via the BMP pathway.

BMP signalling is important in differentiation and 34|Leiden BioMedical Journal |#5 | June 2014

apoptosis in normal colonic tissue and is mostly active at the top of the crypt. BMP ligands initiate signalling by binding cooperatively to transmembrane serine-threonine kinase receptors types 1 and 2, triggering phosphorylation and activation of the type 1 receptor by the type 2 receptor kinase. The activated type 1 receptor phosphorylates SMADs1,5,8 (receptor-regulated Smads(R-Smads)), which allows their association with SMAD4 (common-partner Smads (Co-Smads)). This heterodimer complex then translocates to the nucleus and modulates the transcription of genes specific for the BMP pathway.8 Materials and Methods Fibroblasts treated with conditioned medium of colorectal cancer cells HT-29 human colon cancer cell line and 18-CO human normal colon fibroblasts and were cultured in Dulbecco’s Modified Eagle’s medium (DMEM), supplemented with 50U/ml penicillin, 50µg/ml streptomycin and 10% heat-inactivated fetal calf serum (FCS) (Gibco). Cells were grown in a monolayer at 37⁰C in a humidified atmosphere containing 5% CO₂. HT-29 cells were grown to a confluency of 90% and then incubated with DMEM supplemented with 1% FCS for 24 hours. After incubation the medium was centrifuged for 5 minutes at 1200 rpm. The supernatant was mixed in a 1:1 ratio with DMEM containing 10 % FCS, which resulted in a final product that contained a 5,5% FCS mixture. Fibroblasts were treated with 2 ml conditioned medium of cancer cells

Nominatie Geneeskunde and incubated overnight, in the control situation fibroblasts were treated with 2 ml DMEM containing 5,5% FCS. The next day the cells were harvested and RNA was isolated using Trizol (Invitrogen) according to manufacturer’s instructions and a TGF-β / BMP specific PCR array (SABiosciences) was performed.

Embedded tumours were sectioned and stained for SMAD 4 (mouse anti-SMAD4, 1:400, Santa Cruz Biotechnology) and BMP2 (goat-ant-BMP2, 1:200, Santa Cruz Biotechnology). Two researchers scored all slides independently and in case of discrepancies a consensus was reached.

Invasion assay CRC cell line cells were labelled with 5µM cell tracker green, CMFDA (Invitrogen) according to the manufacturer’s instructions. UV-block inserts with 8 µm pores were filled with a mixture of 50 µL DMEM and 50 µL matrigel (BD Biosciences, Bedford, UK). The inserts were placed in a 24 well plate and incubated for one hour at 37⁰C. 1,0 x 105 labelled cancer cells diluted in 100 µL DMEM (0% FCS) were seeded in the upper chamber of the transwell system.

The lower chamber was filled with 600 µL DMEM (10% FCS, no phenol red). Migration was quantified by measurement of fluorescence at the bottom of the well, every hour for 24 hours, using the Biotek FLX800 fluorescence microplate reader (Biotek, Bad Friedrichshall, Germany). Results are corrected for background fluorescence and presented as a relative increase in fluorescence. Controls were set to 1.

Statistical analysis

Inclusions (%) 64.9 (12,5)

Gender

BMP PCR array 3

*

2

1

Table 1 - Baseline characteristics

7 P

6

BM

P

5

BM

P

4

BM

P

3

BM

P

88

2

Total

BM

40 (45,5)

P

Female

P

48 (54,5)

BM

Male

1

0 BM

Mean age (SD)

Survival The results of the scoring of the immunohistochemistical stainings were processed using SPSS Statistics 20 (IBM Corporation, New York, USA). Kaplan Meier analyses and Log rank tests were used to analyse differences in survival.

Fold change

Immunohistochemical staining Formalin-fixed paraffin-embedded primary stage III colorectal cancer tissues from patients operated between1995 and 2011 were selected. Blocks were obtained from the archives of the pathology department at Leiden University Medical Center, Leiden. The patient cohort included 88 cases; patient’s characteristics are shown in Table 1.

qPCR Microsoft Excel was used to calculate the fold changes. The mean and standard deviation (SD) of three samples was calculated. The experiment was carried out in triplet. Graphs were made using GraphPad (GraphPad Prism Incorporated, Texas, USA) all figures show the mean of the samples and SEM. Unpaired 2-tailed t-test were performed as appropriate using GraphPad. A p-value ≤ 0,05 was considered significantly different.

Figure 1 - Fold change of mRNA in conditioned medium treated fibroblasts versus controls. An unpaired t-test was performed to test if the mean mRNA level was significantly different. *) p = 0,037

June 2014|#5|Leiden BioMedical Journal |35

Thijs de Vos

Figure 2 – (A) shows the invasion of HT-29 with BMP2 ligands in the upper well (100 ng/ml). (B) shows the invasion of HT29 CRC cell line, when BMP 2 was used as an attractant (100 ng/ml) in the lower well). *) Asterisk show significant differences of the mean.

Results Fibroblasts produce BMP2 when treated with conditioned medium of SMAD4 negative colorectal cancer cells We previously mentioned that BMPs can induce EMT in SMAD4 negative colorectal cancer cells, but the source of BMPs is still unclear. To evaluate whether fibroblasts produce BMP ligands when in contact with SMAD4 negative colorectal cancer cells, we treated colonic fibroblasts with conditioned medium from HT-29 colon cancer cell line, which is aSMAD4 negative cell line, and performed a mRNA TGF-β/ BMP specific PCR array. The PCR array showed an upregulation of BMP2 in fibroblasts when treated with conditioned medium of colorectal cancer cells (Figure 1).

BMP2 increases migration of HT-29 cancer cells Next, we measured the effect of BMP2 treatment on invasion of cancer cells. BMP2 ligands were used in the upper chamber as a treatment. When the HT-29 CRC cell line was treated with BMP2 ligands the cells migrated faster than the control situation without the BMP2 ligands (Figure 2A). We were also interested in whether the HT-29 CRC cells would invade towards a source of BMP2 ligands. To investigate this we used BMP2 ligands in the lower chamber as an attractant.

Interestingly cells migrated significantly faster towards the BMP2 ligands compared to the control without BMP2 ligands as attractant (Figure 2B).

Figure 3 - Plot of Kaplan-Meier survival analysis, survival of the SMAD 4 negative tumours is worse compared to SMAD 4 positive tumours, a Log Rank test was performed p = 0,037.

produce BMP2 ligands that increase invasion in the SMAD4 negative colorectal cancer cell line HT29.

To prove the clinical relevance of these experiments, we investigated the BMP2 expression in the stroma of CRC’s. We stained and scored the invasive front of 88 primary tumours from patients with stage III colorectal cancer for SMAD4 and BMP2 using immunohistochemistry. Patient characteristics are displayed in Table 1.

BMP2 expression in tumour stroma is associated with a poorer prognosis in SMAD4 negative colorectal cancer In 56% of the cases tumours were SMAD4 negative, In vitro experiments demonstrate that fibroblasts we linked SMAD4 status to survival. In accordance 36|Leiden BioMedical Journal |#5 | June 2014

Nominatie Geneeskunde to earlier research11, SMAD4 negative tumours show significant worse survival compared to the SMAD4 positive tumours (Figure 3).

normal epithelial cells14, but they can also induce EMT in SMAD4 mutant tumours via the non-canonical BMP signalling pathway7,15 Whether the BMPs function via an autocrine or a paracrine manner We observed that in 34% of the cases BMP2 is is still unknown, but one of the theories is that the expressed in stroma, surrounding the invasive front stroma surrounding the epithelial (cancer) cells is the of the tumour. We stated that the BMP signalling source of the BMPs. induces EMT only in SMAD4 negative cancers, therefore we stratified our cases for SMAD4 status and We showed that fibroblasts produce BMP2 ligands then linked the BMP2 status to survival data. Survival when exposed to soluble factors excreted by SMAD of the BMP2 negative stroma was significantly worse 4 negative CRC cells. These BMP2 ligands can induce compared to the survival of the BMP2 positive stroma EMT and invasion in SMAD4 negative cancers and with a Log Rank test p = 0,007 (Figure 4A). In the stromal BMP2 expression. With IHC we demonstrated SMAD4 positive tumours no difference was found that presence of BMP2 at the invasive front of the in survival in the BMP2 positive and BMP2 negative tumour is associated with a poor prognosis in SMAD4 groups (Figure 4B). These results support our in vitro negative cancers. data that fibroblasts produce BMP2 and that BMP2 can induce EMT in SMAD4 negative cancers.

We revealed BMP2 as a possible factor in the fibroblasts-cancer cell interaction, but many factors have been found previously.16 Most likely it is a Discussion combination of several factors that if combined results To maintain morphological integrity colonic epithelial in the tumour progressive effects of the fibroblasts. cells need the surrounding stroma. Although stromal To further investigate all the possible factors, the cells initially contribute to the intestinal homeostasis, cancer associated fibroblasts should be investigated in malignancies, stromal fibroblasts can also augment more specifically in the future. Using Genome Wide tumour progression.4,12 Over the past years many Association Studies (GWAS), sequencing techniques, studies investigated the crosstalk between the tumour miRNA arrays and CpG methylation arrays the and fibroblasts. One topic of extensive study is the molecular footprint of cancer associated fibroblasts effects that fibroblasts have on cancer cells. can be identified.

An important finding is that fibroblasts surrounding a tumour can induce EMT in cancer cells, which can result in worse prognosis for CRC patients.13 Our aim is to define one of the factors the fibroblasts excrete in order to effectuate tumour progression.

Many mechanisms and factors have been proposed to play an important role in the fibroblast-cancer cells interaction, but the exact interrelationship is still unclear. Based on our previous results that BMPs can induce EMT and tumour progression in SMAD4 negative CRC we suggest a role for the Bone Morphogenetic Protein signalling pathway in the fibroblasts-cancer cell cross talk.

In our experiments we proved that BMP 2 induces invasion and is secreted by fibroblasts upon stimulation

by HT-29 CRC cell line. However we could not prove BMP2 as a requisite in fibroblast induced EMT. If the effect of fibroblasts fades out upon inhibiting BMP2 we could argue the causal role of BMP2 in inducing EMT in HT-29. To prove this causality experiments should be performed with Noggin, this inhibitor binds directly to BMP2 and prevents the interaction between BMP2 and the BMP receptor.17

We looked at expression of BMP in the stroma surrounding the tumour; earlier BMP expression in the tumour itself was assessed.18 We depicted BMPs Transforming Growth Factor-β, which is part of as a tumour progressor secreted by fibroblasts, we did the same signalling family as BMP, is showed to be not look at the BMP2 status of the tumour itself, in a tumour progressor produced by cancer associated contrast with studies earlier preformed. To exactly fibroblast previously,12 but BMPs are new candidates understand the role of BMPs as a tumour progressor, in fibroblasts-cancer cells interaction. BMPs are often the BMP production of the tumour itself must also be depicted as tumour suppressors in colorectal cancer taken into account. because they induce maturation and apoptosis in

June 2014|#5|Leiden BioMedical Journal |37

Thijs de Vos We conclude that BMP2 is excreted by fibroblasts and is a tumour progressor in SMAD4 negative tumours.

If the interaction between the tumour and stroma is better understood it might be a target for tumour suppressive therapy in future. We might inhibit the effect of BMP2 ligands in SMAD4 negative tumours we could void the drift of tumour cells to migrate and metastasize eventually and improve survival prognosis of colorectal cancer patients.

38|Leiden BioMedical Journal |#5 | June 2014

Nominatie Geneeskunde References 1. Mesker,W.E. et al. The carcinoma-stromal ratio of colon carcinoma is an independent factor for survival compared to lymph node status and tumour stage. Cell Oncol. 29, 387-398 (2007).

2. Mueller,M.M. &Fusenig,N.E. Friends or foes bipolar effects of the tumour stroma in cancer. Nat. Rev. Cancer 4, 839-849 (2004).

3. Galon,J. et al. Type, density, and location of immune cells within human colorectal tumors predict clinical outcome. Science 313, 1960-1964 (2006).

4. De,W.O. &Mareel,M. Role of tissue stroma in cancer cell invasion. J. Pathol. 200, 429-447 (2003).

5. Kalluri,R. &Weinberg,R.A. The basics of epithelial-mesenchymal transition. J. Clin. Invest 119, 1420-1428 (2009).

6. Brabletz,T. et al. Invasion and metastasis in colorectal cancer: epithelial-mesenchymal transition, mesenchymal-epithelial transition, stem cells and beta-catenin. Cells Tissues. Organs 179, 56-65 (2005).

7. Voorneveld,P.W. et al. Loss of SMAD4 Alters BMP Signaling to Promote Colorectal Cancer Cell Metastasis via Activation of Rho and ROCK. Gastroenterology(2014).

8. van den Brink,G.R. &Offerhaus,G.J. The morphogenetic code and colon cancer development.

Cancer Cell 11, 109-117 (2007).

9. Hardwick,J.C., Kodach,L.L., Offerhaus,G.J., & van den Brink,G.R. Bone morphogenetic protein signalling in colorectal cancer. Nat. Rev. Cancer 8, 806-812 (2008).

10. Kodach,L.L. et al. The bone morphogenetic protein pathway is active in human colon adenomas and inactivated in colorectal cancer. Cancer 112, 300306 (2008).

11. Mesker,W.E. et al. Presence of a high amount of stroma and downregulation of SMAD4 predict for worse survival for stage I-II colon cancer patients.

Cell Oncol. 31, 169-178 (2009).

12. Hawinkels,L.J. et al. Interaction with colon cancer cells hyperactivates TGF-beta signaling in cancer-associated fibroblasts. Oncogene(2012).

13. Sheehan,K.M. et al. Signal pathway profiling of epithelial and stromal compartments of colonic carcinoma reveals epithelial-mesenchymal transition.

Oncogene 27, 323-331 (2008).

14. Beck,S.E. et al. Bone morphogenetic protein signaling and growth suppression in colon cancer. Am. J. PhysiolGastrointest. Liver Physiol 291, G135-G145 (2006).

15. Freeman,T.J. et al. Smad4-mediated signaling inhibits intestinal neoplasia by inhibiting expression of beta-catenin. Gastroenterology 142, 562-571 (2012).

16. Espina,V. et al. Laser-capture microdissection.

Nat. Protoc. 1, 586-603 (2006).

17. Zimmerman,L.B., De Jesus-Escobar,J.M., &Harland,R.M. The Spemann organizer signal noggin binds and inactivates bone morphogenetic protein 4.

Cell 86, 599-606 (1996).

18. Deng,H. et al. Bone morphogenetic protein-4 is overexpressed in colonic adenocarcinomas and promotes migration and invasion of HCT116 cells.

Exp. Cell Res. 313, 1033-1044 (2007).

June 2014|#5|Leiden BioMedical Journal |39

Tjeerd Muurling

Metabolomics analysis of pharmacotherapies for sensorineural hearing loss T. Muurling, Bsc; K. M. Stankovic, M.D., PhD

Abstract Hypothesis Different pharmacotherapies for sensorineural hearing loss (SNHL) are interconnected in metabolic networks with molecular hubs. Background Sensorineural hearing loss is the most common sensory deficit worldwide. Dozens of drugs have shown efficacy against SNHL in animal studies, and a few in human studies. Analyzing metabolic networks that interconnect these drugs will point to and prioritize development of new pharmacotherapies for human SNHL. Methods Drugs that have shown efficacy in treating mammalian SNHL were identified through PubMed literature searches. The drugs were analyzed using Ingenuity Pathway Analysis. The 3 most interconnected molecules and drugs within the generated networks were considered important targets for treatment of SNHL. Results A total of 70 drugs were investigated with IPA. The metabolomic analysis revealed two statistically significant networks (network 1 and 2). A network analysis using the ‘grow-tool function’ generated one statistically significant network. Hubs of these networks were: P38 MAPK, ERK1/2 and Glutathione for network 1; Akt, NFkB and ERK for network 2; and Dexamethasone, Tretinoin and Cyclosporin A for network 3. Conclusions Metabolomic and network analysis of the existing pharmacotherapies for SNHL has pointed to and prioritized a number of potential novel targets for treatment of SNHL.

Introduction Sensorineural hearing loss (SNHL) is an etiologically heterogeneous disorder resulting from many genetic and environmental insults (1). As the most common birth defect in industrialized countries and the most prevalent sensory deficit (2), SNHL is one of the biggest public health problems. Over 275 million people globally have moderate-to-profound hearing impairment (3). In the USA alone, the number of people with hearing loss has doubled over the past 30 years, currently affecting approximately 30 million people (4). With the growing number of individuals affected by this disability, there is a mounting need to develop effective therapies for SNHL. Over the years, many drugs and compounds have been tested to try to identify pharmacologic agents to treat and prevent SNHL. From here on, we refer to them collectively as “drugs.” These drugs include endogenous mammalian chemicals that are part of normal metabolism, such as antioxidants, as well as synthetic chemical drugs, such as dexamethasone.

Dozens of the tested drugs have shown some efficacy against SNHL in animals, and a few in humans. Despite these efforts, medical therapies for human SNHL 40|Leiden BioMedical Journal |#5 | June 2014

are sorely lacking. The molecular pathophysiology underlying different kinds of SNHL is complex and the number of identified genes related to hearing loss is increasing (1, 5, 6). A better insight into promising pharmacotherapies for SNHL may be obtained by analyzing molecular networks through which drugs with a known effect on SNHL interact with the body’s metabolome. The term ‘metabolome’ refers to all of the low molecular weight metabolites of biological origin that are present in a biological organism and represent the metabolic status of an organism at the time of a sample collection (7, 8). The detectable small molecules in a sample may arise from endogenous metabolites and their intermediates, signaling molecules, or exogenous metabolites from the diet, environment, gut microflora or intake of drugs (7).

Metabolomics refers to the study of some or all of these small molecules.

In this study, we performed metabolomic analysis of the endogenous chemicals used for successful treatment of SNHL in mammals, and combined it with a network analysis to include chemical drugs with a significant effect on mammalian SNHL. Network-

Nominatie Geneeskunde based analysis, as opposed to traditional, linear, reductionist analysis, is gaining momentum in all biomedical fields (8, 9, 10) because networks provide a more holistic approach to capture the complexity of intracellular and intercellular interactions. The rapidly emerging field of network medicine, which integrates systems biology and network science, promises to transform our understanding of disease etiology, classification and treatment (10, 11). The new pharmacologic treatment strategies aspire to capitalize on the understanding of networks to overcome increasing limitations of traditional high affinity, high specificity compounds (12). Therefore, pharmacologic targeting of entire pathways, promises to overcome limitations of compounds that target individual proteins (12,13).

To study the existing pharmacotherapies for SNHL within the global environment of interacting molecules, and thereby identify most promising compounds and pathways for clinical applications, we used Ingenuity Pathway Analysis (IPA). IPA has been extensively used in biomedical research, to analyze big data because its Ingenuity Knowledge Base is the largest curated database of biological and chemical interactions, extracted from the scientific literature (7,14). By using IPA’s bioinformatics tools to search the knowledge base, we have identified several potential novel molecular targets for treatment of SNHL, and prioritized human testing of drugs that have shown promise in animal studies.

Materials and Methods Literature Review To identify pharmacotherapies for SNHL tested in mammals, a PubMed English literature search (from January 1960 to August 2012) was performed using the Mesh terms hearing loss, deafness combined with the MESH terms prevention and control, drug and drug therapy. We considered drugs that improved SNHL, as reflected in lowering of sound level required to elicit auditory brainstem response (ABR) or evoked otoacoustic emissions (OAEs), or decreased loss of hair cells or stereocilia. After elimination of duplicates and articles of unrelated topics, only studies that included control groups with at least three subjects, and reported a statistically significant difference due to experimental treatment (p<0.05), were selected (Figure 1 and Table Supplemental Digital Content 1 listing drugs used in the analysis, with indication, species and criteria of efficacy [http://links.lww.com/ MAO/A193]).

Metabolomic Analysis Metabolomic Analysis of the drugs identified above(Table Supplemental Digital Content 1 [http:// links.lww.com/MAO/A193]), was performed using IPA (7). IPA uses the Ingenuity Knowledge Base, which includes millions of measured interactions between genes, proteins, cellular complexes, cells, tissues, drugs, pathways, and diseases, based on published biomedical literature and textbooks. Ingenuity Knowledge Base is the largest curated database, which is updated weekly to include information published as recently as the prior week (7).

We used IPA to map out interactions between drugs implicated in improvement of SNHL and molecular mechanisms by constructing metabolic networks.

Networks are created from “seed molecules”, also known as seeds – in our case, the drugs we provided – by IPA searching the knowledge base for molecules that are known to biologically interact with the seeds, and connecting these molecules with the seeds.

Any two molecules that bind, act upon one another, or that are involved with each other in any other manner would be considered connected or linked.

The most highly connected seeds and molecules from the knowledge base are consolidated into networks, which graphically represent the regulatory interactions that exist among user-specified seeds, based on characterized interactions within the knowledge base (7). Within networks, molecules are referred to as nodes, and the biological interaction between two nodes is represented as a line; the line is solid if the interaction is direct, and dashed if the interaction is indirect. Direct interactions require that two molecules make direct physical contact with each other, without an intermediate step. Indirect interactions do not require a physical contact between the two molecules, such as a hormone causing a change in expression of a downstream protein via a signaling cascade (7). Nodes are displayed using various shapes that represent the functional class of the gene product. The most interconnected molecule (or class of molecules) in a network is considered the hub.

Within IPA, statistical analysis of networks and pathways was done with the right-tailed Fisher’s exact test. The networks and pathways are scored by significance value, which is a measure of the likelihood that the associations forming the network are due to chance alone (7). The score is the negative logarithm of the p value, which represents significance. We considered networks with a score greater than 5 (i.e. p <10-5) to be significant.

June 2014|#5|Leiden BioMedical Journal |41

Tjeerd Muurling chemical drugs, were uploaded as independent nodes. These nodes were then linked or connected by growing the network with 100 molecules from the Ingenuity Knowledge Base. IPA calculated which 100 molecules from the knowledge base had the highest degree of connectivity to the drugs from our database, to form a network with these drugs.

Figure 1 - Diagram of the strategy for identifying, from the literature, relevant studies on pharmacologic therapies for SNHL

A metabolomic analysis of our dataset containing drugs implicated in improvement of SNHL was performed on September 12th 2012 with IPA version 192063. IPA allows the maximal number of molecules per network to be selected as 35, 70 or 140; we chose 35 to allow the possibility that all drugs that we studied and that were considered metabolites by IPA (i.e. 20 drugs) were a part of a single, small network. An additional advantage of a small metabolic network is that it remains visually tractable even when augmented with 100 additional chemical drugs. Specifically, the IPA networks through a metabolomic analysis can be formed only with “endogenous mammalian chemicals” (7), which represent nodes that are linked if they participate in the same biochemical reactions (9). To add “chemical drugs” from our dataset to the metabolic networks, an extra feature of IPA was used: the “Grow Tool function” from “My Pathway” toolbox. The “Grow Tool function” makes it possible to add new molecules and their interactions, where priority is given to those molecules that, according to the Ingenuity Knowledge Base, have a high degree of connectivity with the other molecules (7). In our case, the chemical drugs with a high degree of connectivity to the molecules in the metabolic networks were added to the networks. An additional network analysis was done using the Grow Tool function only, without starting with a metabolic network first. For this analysis all pharmacotherapies, including endogenous mammalian chemicals and 42|Leiden BioMedical Journal |#5 | June 2014

Results The literature searches (Figure 1) identified 135 studies that met our inclusion criteria. These studies generated 85 drugs, 70 of which were present in the Ingenuity Knowledge Base, and therefore were eligible for our study. Twenty of the drugs were endogenous mammalian chemicals and the rest were chemical drugs. The drugs were studied for different types of SNHL, including noise induced hearing loss (NIHL), presbycusis, cisplatin ototoxity, aminoglycoside ototoxity or unspecified SNHL. The drugs were mostly tested in rodents (guinea pigs, rats, mice, chinchillas, gerbils or hamsters), and a few were tested in humans (Table Supplemental Digital Content 1 [http://links. lww.com/MAO/A193]).

Metabolomic analysis of the pharmacotherapies for SNHL that consisted of endogenous mammalian chemicals revealed 3 networks. The first two were highly significant (p=10-28 and p=10-11, respectively) while the third network did not meet our criterion for significance (p=10-2). After growing the top ranking network with the chemical drugs from our database, the resulting network was highly interconnected (Figure 2), reflecting the complexity of the underlying molecular and chemical interactions.

This complexity can be simplified by focusing on the most interconnected molecules, or hubs, as they are thought to hold the whole network together (16). The main hub of the top ranking network, which had been augmented with chemical drugs, was P38 MitogenActivated Protein Kinase (P38 MAPK), having 47 connections or links to other nodes. The subsequent 4 hubs were p44/p42 MAP kinase (ERK) with 45 connections, glutathione with 34 connections, Insulinlike Growth Factor 1 (IGF1) with 27 connections and Activator protein 1 (Ap1) with 26 connections (Table 1, Figure 2 and Figure S1 Supplemental Digital Content 2 [http://links.lww.com/MAO/A194]). After growing the second top ranking metabolic network with the chemical drugs from our database, the top five hubs were Protein Kinase B (Akt) with 54 connections, nuclear factor kappa-light-chain-enhancer of activated B cells (NFkB) with 50 connections, p42/44 MAPK (ERK) with 42 connections, Insulin with 40

Nominatie Geneeskunde connections and Phosphatidylinositide 3-kinases (PI3K) also with 40 connections (Table 1 and Figure S2 Supplemental Digital Content 3 [http://links.lww. com/MAO/A195]).

In addition to networks, which reveal interactions among user-specified molecules, the metabolomic analysis also identified canonical pathways, which are well established cell signaling and metabolic pathways such as the Krebs cycle (6, 7). It is remarkable that 4 out of 5 top canonical pathways (Table 2) discovered

by IPA are directly involved with metabolism of retinoic acid, which is a metabolite of vitamin A. Even the remaining pathway, aryl hydrocarbon receptor signaling, has recently been shown to be controlled by all trans retinoic acid (tretinoin) (17).

For network analysis using the Grow Tool function only to establish connections between all reported pharmacotherapies for SNHL by growing the network with 100 additional molecules, a highly complex network emerged (Table 1 and Figure S3 Supplemental Digital Content 4 [http://links.lww.com/MAO/A196]).

This network included 62 of the 70 drugs from our database, suggesting substantial interaction among the drugs. By focusing on the hubs of this network, we identify drugs that are most likely to be effective in treating different kinds of SNHL: dexamethasone (61 connections), tretinoin (45 connections), cyclosporin A (40 connections), N-acetyl-L-cysteine (37 connections) and resveratrol (36 connections).

Discussion A comprehensive metabolomic, network and pathway analysis of the reported pharmacotherapies for mammalian SNHL has highlighted several molecules and pathways that are attractive targets for development of novel pharmacotherapies for SNHL. These molecules include signaling kinases, such as P38 MAPK, ERK1/2, Akt and PI3K (Table 1). Of these, ERK1/2 is particularly appealing because it was also identified as a key node in our comprehensive network analysis of human deafness genes (6). We have suggested that mutations in the gene encoding ERK1, known as MAPK1, may account for human nonsyndromic hearing loss at the DFNB40 locus because MAPK1 is located within that locus, which awaits identification of the causative gene (6).

Given that ERK1/2 can signal death or survival of cochlear cells, depending on stimulus (18, 19), potential therapeutic targeting of ERK1/2 may need to be personalized to avoid toxic side effects.

Otology is likely to benefit from and impact oncologic

Figure 2 - The central part of the most significant network generated

by IPA metabolomic analysis of the drugs that ameliorate SNHL and are known to be endogenous mammalian chemicals augmented with a grow tool to include chemical drugs. The full network is shown in supplemental Figure 1, http://links.lww.com/MAO/A194. The input drugs, framed in blue, are connected with each other through complex interactions that are direct (solid line) or indirect (dashed line) and involve molecules provided by the Ingenuity Knowledge Base. This complexity can be simplified by focusing on the most interconnected molecules, that is, hubs, which are highlighted with yellow frames

research because of prominent role that ERK plays in pathophysiology of diverse neoplasms, and because ERK inhibitors hold chemotherapeutic promise in clinical trials (20).

Our comprehensive analysis of the reported pharmacotherapies for mammalian SNHL has also prioritized future clinical testing of the drugs that have shown promise in animal studies.

Of these, tretinoin and modulators of vitamin A signaling are especially attractive because tretinoin is not only the second hub in the interactome of the reported pharmacotherapies for SNHL, but retinoic acid signaling also dominates pathway analysis of the 20 endogenous mammalian chemicals with a reported efficacy in alleviating SNHL (Table 1, 2). In line with the modern pharmacologic emphasis on targeting entire pathways, rather than single proteins (12, 13), our study suggests that pharmacologic targeting of the retinoic acid pathway may be promising in preventing and treating SNHL. Vitamin A is a family of hydrocarbons, which include retinol, retinal, retinoic acid and several provitamin A carotenoids, such as beta-carotene (21). Retinoic acid, both alltrans retinoic acid (Tretinoin) and 9-cis retinoic acid (Alitretinoin), signals via the retinoic acid receptor (RAR). The critical role of retinoic acid signaling for normal development of the inner ear is demonstrated in studies showing that either excess or deficiency in retinoic acid leads to inner ear dysmorphogenesis (22). Although retinoic acid signaling has not been studied in the adult inner ear, it has been studied in the adult hippocampus where retinoid hyposignaling June 2014|#5|Leiden BioMedical Journal |43

Tjeerd Muurling Network 1 ANALYSIS Molecule (No.connections) 1. Metabolomics analysis P38 MAPK (47) ERK1/2 (45) Glutathione (34) IGF1 (27) Ap1 (26) 2. Network analysis using the grow tool Dexamethasone (61) Tretinoin (45) Cyclosporin A (40) N-acetyl-L-cysteine (37) Resveratrol (36)

Network 2 Molecule (No.connections) Akt(54) NFkB (50) ERK (42) Insulin (40) PI3K (complex) (40) -

Table 1 - The top 5 nodes of the networks that met our criterion for significance based on metabolomics analysis (Analysis 1), or network analysis via grow tool of all reported pharmacotherapies for SNHL (Analysis 2). The molecules in bold were not part of our original data base whereas the other molecules were.

contributes to age-related decline in memory (23).

A recent prospective study has shown that oral isotretinoin, a drug widely used for treatment of extensive and nodulocystic acne, improved hearing of all 38 treated adults who had both acne and hearing loss (24). This study, combined with our data, strongly motivate future large-scale prospective randomized study to establish long-term efficacy of vitamin A derivatives in alleviating human SNHL; multi-drug targeting of the retinoic acid pathway may be especially promising. Localized drug delivery to the inner ear would circumvent potential systemic toxicity.

It is reassuring that dexamethasone, has emerged as the central node within the interactome of reported pharmacotherapies for mammalian SNHL, given the well documented therapeutic effect of intratympanic dexamethasone in alleviating sudden SNHL in humans. The marked interconnectedness of dexamethasone is not surprising because dexamethasone is known to greatly change the global expression profiles of genes and microRNA in various tissues, including the inner ear (25, 26, 27). The prominence of cyclosporine A, and the emergence of the pro-inflammatory transcription factor NFkB as a major node in our metabolomics analysis highlight therapeutic significance of controlling inflammation to prevent hearing loss.

The methodology we have introduced has important implication for development of novel pharmacotherapies for SNHL. Otologic patients may benefit from the drugs originally developed for other indications. Indeed, a systems-level exploration of off-target drug effects is gaining momentum to not only understand side-effect profiles of the various 44|Leiden BioMedical Journal |#5 | June 2014

Top Canonical Pathways Retinoic Acid Mediated Apoptosis Signaling Retinoate Biosynthesis I Retinoate Biosynthesis I Aryl Hydrocarbon Receptor Signaling Retinoic Acid Receptor (RAR) Activation

P value 3.32E-04 6.6E-04 6.6E-04 1.09E-03 1.63E-03

Table 2 - The top canonical pathways discovered by metabolomics analysis of reported pharmacotherapies for mammalian SNHL (Analysis 1).

drugs, but also to discover potent and unexpected therapeutic targets for drug repurposing (28, 29).

A shortcoming of our approach is that it is inevitably biased by the most extensively studied molecules and molecular interactions, which dominate the biomedical literature across fields, and hence dominate the bioinformatic knowledge base. In addition, our analyses are constrained in time because the knowledge base is continuously evolving to reflect the latest discoveries from the literature. Nonetheless, our study outlines a network-based approach to systematizing the literature on pharmacotherapies for SNHL, which can be reapplied as the knowledge base expands. The network based approach provides a new insight into, and prioritizes testing of the most promising pharmacologic strategies for future prospective randomized clinical trials.

Conclusions We provide the first metabolomic and network analysis of pharmacotherapies for SNHL. By studying molecular interconnectedness between the reported

Nominatie Geneeskunde pharmacotherapies for SNHL, we define and prioritize potential novel targets and pharmacotherapies for human SNHL. Our analyses suggest that modulation of inner-ear signaling via ERK1/2 and vitamin A, either alone or in combination with therapies targeting inflammation, are promising strategies for future prospective randomized clinical trials for SNHL. This article is an abridged version of the paper published in Otology&Neurotology (Otol Neurotol.

2014 Jan;35(1):1-6).

References

1.

Nance WE. The genetics of deafness. Mental retardation and developmental disabilities research reviews 2003; 9: 109–19.

2.

Morton CC, Nance WE. Newborn hearing screening--a silent revolution. New England J Med 2006; 354: 2151–64.

3.

World Health Organisation. Deafness and Hearing Impairment at http://www.who.int/mediacentre/factsheets/ fs300/en/index.html 4.

American Speech-Language-Hearing Association. The Prevalence and Incidence of Hearing Loss in Adults. at http:// www.asha.org/public/hearing/disorders/prevalence_adults.htm 5.

Hereditary Hearing Loss Homepage. at http:// hereditaryhearingloss.org 6.

Stamatiou GA, Stankovic KM. A Comprehensive Network and Pathway Analysis of Human Deafness Genes. Otol Neurotol 2013 Jul;34(5):961-70 7.

Ingenuity Systems. at www.ingenuity.com 8.

Goodacre R. Metabolomics – the way forward.

Metabolomics 2005; 1:1-2.

9.

Jeong H, Tombor B, Albert R, Oltvai ZN, Barabási AL. The large-scale organization of metabolic networks. Nature.

2000;407(6804):651-4.

10. Barabási AL, Gulbahce N, Loscalzo J. Network medicine: a network-based approach to human disease. Nat Rev Genet. 2011;12(1):56-68.

11.

Pawson T, Linding R. Network medicine. FEBS Lett.

2008;582(8):1266-70.

12.

Schrattenholz A, Groebe K, Soskic V. Systems biology approaches and tools for analysis of interactomes and multitarget drugs. Methods Mol Biol.2010;662:29-58.

13.

Pujol A, Mosca R, Farrés J, Aloy P. Unveiling the role of network and systems biology in drug discovery. Trends Pharmacol Sci. 2010;31(3):115-23.

14. Valencia-Cruz AI, Uribe-Figueroa LI, GalindoMurillo R, Baca-López K, Gutiérrez AG, Vázquez-Aguirre A, Ruiz-Azuara L, Hernández-Lemus E, Mejía C. Whole genome gene expression analysis reveals casiopeína-induced apoptosis pathways. PLoS One. 2013;8(1):e54664.

15. Durga J, Verhoef P, Anteunis LJ, Schouten E, Kok FJ. Effects of folic acid supplementation on hearing in older adults: a randomized, controlled trial. Ann Intern Med. 2007;146(1):1-9.

16. Jeong H, Mason SP, Barabási AL, Oltvai ZN. Lethality and centrality in protein networks. Nature. 2001;411(6833):41-2.

17. Ohno M, Ikenaka Y, Ishizuka M. All-trans retinoic acid inhibits the recruitment of ARNT to DNA, resulting in the decrease of CYP1A1 mRNA expression in HepG2 cells. Biochem Biophys Res Commun 20126;417(1):484-9.

18.

Battaglia A, Pak K, Brors D, Bodmer D, Frangos JA, Ryan AF. Involvement of ras activation in toxic hair cell damage of the mammalian cochlea. Neurosci 2003;122:1025-35.

19.

Lahne M, Gale JE. Damaged-induced activation of ERK 1/2 in cochlear supporting cells is a hair cell death-promoting signal that depends on extracellular ATP and calcium. J Neurosci 2008;28:918-28.

20.

Kohno M, Tanimura S, Ozaki K. Targeting the extracellular signal-regulated kinase pathway in cancer therapy.

Biol Pharm Bull 2011;34(12):1781-4.

21.

Fennema O (2008). Fennema’s Food Chemistry. CRC Press Taylor &Francis Group. pp. 454–455.

22.

Frenz DA, Liu W, Cvekl A, Xie Q, Wassef L, Quadro L, Niederreither K, Maconochie M, Shanske A. Retinoid signaling in inner ear development: A “Goldilocks” phenomenon. Am J Med Genet 2010;152A(12):2947-61.

23.

Mingaud F, Mormede C, Etchamendy N, Mons N, Niedergang B, Wietrzych M, Pallet V, Jaffard R, Krezel W, Higueret P, Marighetto A. Retinoid hyposignaling contributes to aging-related decline in hippocampal function in shortterm/working memory organization and long-term declarative memory encoding in mice. J Neurosci 2008;28(1):279-91.

24.

Karabulut H, Karadag AS, Acar B, Dagli M, Karabulut I, Ozmen E, Babademez MA, Karaşen RM. The effect of oral isotretinoin (13-cis retinoic acid) on hearing systems in patients with acne vulgaris: a prospective study. Int J Dermatol 2011;50(9):1139-43.

25.

Maeda Y, Fukushima K, Hirai M, Kariya S, Smith RJ, Nishizaki K. Microarray analysis of the effect of dexamethasone on murine cochlear explants. Acta Otolaryngol 2010;130(12):132934.

26.

Liu L, Walker EA, Kissane S, Khan I, Murray PI, Rauz S, Wallace GR. Gene expression and miR profiles of human corneal fibroblasts in response to dexamethasone. Invest Ophthalmol Vis Sci 2011;52(10):7282-8.

27.

Lee MJ, Gong DW, Burkey BF, Fried SK. Pathways regulated by glucocorticoids in omental and subcutaneous human adipose tissues: a microarray study. Am J Physiol Endocrinol Metab 2011;300(3):E571-80.

28.

Scheiber J, Chen B, Milik M, Sukuru SC, Bender A, Mikhailov D, Whitebread S,Hamon J, Azzaoui K, Urban L, Glick M, Davies JW, Jenkins JL. Gaining insight into off-target mediated effects of drug candidates with a comprehensive systems chemical biology analysis. J Chem Inf Model 2009;49(2):308-17.

29.

Chang RL, Xie L, Xie L, Bourne PE, Palsson BØ.

Drug off-target effects predicted using structural analysis in the context of a metabolic network model. PLoS Comput Biol 2010;6(9):e1000938.

June 2014|#5|Leiden BioMedical Journal |45

Overzicht Onderwijsprijzen 2008 â&#x20AC;&#x201C; 2013

Winnaars Prof.dr. G.J. Tammeling Onderwijsprijs 2008-2009:

Dr. M. Gosselink 2009-2010: Dr. B.P. Hierck 2010-2011: Dr. O.C. Meijer 2011-2012: Prof. dr. Bolk 2012-2013: Dr. B.P. Hierck

Beste Blok Geneeskunde

Student Research Award

2008-2009: 2009-2010: 2010-2011: 2011-2012: 2012-2013:

2008-2009 BW: Jolien Suurmond GNK: Marc Lobatto

Bewegingsapparaat Buik Trauma Kind en Jongere Van Cel tot Molecuul

Beste Blok Biomedische wetenschappen

2009-2010 BW: Marie-Louise van der Hoorn GNK: Dario Valerio

2008-2009: 2009-2010: 2010-2011:

2010-2011 BW: Hellen Buijze GNK: Sjoerd Joustra

2011-2012: 2012-2013:

Hormones and the Nervous System FoS Signal Transduction FoS Regulation of the Immune response Hormones and the Nervous System Molecular Biology and Oncology

Overzicht Beste Co-schap 2008-2009 2009-2010 2010-2011 2011-2012 2012-2013

Heelkunde, Bronovo Heelkunde, MCH Westeinde Gynaecologie, GHZ Heelkunde, MCH Westeinde Gynaecologie, Rijnland

46|Leiden BioMedical Journal |#5 | June 2014

2011-2012 BW: Wil Stutterheim GNK: Thomas Vellinga 2012-2013 BW: Gido Gravesteijn GNK: Annemijn Algra

Scriptieprijs voor het beste stageverslag Geneeskunde of Biomedische wetenschappen “Het winnen van een scriptieprijs is erkenning voor het werk dat je er aan besteed hebt!”

-Y. de Jong.

“De LAG prijs motiveerde ons erg om door te zetten met publiceren van ons artikel!”

-J.H.F.M. Pinckaers “Het winnen van deze prijs vormt voor ons een mooie kroon op het werk wat wij de afgelopen maanden verzet hebben. Wat ons betreft is het een mooie opstap naar meer kwalitatief hoogstaand onderzoek in de toekomst!”

-R.M. ten Brinck

Dit zijn uitspraken van de winnaars van het afgelopen jaar. Zij vonden dat bij patiënten met LUTS klachten, een verbeterd urodynamisch profiel werd bewerkstelligd in een zittende positie. Met het verdiende prijzengeld betalen ze de publication fee bij PLOS ONE, het blad waarin hun artikel wordt gepubliceerd.

Wil jij ook deze Scriptieprijs in de wacht slepen? Dat kan, want in 2014 reikt de Leidse Alumnivereniging Geneeskunde voor de zesde maal de jaarlijkse scriptieprijs uit aan de student met het beste stageverslag Geneeskunde of Biomedische Wetenschappen (Ba/Ma).

Het verslag moet aan de volgende criteria voldoen: - origineel qua onderzoek; - maatschappelijk relevant onderwerp; - heeft een kop en een staart; - bevat een heldere hypothese, onderzoeksopzet, methoden, resultaten, discussie, conclusies

en aanbeveling(en).

De winnaar ontvangt € 1.000,- en houdt een voordracht van ca. 15 minuten over het onderzoek op de jaarlijkse LAG-lezing op donderdag 20 november 2014.

Informatie of vragen? Insturen kan tot woensdag 1 oktober 2014. Meld je aan door een e-mail, met uw onderzoeksrapport of scriptie, en uw volledige naam, adres en telefoonnummer, naar lag@lumc.nl te sturen.

Zend hierbij ook een bewijs van het eindresultaat van uw scriptie mee.

Meer informatie: www.lumc.nl/lag

June 2014|#5|Leiden BioMedical Journal |47