Allergenic risks of mealworm and other insects An approach to assess the risks of new food proteins in allergic patients Allergene risico’s van meelworm en andere insecten Een aanpak voor de benadering van de risico’s van nieuwe voedingseiwitten in allergische patiënten (met een samenvatting in het Nederlands) Proefschrift ter verkrijging van de graad van doctor aan de Universiteit Utrecht op gezag van de rector magnificus, prof.dr. G.J. van der Zwaan, ingevolge het besluit van het college voor promoties in het openbaar te verdedigen op dinsdag 23 mei 2017 des middags te 2.30 uur door Henrike Casparina Helena Petronella Hustinx-Broekman geboren op 28 november 1980 te Utrecht
Transcript
Allergenicrisksofmealwormandotherinsects An approach to assess the risks of new food proteins in allergicpatientsAllergenerisico’svanmeelwormenandereinsectenEen aanpak voor de benadering van de risico’s van nieuwevoedingseiwitteninallergischepatiënten(meteensamenvattinginhetNederlands)
Proefschrift
ter verkrijging vandegraadvandoctor aandeUniversiteitUtrechtop gezag van de rector magnificus, prof.dr. G.J. van der Zwaan,ingevolgehetbesluitvanhetcollegevoorpromotiesinhetopenbaarteverdedigenopdinsdag23mei2017desmiddagste2.30uur
Chapter1:GeneralIntroductionInnovationsinfoodsupplyandallergyrisksThe growing world population and the increased impact on theenvironmentcreatesthedemandforchangesinagriculturalpracticesandfood supply. New or improved (climate or pest resistant) crops (e.g.geneticallymodifiedmaizeorsoy),currentlyunusedby-products(e.g.beetleaves)andalternativesourcesoffoodprotein(e.g.insects,algae)canhelpmakeour foodsupplymuchmore sustainable. Insectsare suggestedasamoresustainablealternativeformeatandfish,duetoshorterproductionduration,comparedtootheranimalproteinsources,andamorefavorablebiomass conversion rate. This is combined with a high nutritional andenergycontent.Insectproductionhasadditionaladvantagesofinnovationsin agricultural practices. However, innovations in food supply will onlysucceed if they are healthy and safe. Solving one problem should notcreate a new one. Therefore, when new foods are introduced on themarked,precautionisneededtoavoidfoodsafetyissues.Legislationintheform of the General Food Law is in place to ensure safety and hold theproducer responsible for a high safety standard of the product [1].Additionally,foodsafetyaspectsshouldbeassessedbeforeanewfoodcancommerciallyentertheEuropeanmarket.Guidelinesarepresenttohelpinthe safety assessment of food additives, flavorings, food enzymes, foodsbasedongeneticallymodifiedorganisms(GMO’s)andinfantformula[2,3].Whenanew food is introduced thathasnotbeenconsumed regularly inEuropebeforeMay1997,andisnotregulatedbyanyspecificregulation,itis considered a ‘Novel food’. The EU Novel Food Law prescribes specificprocedures and requirements for marketing approval of new foods,ingredientsandprocesses[4].Aspects of food safety tobe addressed according to theNovel Food Lawarenutritional,microbiological,toxicologicalandallergenicsafety.Whenitcomes to products that are based on – or contain – new or modifiedproteins, allergenicity especially poses a potential health risk. To preventthe emergence of food allergies to new foods introduced into our diet,adequate and accepted methods and standards for assessing theconsumers’ expected health response to new or modified products areneeded.Suchmethodsandstandardsarecurrentlylacking.
Chapter1
4
Food allergy, and particularly Immunoglobulin E (IgE)-mediated foodallergy, isamajorhealthand foodsafetyproblemworldwide,andaffects2-4%ofadultsand5-8%ofyoungchildren[5,6].Foodallergyprevalencesin Europe vary between the different foods, and have been estimatedaroundpercentagesof0.2%forpeanut,0.6%forcow’smilk,0.5%fortreenuts and 0.1% for shrimp, based on positive double blind placebocontrolled food challenge (DBPCFC) [7]. IgE-mediated food allergy is animmunologic,non-toxicadversereactiontootherwiseharmlesssubstances(allergens) in food, generally food proteins. The mechanisms underlyingIgE-mediated food allergy consist of a sensitization and an elicitationphase. Sensitizationmay occur upon contactwith the food allergen, andresults in the generation of allergen-specific IgE (sIgE). Elicitation ofsymptomsmayoccuruponsubsequentcontactwiththeallergenleadingtosymptoms.Symptomsoccurwithinminutestohours(usuallylessthantwo)after allergen ingestion [8], and involve one or more of the followingsystems;theskin(pruritus,urticaria,orangioedema),thegastro-intestinaltract (diarrhea, vomiting, contractions, increased bowel movement), therespiratory tract (asthma attack, hoarseness, stridor/laryngealangioedema) or the cardiovascular system (dizziness, drop in bloodpressure, loss of consciousness) [9]. Some foods aremore likely to causemildsymptoms(e.g.apple)thanothers(e.g.peanut).Labelinglegislationisinplace inmost regionsof theworld thatprescribe that thepresenceofspecificmajorallergenic ingredients in foodproducts is tobedeclared toinformallergicconsumers[10,11].Assessmentofcross-reactivityandprimarysensitizationandallergybynewfoodproteins:thegoalandcontentsofthisthesisAllergies to new food proteins can result from cross-reactivity in existingsensitized or allergic individuals and thereby thus immediately manifestthemselvesinelicitationofallergicreactionsuponconsumption(elicitationphase). However,allergies tonew foodproteinsmayalso result fromdenovo sensitization of - and development of new allergies in - susceptibleindividuals, in this case thus first requiring these individuals to becomesensitized(sensitizationphase)beforesubsequentexposureorintakemayelicit allergic reactions (elicitation phase). The aim of this thesis was todevelopandtestastructuredapproachtoassessthecross-reactiveanddenovo allergenicity of potential new food proteins, and examine theallergenic risksof insectsaspotentialnew foodprotein sourcesasacasestudyforthestructuredapproach,withspecificfocusonmealwormasone
Chapter1
5
ofthecandidateinsectsasnewfoodproteinsource.ThereisnosubstantialhistoryofuseofinsectsinfoodinEurope.However,insomeotherregionsoftheworld,consumptionofinsectsismorecommon[12].Reportsoffoodallergy to insectsareknown,mostly fromAsia,where insectsarehighonthe list of causing anaphylactic symptoms [13]. Detailed descriptions andassessment of the relevance with respect to a possible introduction ofinsectsasanewfoodproteinsourceinEuropewerenotavailablepriortothestartofourstudy.In Chapter 2, the background and proposed structured approach forallergenicity assessment of new food proteins and protein sources ispresentedandstudiesonmealwormandother insectsarepresentedanddiscussed inthefollowingchapters.Both incross-reactiveallergy,suchasappleallergy inpollen-related-foodallergy [14],andprimaryallergy, suchasgenerallyisthecaseforpeanut[15],changesinallergenicitycausedbyheatinghavebeendescribed.Differentformsofprocessingthereforeneedtobeconsideredinallergenicityassessment.InChapter3,thisisillustratedfor the case for mealworm. Chapters 4 and 5 address risks of cross-reactivity of mealworm proteins in patients with allergies tophylogenetically related allergenic sources (shrimp and house dust mite)and risks of sensitization in patients with unrelated allergies. Primarysensitization and allergy caused by mealworm exposure is described inChapter6. InChapter7 it isaddressedwhetherand towhatextend risksidentified for mealworm may also apply to other insects that maycandidateasnewfoodproteinsource,withemphasisontwoidentifiedat-riskpopulations:shrimpandprimarymealwormallergicindividuals.Theresultsofthisthesisaresummarizedanddiscussedinthefinalchapter,inlightoftherisksofinsectproteinsfordifferentriskgroupsinthegeneralpopulation.
Chapter1
6
References:1. The European parliament and the Council of the European Union.
Regulation(EC)No258/97oftheEuropeanParliamentandoftheCouncilof 27 January 1997 concerning novel foods and novel food ingredients.OfficialJournalL043,14/02/1997P.0001-0006
pdeskapplworkflownutriinfant.pdf4. The European parliament and the Council of the European Union.
Regulation(EU)2015/2283oftheEuropeanParliamentandoftheCouncilof 25 November 2015 on novel foods, amending Regulation (EU) No1169/2011oftheEuropeanParliamentandoftheCouncilandrepealingRegulation(EC)No258/97oftheEuropeanParliamentandoftheCounciland Commission Regulation (EC) No 1852/2001. Official Journal of theEuropeanUnionL327/1,11/12/2015P.01–22
5. Sicherer SH, Sampson HA. Food allergy: Epidemiology, pathogenesis,diagnosis,andtreatment.JAllergyClinImmunol2014;133:291-307.
7. NwaruBI,HicksteinL,PanesarSS,RobertsG,etal.onbehalfoftheEAACIFood Allergy and Anaphylaxis Guidelines Group. Prevalence of commonfood allergies in Europe: a systematic review andmeta-analysis. Allergy2014;69(8):992–1007.
8. Bischoff S, Crowe SE. Gastrointestinal food allergy: new insights intopathophysiology and clinical perspectives. Gastroenterology 2005; 128:1089-113.
10. The European parliament and the Council of the European Union.Regulation (EC) No 2003/89 of the European Parliament and of theCouncilof10November2003amendingDirective2000/13/ECasregardsindicationofthe ingredientspresent infoodstuffs.Official JournalL308,25/11/2003P.0015-0018
11. The European parliament and the Council of the European Union.Regulation (EC) No 2006/142 of the European Parliament and of theCouncil of 22 December 2006 amending Annex IIIa of Directive2000/13/EC of the European Parliament and of the Council listing theingredientswhichmustunderallcircumstancesappearonthelabellingoffoodstuffs.OfficialJournalL368,23/12/2006P.0110-0111
12. Van Huis A, Potential of Insects as Food and Feed in Assuring FoodSecurity.AnnualReviewofEntomology,2013;58:563-583.
Chapter1
7
13. JiK.M,ChenJ,LiM,LiuZ.,etal.Anaphylacticshockandlethalanaphylaxiscaused by food consumption in China. Trends in food science &Thechnology,2009;20:227-231
14. DatemaMR, Zuidmeer-Jongejan L, Asero R, Barreales L, et al. Hazelnutallergy across Europe dissected molecularly: A EuroPrevall outpatientclinicsurvey.JAllergyClinImmunol.2015;136(2):382-91.
Chapter2:Allergenicityassessmentstrategy fornovel foodproteinsandproteinsourcesAbstractTo solve the future food insecurity problem, alternative and sustainableproteinsources(e.g.insects,rapeseed,favabeanandalgae)arenowbeingexplored for the production of food and feed. To approve these novelproteinsourcesforfuturefoodacomprehensiveriskassessmentisneededaccording to the European food legislation. Allergenicity risk assessmentmight pose some major difficulties, since detailed guidance on how toassess theallergenicpotentialofnovel foods isnot available.Atpresent,theapproachreliesmostlyontheguidanceofallergenicityassessmentforgeneticallymodified(GM)plantfoods.Themostrecentonewasproposedby EFSA [1,2]; “weight-of-evidence approach”. However, this guidance isdifficulttointerpret,notcompletelyapplicableorvalidatedfornovelfoodsand therefore needs some adjustments. In this paper, we propose aconceptualstrategywhich isbasedonthe“weight-of-evidenceapproach”forfoodderivedfromGMplantsandotherstrategiesthatwerepreviouslypublishedintheliterature.Thisstrategywillgivemoreguidanceonhowtoassesstheallergenicityofnovelfoodproteinsandproteinsources.1.IntroductionStrategiesarebeingdevelopedtochangethecurrentagriculturalpracticesby creating more sustainable and new climate resistant crops and toensure an adequate, safe, sustainable and nutritious food supply (e.g.alternativeproteinsources)inthenearfuture.Beforenovelfoodproteinsorproteincontainingproductscanbebroughttomarket,weneedtotakeprecautions to avoid that novel products will add to the burden of foodallergy.Atleastwehavetotakecarethatwewillnotintroduceallergensaspotentasthemajorallergenicfoodssuchaspeanut.Ontheotherhand,wealso have to be aware that any (novel) proteinmight have some risk ofallergenicity. Therefore, we need to take care that we will not excludepromising new protein sources with low or virtually absent allergenicpotential from the market. The EU novel food law requires that acomprehensive food safety assessment (addressing nutritional value,microbiological,toxicological,andallergenicrisks)hastobeperformedfor
Chapter2
12
allnovel foodsor food ingredients thatwerenotcommonlyconsumed inthe EU before May 1997, before they can be launched onto the foodmarket (EC regulation No 258/97 and EU recommendation 97/618 EC;http://eur-lex.europa. eu/). For the assessment of nutritional,microbiologicalandtoxicologicalrisks,standardandwelldefinedmethodsdoexist.Theassessmentofallergy risks foranovelproteinsource is lessstraightforward.Atpresent,theapproachreliesmostlyontheguidanceofallergenicity assessment for genetically modified (GM) plant foods. Themost recent one was proposed by EFSA [1,2]: the so-called “weight-of-evidence approach”. The purpose of these guidelineswas to prevent theintroductionofanallergenicproteinintoafoodsource,whichmightposeariskforconsumersallergicforthisproteinortopreventtheintroductionofa protein that is similar to an allergenic protein, so that cross reactivitymight occur. The applicability of these guidelines for the assessment ofnew and modified proteins or protein containing products (e.g. insects,algae, alternatively processed products) is hampered, since there is nogenerallyaccepted,validatedandbroadlyapplicablemethodavailable forallergenicityhazardandriskassessment.Theshortcomingsof thecurrentguidelines for this latter purpose will be discussed in this paper. Foodallergy is an adverse reaction of the human immune system to anotherwiseharmlessfoodcomponentandtheprevalenceoffoodallergyinEurope is up to 3% according to the EAACI food allergy and anaphylaxisguidelines group [3]. Food allergy develops in two phases. In the firstphase, susceptible subjects become sensitized to specific food proteinsafterdietaryexposure,orpossiblyviaotherroutesofexposure(inhalationand/or skin contact). Thismay result in theproductionof specific IgE tothe food protein [4,5].When sensitized subjects subsequently encounterthe respective allergen(s) again, cellular bound specific IgEwill recognizethe allergens and an allergic reactionmay be elicited. Allergic symptomsmayvaryconsiderablyandcanrangefrommild,localandtransienteffectstopotentialfatalreactions likesystemicanaphylaxis[6,7].Generally, foodallergensareproteinsbut the vastmajorityof foodproteins areweakorvirtually non-allergenic [8,9]. Most cases (90%) of food allergic reactionsarecausedbyalimitedrangeofproducts;milk,egg,peanut,treenuts,fish,soy, wheat and crustaceans [10-12]. Furthermore, the manifestations offood allergies can be dependent on geography, dietary habits, foodpreparation and age atwhich food is first consumed [13]. It is thereforepossible that a food product that was not reported to be common orknown as allergenic in Asia can be an allergenic food in Europe, for
Chapter2
13
examplekiwifruit[13].Anotherexampleistheallergytopeach,amemberof the Rosacea family which is attributed to birch pollen in Central andNorthernEurope(Prup1,theBetv1homologue,PR-10)andleadstomildreactions(oralallergysyndrome),whileintheMediterraneanareaswherebirchtreesarelesscommon,peachallergymayresultfromsensitizationtoPru p 3 (lipid transfer protein, LTP) and/or Pru p 4 (profilin)whichmorecommonly leads to severe allergic reactions [14]. At the moment, novelfoods such as insects and rapeseed are entering the market without aproper allergenicity risk assessment. For mealworms, larval stage of theyellowmealwormbeetle, itwas recently demonstrated in a double-blindplacebocontrolled foodchallenge (DBPCFC) that87%ofa shrimpallergicpatientpopulationshowedallergicreactionuponeatingYellowmealwormandthatdenovosensitizationtoYellowmealwormproteinsispossible[15Broekmanet al., 2015a] [ Broekman, JACI, in press]. In caseof rapeseed,whichwas formally inuse in theEUonly in the formof rapeseedoil, theEFSA panel concluded, that a risk of sensitization to rapeseed proteinisolate cannot be excluded and that it is likely that rapeseedwill triggerallergic reactions in mustard allergic subjects [16 EFSA NDA Panel (EFSANDAPanel, 2013)]. This conclusionwas based on a food challenge and askinpricktestswithcrushedrapeseed(notproteinisolate)inatopicFinnishchildrenwithatopicdermatitisandsuspectedfoodallergies.10.9%ofthechildren showed sensitivity in the SPT and89%of these children reactedpositive in the food challenge. Cross reactivity with mustard seeds wasdemonstrated using IgE binding tests with serum form mustard allergicpatients. Furthermore, structural homology of 95% of seed storageproteinsofvariousmembersofthebrassicaceae,incl.mustardwasshown.Inthisassessment,clinicallyrelevantstudieswereperformedwithcrushedrapeseed but not with rapeseed protein isolate. In the latter, a higherprotein concentration can be expected and furthermore, the effect ofprocessing was not taken into account. Other novel food dossierssubmitted in the last five years for approval by the EFSA (e.g. Chia seed,Lentinus edodos and alfalfa) were lacking properly conducted clinicallyrelevant tests (e.g. SPT, or basophil activation tests (BAT)) and in mostcases no formal proof of absence of allergenicity using double-blindplacebocontrolledfoodchallenge(DBPCFC)wasgiven,norwastheeffectofprocessingor the sensitizingpotency tested [17,8,19]. FoodchallengesareessentialfordeterminingifIgEbindingmeasuredwithtechniquessuchas immunoblot,BATandSPT isclinically relevant. IgEbindingor IgEcrossreactivity does not automatically indicate that an allergic reaction will
Chapter2
14
occur. For instance, some proteins have cross reactive carbohydratedeterminants(mostlyfoundinplants)thatbindtoIgEbutdonotelicitanallergicreaction[20].Furthermore,crossreactivitybetweentaxonomicallyrelatedfoods,suchasthelegumefamily(peanut,soy,lupine,whitebean)does not automatically indicate clinical cross reactivity [21]. Ibaῆez et al.showed that white bean and overall green bean are well tolerated bychildrenallergictootherlegumes[22].Itisintheinterestoftheproducerofnovelfoodproductstopredictallergenicityinanearlystageofproductdevelopmenttoavoidwithdrawalofthenovelfoodfromthefoodmarketafter introduction. For this reason, it is necessary toassess theallergenicpotentialofnovel foodsbeforeawell-informeddecisioncanbemadeontheallergenicpotentialofanovelfoodandtoguidetheimplementationofriskmanagementstoolssuchaslabelling.Riskmanagementaspectsarenotaddressedinthispaper.Inthispaper,thecurrentriskassessmentstrategyand guidelines will be discussed and a conceptual strategy is suggested,aimed to givebetter guidance in how to assess the allergenicity of novelfoodproteinsandproteinsources.2.CurrentstrategyandguidelinesAsalreadymentionedabovethereisnopredictiveandvalidatedmethodtoassess the allergenicity of novel proteins (sources) or protein containingproducts. In most recently filed novel food dossiers, parts of theallergenicityriskassessmentguideline forGeneticallyModifiedOrganisms(GMO) which was drafted in 2010 by the EFSA's Genetically ModifiedOrganisms(GMO)Panel[1]andupdatedin2011[2],wereused.Theinthisguideline suggested weight-of evidence approach (Fig. 1) involves anintegrated case-by-case approach to be used in the allergenicity riskassessmentofnewlyexpressedproteinsingeneticallymodified(GM)feedandfoods.Thesafetyevaluationmainlyfocusseson:1)Evaluationofthesourceofthegene2)Sequencehomologywithknownallergens3)BindingtoIgEfromallergicindividuals4)Stabilityoftheproteininapepsinresistancetest.2.1.SourceofthegeneAllergenicity assessment of GM food starts with the evaluation of thesource of the gene. If the source of the gene has a proven allergenicpotentialthenacarefulassessment ismandatorytoensurethatthegene
Chapter2
15
of interest does not encode for an allergen. The relevance of thisevaluation is apparent from the incidence that a GM soybean wasproducedthatcontainedagenefromBrazilnut.ThisGMsoybeanshowedallergenicreactionsinBrazilnutsensitiveindividuals[11].
Figure 1. Flow chart summarizing the Weight-of-evidence approach forallergenicityassessmentofnewlyexpressedproteinsinGMO.2.2.Sequencehomologytoknownallergen(s)Bioinformatic tools are used to compare the amino acid sequenceof thenewly expressed protein with the sequences of known allergens todetermine sequence homology. High sequence homology is associatedwith a high risk of a cross-reactive allergic reaction. FAO/WHO2001 andCodex2003recommendedthat35%sequenceidentitytoaknownallergenover a window of at least 80 amino acids is considered a minimalrequirement to regardaproteinallergenic innature [23].Thiscriterion isstillsupportedbytheEFSAGMOpanel[2].
Chapter2
16
2.3.IgEbindingtestsSpecificserumscreeningisrecommendedbyCodexandEFSAguidelinesincaseswherethesourceofthegene/proteincommonlycausesallergies,orwhen there is ahighdegreeof sequencehomologyof theprotein (>35%homology,see2.2)toaknownallergen.Inaspecificserumscreenbindingofthetransgenicproteinwithserafrompatientswithaclinicalfoodallergyto a specific allergen/food is tested todeterminewhether the transgenicproteinisnotcrossreactivewithaknownfoodallergen.2.4.PepsinresistancetestResistance to pepsin is proposed as a criterion for a protein to beconsidered as a potential allergen.However, it hasbeenestablished thatno absolute correlation exists [24-26] between pepsin resistance andallergenicityandthere isno internationallyacceptedprotocolavailabletoperform such in vitro digestibility tests. Improvement and good guidancefor the interpretation of pepsin resistance test and validation of the testwithallergensand(virtually)non-allergensiscurrentlyunderreviewoftheEFSA GMO panel (EFSA workshop June 17th Brussels,www.efsa.europa.eu/en/events/event/150617).2.5.AdditionaltestTheEFSAguidelinessuggestadditional test suchasTcellepitopescreensand animal models to be applied, once developed and validated.Unfortunately,novalidatedandpredictivemodelshavebeendevelopeduptillnowandnofurtherguidanceisgivenonhowtousethesemodels.Thetests and interpretation of outcomes of the aforementioned strategy aredifficulttoapplytonovelfoodproteinsandproteinssources.Especiallytheelements “source of the gene” and “homology testing” are difficult toperformwhenassessingtheallergenicityofcomplexproteinmixturesfromneworganisms.Thenwearenottalkingaboutaninsertionofjustonegenewitha known sequencebutaboutnewcomplexmixturesofhundredsorthousands of proteins for which, inmost cases, gene sequences are notknown. This makes comparison with known allergen sequences a verycomplex, time consuming and most likely a non-feasible process. IgEbindingtestsandpepsinresistancetestsarepartsof theassessmentthatmight be useful in a strategy for novel food proteins, however, moreguidanceonproceduresandinterpretationofoutcomes,particularlywhere“non-negative” findings are observed (e.g. some degree of pepsindigestion) isneeded.Furthermore, IgEbindingor IgEcross reactivitywith
Chapter2
17
cross reactive carbohydrate determinants or taxonomic related allergensdoesnotautomaticallymeanthatanallergicreactionwilloccur[20-22].Inaddition, other factors that might influence allergenicity, such asprocessingandmatrix,shouldbetaken intoaccountaswellashistoryofsafeuseandwherepossiblefoodchallenges.TheseitemsarelackinginthecurrentEFSAstrategy.Below,aframeworkissuggestedforallergenicityassessmentofnovelandmodified food proteins. Cross-reactivity can mostly be assessed usingcurrently available techniques and tests. Risks associated with de novosensitization can partly be assessed using currently available techniquesandtestsbutwillalsorequirenewapproachesyettobedeveloped.3.ConceptualstrategyThe need for an allergenicity assessment strategy for novel protein(sources)wasalreadymentionedbyGubeschandcoworkers,whoassessedtheallergenicityofthreenovelvegetables,namelywaterspinach,hyacinthbeanandEthiopianeggplant[27].Forthisassessment,theyusedathree-stepstrategy.The first stepwas toanalyze thepresenceofpan-allergensby immunoblot with specific animal antibodies. In the second step IgEbinding to the extracts of these vegetableswas testedby EAST (Enzyme-allergo-sorbenttest)andimmunoblotanalysisusingserawithIgE-reactivitytoknownpan-allergensortophylogeneticallyrelatedfoods.In the final third step the clinical relevanceof the IgEbindingwas testedusing SPT and open oral food challenge (OFC). This stepwise procedureseemed successful to confirm the presence of allergenic proteins in thevegetableextractsandtheir IgEbindingcapacity.Also, the invivostudiesshowedthepotentialofthevegetablestoelicitaclinicallyrelevantallergicreaction. Another important example is the allergenicity assessment ofNangainuts [28]. In this study, the relevanceofa foodchallengebecameveryclearsincenoneofthe12patientswhoshowedNangaisensitization(RAST,SPTorhistaminerelease),hadapositivefoodchallengewithNangainuts.Theaforementionedstrategiestogetherwiththeweight-ofevidenceapproach for foodsderived fromGMplantswasagoodstartingpoint forthe development of a generic food allergy assessment strategy for novelproteins(Fig.2).Thedifferentaspectsofthestrategyaredescribedinmoredetail below. It should be noted that elements of the scheme may beomittedonacasebycasebasiswhenthiselementisnotapplicable.
Chapter2
18
3.1.Productinformation3.1.1.HistoryofexposureA thorough investigation on exposure history can provide moreinformation on previous adverse effects of (dermal, respiratory and/ororal) exposure to the protein (source), for instance in an occupationalsetting.Also,historyofsafeuseinotherpartsoftheworldcanbehelpful,howeveritshouldbekeptinmindthatinfoodallergy,environmentalandgeographical dependent factors are important and that safe use in someparts of the world does not exclude allergenicity in others. (e.g. pollenrelated Rosacea fruits allergy) [14]. However, further guidance may beneeded on how much and which information on history of safe use isneededandhowtouseandinterpretthisinformation.3.1.2.TaxonomyandrelationshipSince knowledge on the allergenic potential of the novel protein sourcemight be scarce or not available at all, gathering information on thepossible allergenicity of biologically related species is requisite. For thispurpose, the phylogenetic tree or evolutionary tree can be used. Thephylogenetictreeisabranchingdiagramshowingtheinferredevolutionaryrelationships among various biological species or other entities and isbased upon similarities and differences in their physical or geneticcharacteristics. Relationships to known allergenic sources might giveindications for allergenic risks, based on cross reactivity, and thus theallergicpopulationsatrisk.However,taxonomicrelationshiponlydoesnotlead to conclusive evidence. For instance, serological cross reactivitybetweenfishspeciesisfrequent,butinasignificantproportionofpatients,clinicalrelevanceappearedtobelimitedtoonlycertainspecies[29].3.1.3.ProteinidentificationTheidentificationofproteinsinthenovelproteinsourcecanbehelpfultoassigns putative allergens in the novel protein source and thus provideuseful information to define the population at risk. For instance,identificationofa tropomyosinprotein inan insectextractmight indicatethat shrimp allergic patients might be at risk when eating insects, sincetropomyosin is the major allergen in shrimp and other crustaceans.Identificationofproteins istypicallyperformedbyLC-MS/MSanalysisanddatabasesearchesafterdigestionoftheproteinswithtrypsin.Theproteinsare separatedby reversedphase chromatographyprior to in-lineanalysisoftheirmassesandfragmentationpatternsinthemassspectrometer.The
Chapter2
19
masses of the parent ions and their fragments are used to searchdatabases of known protein sequences to match to in silico digestionpatternsoftheseproteins. Identificationofnovelproteins (notpresent inthedatabase) isbasedonhomologieswithknownproteinspresent inthedatabaseandthemorepeptidemassesthatmatchthepredictedmasses,themorecertainoneisofthelikelihoodthattheproteinisidentified.3.1.4.InformationonusageInformationonhowthenovelprotein(source)isintendedtobeusedinafoodproduct is importantfordeterminingwhichextractsand(processed)formsof theproducthave tobe tested in theallergenicityassessment. Isthenovelprotein (source)embedded inacertainmatrix (high fatorhighsugar) and is it processed (e.g. backing, frying)? So, what is the finalproductandformthatwillbeconsumed.Thisinformationisneeded,sincematrix and processing may have an effect on solubility, digestibility andallergenicity [30,31].Weknow, for instance, thatheatedapplehas lowerallergenic properties than raw apple [32] and heat processing has asignificantimpactonthedigestibilityofovalbuminfromegg[33].Anotherimportantissueistheintendedlevelofuse.Whatistheexpectedamountofproteinsthatwillbepresentinthefoodproductandhowoftenwillthisproductbeconsumed?Thisismeaningfulinformationthatcanbeusedtodetermine how many protein should be used in a food challenge todeterminedose-responsecurvesandthustheallergenicrisk.3.1.5.Researchmaterial:Extract(s)All information obtained in the previous sections is needed tomake thecorrect choices concerning the preparation of extracts to be tested (e.g.processedornot)ininvitroandinvivoassays,butalsowhichfoodproducthastobeused in foodchallengestudies.Theextractsandchallengefoodshouldbeagoodrepresentationofthesubstancesthatwillbepresent inthecommercial foodproduct.Matrixandprocessingmayhavean impacton the solubility of proteins, which may change due to aggregation orunfolding of proteins. Special attention has to be paid to the extractionbufferstoensurethatarelevantproteinsetistestedintheassessment.Inmoststudiesonlyonebuffer,mostlyaTRISorphosphatebufferisused.Asa consequence, only readily soluble proteins will be extracted from thefood product and in this way an incomplete protein panel is tested forallergenicity.Itisthereforerecommendedtouseforinstanceasequentialextraction procedure using in succession TRIS buffer, urea, and SDS/DTT
Chapter2
20
bufferaswasusedbyBroekmanetal.[34]andmentionedbytheILSIpanelon Processing and allergenicity [31]. Furthermore, identification of theproteinspresentinthedifferentextractsisrequisite.3.2.Cross-reactivitytesting3.2.1.IgEbindingstudiesTargeted IgE binding screens canbe used to identify putative allergen(s).Serumfromwell-characterizedallergicpatientsisneededforthistargetedserum screen. Based on the previously obtained information a smartselection of allergic patients that might be at risk can be made. Forinstance, sera from individuals previously sensitized againstphylogenetically related foods, (e.g. serum from shrimp allergic patients,when testing insects) can be used. Also, negative control group(s), serafrom non-phylogenetically related allergies should be used (e.g. peanut,when testing insects) to exclude non-specific IgE binding.When it is notclearwhichallergicindividualshavetobetested,apanelofseraobtainedfrom patients with different allergies/allergy profile can be used. Forexample; pollen (birch/grass, mite), plant (e.g. peanut, soy, tree nuts,wheat), animal (egg, milk, fish, crustaceans). Preferably individual serafrompatientswithawell-documentedallergyshouldbeusedrather than(pooled)seratoimprovethesensitivityofthetest.Dependingonthestudydesign it is important to consider the amount of patient sera to be usedand the selection of sera should be critically evaluated, since patientselectioncanhaveabigimpactontheoutcomeofthetest.IgEbindingcanbe tested using different techniques; ELISA, RAST, immunoblot etc.Immunoblot has the advantage over the other techniques, becausewiththis test more than one protein can be visualized simultaneously.Moreover, this technique gives more information on the presence ofdifferent allergenic proteins and differences between patients. Thedisadvantage is, that with SDS-PAGE, proteins will lose their naturalstructure (denaturing buffers) and thus false negative results can beobtained. Using another IgE binding test, for instance ELISA or RAST,simultaneouslyisthereforepreferred.3.2.2.FunctionalIgEtestingBindingofproteinswithIgEfromanallergicpatientmayindicatethatthenovelprotein(source)caninduceanallergicreactionintheallergicpatientunder investigation.However, IgEbinding as suchdoesnot automaticallyindicate that a clinically relevant reactionwill take place. Therefore, it is
Chapter2
21
necessary to test the clinical relevance of the in vitro IgE binding withfunctionalIgEtestingstrategies;suchasbasophilactivationtest(BAT)andskinpricktests(SPT).However,aDBPCFC,whichisregardedworldwideasthe‘goldstandard’infoodallergydiagnosis,ispreferable.Forallthesetestanapprovalofamedicalethicalcommittee isneededandthetesthastobeperformedinaspecializedclinicalandsafesetting.Alternatively,totheDBPCFCorasapre-screening,SPTandBATcanbeused.TheSPT isoftenused in clinical diagnosis of food allergy andwidely accepted. TheBAT isnotusedinroutinediagnosticsandthepredictivevalueisnotprovenyet,however the advantage over the SPT is that different protein extracts,including stringent buffers such as Urea and SDS/DTT can be used. Analternative for the BATwith human cells is the Rat Basophilic Leukaemiacells (RBL) assay. This in vitro assay uses rat basophilic leukaemia cellstransfectedwiththeFcεreceptortypeI,whichisprimedwithhumanIgE.However,somedisadvantagesofthisassayhavebeenreportedsuchaslowIgE receptor expression on RBL with respect to human basophils, whichleadtolowsensitivityoftheassayandtheneedforserawithahighratioofspecific/non-specificIgE.Unfortunately,onlyaminorityoftheserafromfoodallergicpatientswillmeetthisspecificitycriterion.3.2.3.IdentificationIgEbindingproteinsThe IgE binding proteins can be identified as described under proteinidentification after isolation of the proteins using for instance magneticbeads immobilizedwith IgEorexcisionofproteinbandsfromaSDS-PAGEgelafterdetectingtheIgEbindingbandsusingimmunoblotting.3.3.Sensitizingpotency(incaseofhistoryofsensitization)3.3.1.HistoryofuseAnother importantaspect intheallergenicityassessmentofnovelprotein(sources) isthepotencytosensitizeanindividualdenovo. Informationonthe history of use and exposure can be a starting point for this.Identification of work related allergic symptoms (inhalant or dermal)shouldbeassessedusingfor instancespecialquestionnaireswhenvisitingfacilities where the novel protein (source) is produced. Individualssensitizedorallergic to the sourceunder investigationcanbe included infurtherstudies.Furthermore,investigationoftheliteratureoncasereportsand use of the novel protein (source) in other geographically situatedcountriesisimportant.
Chapter2
22
3.3.2.IdentificationIgEbindingproteinsSera from the sensitized or allergic individuals can be used to identifyputativesensitizingproteins.Thisidentificationmighthelptodetermineifde novo sensitization may lead to new allergies or can lead to cross-reactive reactions with already known allergies. For instance, when anindividual isprimarysensitizedto insectproteins, is itpossibletobecomeallergicforhousedustmiteorshrimpaswell?3.3.3.CrossreactivitywithknownallergiesThiscrossreactivitycanbetestedinthesamemannerasdescribedaboveusingIgEserumscreensandIgEfunctionalitytesting.Theuseofinhibitionstudiesisrequisitetodetermineprimarysensitizationtothenovelproteinorcrossreactivitytoalreadyknownallergens.3.3.4.NewallergyWhether de novo sensitization may lead to a new food allergy can betested using functional IgE testing (as described above), preferably aDBPCFCwiththenovelprotein(source).3.4.Sensitizingpotency(nohistoryofsensitization)When no information on a history of sensitization is available, thepredictionofdenovosensitizationwillbedifficult,sincetherearecurrentlynotestsavailablethatcanpredictfordenovosensitization[35].However,some animalmodels that determine immunogenicity and allergenicity ofproteinsdoexist and thesemodelshavebeenused to try to identify theallergenicpotentialofproteins.Unfortunately,thesemodelshavenotbeenvalidated using the appropriate number of allergens and (virtually) non-allergenicproteins.Thesame is true for invitrocellbasedassayssuchasDCactivationorcytokine release fromTcells.Hence, it isnotpossible topredict, usingmethodologies available todate, the sensitizingpotencyofnovelprotein(sources).
Chapter2
23
Figure 2. Schematic overview of suggested allergenicity assessment strategy ofnovelproteinsandproteincontainingsources.4.ThewayforwardInthispaper,agenericstrategy issuggestedtoassesstheallergenicityofnovel protein (sources), based on tools and tests that are currentlyavailable.Withthisstrategy,itispossibletodeterminecross-reactivity/co-sensitization and thus whether and which part of the existing allergicpopulationisatrisk.Thestrategycanalsoaddressdenovosensitizationincase a history of sensitization to the novel protein source is known.However, the strategy is not applicable for the assessment of de novosensitizationwhennosubjectscanbefoundwithahistoryofsensitizationtothenovelprotein(source).Inallergenicityassessmentitcannotberuledout that novel or modified proteins in food may induce de novosensitization which may eventually give rise to new allergies. Theassessmentofdenovosensitization ishardlyornotcovered inanyofthe
Chapter2
24
previously mentioned guidelines and for this reason many safetyassessment dossiers submitted to the EFSA are lacking relevantinformation on this aspect. Therefore, there is an urgent need for astrategy that is capable of predicting the sensitizing potency of proteins.For the development of this strategy, the following key determinants forthedevelopmentofanallergicreactionmayneedtobeconsidered:a)thetiming,doseandrouteofproteinexposure(e.g.mucosalordermal)b)theintrinsicpropertiesofaprotein(e.g.physical/chemicalandbiologicalproperties)c) the context (e.g. lipids) inwhich theprotein is seenby the individual'simmunesystem(e.g.matrix/processing).Itcanbeenvisagedthatnosingletestisavailablethatisabletopredictthedenovo sensitizingpotencyofaprotein (source)andthusasetofassaysshouldbeconsideredandusedintheassessment.Theparameterssuchasmentionedundera,bandcor combinations thereof canbeused to findcorrelationsbetweenpropertiesofproteinsandtheirallergenicpotential.For this strategy, it is important to define how allergens can be rankedbasedon their allergenicpotential. To this end,one shoulddecidewhichcriteriashouldbeusedtoscaleapaneloflow/intermediate/highallergenicproteins.CurrentlyanILSIEuropeExpertGroupandTNOaredevelopinganallergenicity scaling system Prioritizing of allergenic foods according totheirpublichealth importance [36]. Furthermore, aCOSTActionnetwork(ImpARAS,www.imparas.eu)hasrecentlystartedtodiscusswithanout-of-the-boxview,new ideasandmorepredictive invivo, invitroand insilicomodels and approaches to improve the current allergenicity riskassessmentstrategy,withthefocusonsensitization.Athirdinitiativeisthesharedresearchprogram(SRP)Foodallergy,whichisinitiatedbyTNOandisacollaborationbetweenindustry,universitiesandTNO.Inthisprogram,the focus is on the development of a predictive allergenicity assessmentstrategy which is based on an allergenic scaling as described above andpredictivephysicalchemicalandbiologicalmarkers.Thisnewstrategymusthelp regulatory bodies to assess novel protein (sources) and improveallergymanagement.
EFSA Panel on Genetically Modified Organisms (GMO); Draft ScientificOpinion on the assessment of allergenicity of GM plants andmicroorganisms and derived food and feed. EFSA J. 2010: 8(7) (1700),1e168.
3. NwaruBI,HicksteinL,PanesarSS,RobertsG,etal.Prevalenceofcommonfood allergies in Europe: a systematic review andmeta-analysis. Allergy2014:69(8),992e1007.
4. Johnston LK, Chien KB, Bryce PJ, The immunology of food allergy. J.Immunol.192(6),2529e2534.
9. Radauer C, BublinM,Wagner S,Mari A, et al. Allergens are distributedintofewproteinfamiliesandpossessarestrictednumberofbiochemicalfunctions.J.AllergyClin.Immunol.2008:121(4),847e852e847.
10. BoyceJA,Assa'adA,BurksAW,JonesSM.Guidelinesforthediagnosisandmanagement of food allergy in the United States: report of the NIAID-sponsoredexpertpanel.J.AllergyClin.Immunol.2010:126(l),S1eS58.
14. Andersen MB, Hall S, Dragsted LO. Identification of european allergypatterns to the allergen families PR-10, LTP, and profilin fromRosaceaefruits.Clin.Rev.AllergyImmunol.2011:41(1),4e19.
Chapter2
26
15. Broekman H, Knulst A, Den Hartog Jager C, Gaspari, M, et al. 2015a.Shrimpallergicpatientsareatriskwheneatingmealwormproteins.Clin.Transl.Allergy2005:5(3),P77.
16. (EFSAPanelonDieteticProducts,N.a.A)EFSANDAPanel,2013.ScientificOpinion on the safety of “rapeseed protein isolate” as a Novel Foodingredient.EFSAJ.2013:11(10),3420e3443.
17. (EFSAPanelonDieteticProducts,N.a.A)EFSANDAPanel,2009.Opinionon the safetyof ‘Chia seeds (Salviahispanica L.) andgroundwholeChiaseeds’asafoodingredient.EFSAJ.2009:996,1e26.
18. (EFSA Panel on Dietetic Products, N. a. A) EFSA NDA Panel, 2010a.Scientific opinion on the safety of “Lentinus edodes extract” as a novelfoodingredient.EFSAJ.2010:8(7),1685e1700.
19. (EFSA Panel on Dietetic Products, N. a. A) EFSA NDA Panel, 2010b.Scientific opinion on the “safety of sardine peptide product” as a novelfoodingredient.EFSAJ.2010:8(5),1684e1701.
20. Mari A, Iacovacci P, Afferni C, Barletta B, et al. Specific IgE to cross-reactivecarbohydratedeterminantsstronglyaffect the invitrodiagnosisofallergicdiseases.J.AllergyClin.Immunol.1999:103(6),1005e1011.
21. Peeters KA, Nordlee JA, Penninks AH, Chen L, et al. Lupine allergy: notsimplycross-reactivitywithpeanutorsoy.J.AllergyClin.Immunol.2007:120(3),647e653.
22. Ibanez MD, Martinez M, Sanchez JJ, Fernandez-Caldas E. Legumecrossreactivity.Allergol.Immunopathol.Madr.2003:31(3),151e161.
23. Commission, C.A., 2003. Alinorm 03/34: Joint FAO/WHO Food StandardProgramme,CodexAlimentariusCommission,Twenty-fifthSession,Rome,Italy30Junee5July,2003.AppendixIII,guidelinefortheconductoffoodsafety assessment of foods derived from recombinant-DNA plants andAppendix IV, Annex on the assessment of possible allergenicity, pp.47e60.
24. BannonGA,GoodmanRE,LeachJN,RiceE,etal.Digestivestabilityinthecontext of assessing the potential allergenicity of food proteins.CommentsToxicol.2002:8(3),271e285.
25. MillsENC,MarshJT,BoyleR,Hoffmann-SommergruberK.etal.Literaturereview: ‘in vitro digestibility tests for allergenicity assessment’. EFSASupport.Publ.1e52(EN-529).
26. Moreno FJ. Gastrointestinal digestion of food allergens: effect on theirallergenicity.Biomed.Pharmacother.2007:61(1),50e60.
28. StenE,StahlSkovP,AndersenSB,TorpAM,etal.Allergeniccomponentsof a novel food,MicronesiannutNangai (Canarium indicum), shows IgEcross-reactivityinpollenallergicpatients.Allergy2002:57(5),398e404.
29. SchulkesKJ,KlemansRJ,KniggeL,deBruin-WellerM,etal.SpecificIgEtofish extracts does not predict allergy to specific specieswithin an adultfishallergicpopulation.Clin.Transl.Allergy2014:4,27.
30. MillsEN,SanchoAI,RigbyNM,JenkinsJA,etal.Impactoffoodprocessingon the structural and allergenic properties of food allergens.Mol.Nutr.FoodRes.2009:53(8),963e969.
32. Sancho AI, Rigby NM, Zuidmeer L, Asero R, et al. The effect of thermalprocessingon the IgE reactivityof thenon-specific lipid transferproteinfromapple,Mald3.Allergy2005:60(10),1262e1268.
33. TakagiK,TeshimaR,OkunukiH,SawadaJI.Comparativestudyofinvitrodigestibility of food proteins and effect of preheating on the digestion.Biol.Pharm.Bull.2003:26(7),969e973.
34. BroekmanH, Knulst A, denHartog Jager C,Monteleone F, et al. 2015b.Effectofthermalprocessingonmealwormallergenicity.Mol.Nutr.FoodRes.2015:59(9),1855e1864.
35. Dearman RJ, Kimber I. 2009. Animal models of protein allergenicity:potential benefits, pitfalls and challenges. Clin. Exp. Allergy 39 (4),458e468.
Chapter3:EffectofthermalprocessingonmealwormallergenicityAbstractScope: The growing world population requires the exploration of newsustainable protein sources to ensure food security. Insects such asmealwormarepromisingcandidates.Forsafetyreasons,ariskassessment,including allergy risks, is needed. Since allergenicity can be influenced bythermalprocessing,itishighlyimportanttotakethisintoaccount.Methodsand results: Freshmealwormwasheatprocessedandextractedby a sequential extraction method using in succession Tris, urea, and acombined SDS/DTT buffer. Extracts were tested using immunoblot,basophil activation test and skin prick test in 15 shrimp allergic patients,previously indicated as population at risk for mealworm allergy.Immunoblots showed a difference in IgE binding between processed andunprocessed mealworm extracts. However, this was due to change insolubility. Some allergenswere soluble in urea buffer, but becamemoresolubleinTrisbufferandviceversa.IgEbindingwasseenforallextractsinblot and basophil activation test. The results from 13 skin prick testsshowed a skin reaction similar between processed and unprocessedmealworm.Conclusion: Thermal processing did not lower allergenicity but clearlychangedsolubilityofmealwormallergens.Asequentialextractionmethodallowedforassessmentofabroaderproteinpanel.IntroductionA huge shortage of protein sources for human food consumption isexpected in the near future due to the growing world population [1].Sustainable protein sources are being explored to solve the coming foodinsecurity problem. The larvae of the yellowmealworm beetle (Tenebriomolitor)isagoodcandidateandisalreadyforsaleinGreatBritain,theUS,and in the major supermarkets in the Netherlands and Belgium [2,3].However,a thorough safetyassessment,and inparticularanallergenicityrisk assessment, is yet to be performed [4]. Allergenicity is not only atheoretical threat, since 0.1–5.7% of the pediatric and 0.1–3.2% of theadultEuropeanpopulationhasafoodallergy[5].Moreover,previouslywe[6]foundthat IgEfrompatientssensitizedtoshrimpandhousedustmite(Der p 10; closely related species), binds to mealworm proteins. The
Chapter3
32
relevant proteins were identified as the pan allergens tropomyosin andarginine kinase, which are major allergens in shellfish (e.g. shrimp andlobster).Allergenicity canbe influencedby factors suchasmatrix [7] andprocessing—for instance, by changing protein structure and thus IgE-bindingepitopes[8].Thiswaspreviouslyreportedforotherfoodssuchaspeanut, tree nuts, and apple [9–12]. Thermal processing by dry roastingenhanced allergenicity of peanut [12], while for tree nuts the allergenicpropertieschangedinsuchawaythatmostpollenallergicpatientsreactingto treenutshadnoclinical reactionaftereating theheatprocessed food[12,13]. Thus, processing may have an impact on the risk of getting anallergic reaction for mealworm. Since mealworm is closely related toshellfish one might expect that processing may alter the allergenicity ofmealworm proteins in a comparable manner to shellfish. For instance,Nakamura et al. [14] reported that thermal processing resulted in anenhanced IgE-binding capacity of scallop tropomyosin in dot blot andcompetitiveELISAusingserumfromscallopallergicpatients.Thisenhancedcapacity was suggested to be a result of glycation between free aminoacids and aldehydeor ketone groups of sugars during heating. The samegroup found an opposite result after Maillard reaction with squidtropomyosin [15]. Samson et al. [16] found no significant differencebetweenboiledandrawshrimpextractusingimmunoblotwithserumfromshrimp allergic patients. However, inter individual differences in proteinrecognitionwereobserved.Carnesetal.[17]reportedthatboiledextractsofshrimpandlobsterhadhigherIgE-bindingcapacityinELISAandrecordedgreater skin reactivity in skin prick test (SPT). A similar finding wasobserved by Liu et al., [18] when testing shrimp tropomyosin. Takentogether, the results in the above-mentioned papers are to some extentcontradicting. This could be due to solubility issues. Therefore, moreattention should be paid to the preparation of extracts to ensure thepresence of a representative set of proteins for allergenicity assessment.Most studies reported the effect of processing using immunoblot andELISA.Unfortunately, thesemethods lack informationonthefunctionalityof IgE binding, which canbemeasured using SPT and basophil activationtest(BAT).ThesetestsarethereforepreferredoverimmunoblotandELISAin allergenicity assessment. However, they cannot replace foodchallenges—the“gold”standard. Immunoblot,BAT,andSPTwereused inthis study to test the effect of processing on mealworm allergenicity.Shrimpallergicpatientsweretestedduetothelackofasufficientnumberofmealwormallergicpatients.Toensurethatmostrelevantproteinswere
Chapter3
33
covered, a sequential protein extraction method was used and thepresenceofallergenswasconfirmed,usingnanoLC–MS.MaterialsandMethodsPatientselectionandscreeningThree sera frompatients diagnosedwith shrimp allergy at theUniversityMedical CentreUtrecht, theNetherlands,wereused to test the effect ofprocessingonproteinsolubilityusingimmunoblot.Forallergenicitytesting,15 adult patients diagnosed with shrimp allergy, based on suggestivehistory and sensitization were included. All patients reacted positive tomealwormproteininSPTandserology.AllpatientsgaveinformedconsentbeforeansweringthequestionsandfortheperformanceofSPTandbloodcollection.Thestudywasapprovedbythelocalethicscommittee.ThermalprocessingofmealwormRaw and freeze dried Yellowmealworms in final larval stagewere kindlyprovided by Dutch insect farm Kreca (Ermelo, the Netherlands). Rawmealworms(50g)wereheatprocessedbyvariousmethods:Blanchingfor1minat100°C,boilingin300mLwaterfor10minat100°C,bakingfor3.5minat1000Wattonaninductioncooker(PrimaDonnaDonnatsi-199k),orfrying for 30 s at 180°C in peanut oil. All processed and unprocessedmealwormswerestoredat−20°Cuntilfurtheruse.MealwormextractpreparationFivegramsofraw,freeze-dried,andprocessedmealwormswereextractedusing a sequential protein extraction method (see Fig. 1). First, themealwormsweremixedwith25mLice-coldTrisbuffer(20mMTrisbufferpH 7.6 containing 1 mM phenylthiocarbamide (Sigma Aldrich) and HaltProtease InhibitorCocktail (ThermoScientific)).Theamountofmealwormwas corrected for weight gain or weight loss due to processing.Subsequently,themealwormsweredisrupted,usinganultraturrax(3×10s) under continuous cooling. After centrifugation (30min, 15 000 × g at4°C), the supernatant was recovered. The insoluble residue was washedonce with 5 mL Tris buffer (as described above). The 30 and 5 mLsupernatants were combined. Twentyfive milliliter was used for samplecleanupandconcentrationusingTCAprecipitation.Second,theremainingpelletwasextractedovernightat4°Cwith30mLureabuffer(6Mureain20mMTrisbufferpH7.6containing1mMphenylthiocarbamideandHaltProteaseInhibitorCocktail).Thesamplewassubsequentlycentrifugedand
Chapter3
34
the supernatantwas collected. The pelletwaswashedoncemorewith 5mLureabuffer, centrifuged,and the supernatantwascombinedwith the30 mL urea supernatant. Twentyfive milliliters of the extract was TCAprecipitated. Tris and urea extracts were combined (1:1) for the BAT.Finally, the insoluble residue was almost completely dissolved at roomtemperaturein20mLSDS/DTTbuffer(20mMTrispH7.6,2%SDS,and1%DTT) and the supernatant was collected after centrifugation. All TCAprecipitated samples were redissolved in 6M urea buffer and stored at−20°Cbeforefurtheruse.ProteinconcentrationwasdeterminedusingtheBradfordmethod(Bio-Rad,Hercules,CA,USA).Figure1.
PreparationschemeofmealwormextractsSDS-PAGEgelofprocessedmealwormextractsForSDS-PAGE, theCriterionsystemwitha10–20%ReadyGelTris-HClgel(Bio-Rad) was used according to the manufacturer’s instructions. Allmealworm extracts (10 µg per sample) were loaded on the gel underreducing conditions (Laemmli buffer). After protein separation, theproteins were visualized using Coomassie-staining (Instant Blue, Expedi-ton,UK).
Chapter3
35
ImmunoblotwithserumofshrimpallergicpatientsAllmealwormextractswereappliedontheSDS-PAGEasdescribedaboveand transferred to a polyvinyldifluoride membrane using the CriterionBlottersystem(Bio-Rad)accordingtothemanufacturer’sinstructions.Themembranewasblockedovernightwith3%BSAandincubatedfor1hwithserum from a shrimp allergic patient (1:50) in PBS with 0.1% Tween 20containing3%BSA (PBST).After thoroughwashing, themembraneswereincubatedfor1hwithGoatantihumanIgE(KPL,Gaithersburg,MD,USA)1:100000inPBST.Afterwashing,thebandswerevisualizedusingachemi-luminescent peroxidase substrate kit ECL (Sigma) according to themanufacturer’s instructions.BlotswerescannedusingtheChemidocXRS+imagescannerwithImagelabsoftware(Bio-Rad).ProteinidentificationandquantificationusingNanoLC-MS/MSTrypsinDigestionextracts(TRISandurea)Extracts(50μgprotein)weresubjectedtoconventional in-solutiontrypticdigestion as previously described [6]. After reduction and alkylation, theproteinsweredigestedwithtrypsin(enzyme:substrateratioof1:25w/w)overnight at 37°Cwith agitation. Peptidemixtureswere desalted by C18StageTips, fabricatedbyusingC18disks(3M,Neuss,Germany),andusedaccordingtotheoriginalprotocol[19].Briefly,1of10ofeachtrypticdigestsolution was diluted fivefold in 0.1% TFA (solution A), and applied ontoStage Tips, which were previously conditioned with 10 μL of solution B(0.1%formicacid,50%acetonitrile)followedby10μLofsolutionA.Aftersampleloading,StageTipswerewashedwith10μLofsolutionA.Peptideelutionwasachievedbyadding8μLofsolutionB.Purifiedpeptideeluateswere diluted tenfold in mobile phase A (see below) and used for massspectrometricanalysis (0.5%of theoriginal sample foreachpreparation).Three technical replicates of C18 purification and mass spectrometricanalysiswereinjectedforeachsample.Trypsindigestionpellets(SDS/DTT)Pelletsobtainedafterureaextractionandcentrifugationweredissolvedin200μLlysisbuffer(100mMTrispH7.6containing4%SDSand0.1MDTT),incubatedfor5minat95°Candsonicated.Onceclarified,eachsampleformass spectrometric analysis was subjected to filter-aided samplepreparation [20], using a 30 kDa Microcon filtration unit (Millipore).Peptides were recovered by centrifugation at 14 000 g; followed by anadditional washing step to mobilize the peptides retained by the
Chapter3
36
membraneinthefiltrationunit,using50μLNaCl0.5M.Flow-throughwaspooled, desalted by C18 Stage Tips (as described previously) andsubsequently injectedformassspectrometricanalysis(0.5%oftheoriginalsample, corresponding to 400 ng of proteins). Three technical replicateswereinjectedforeachsample.NanoLC-MS/MSanalysisanddatabasesearchThepeptidemixturewasanalyzedaccording toVerhoeckxetal. [6],withsmall changes.ChromatographywasperformedonanEasyLC1000nanoscale liquid chromatography (nanoLC) system (Thermo Fisher Scientific,Odense,Denmark).TheanalyticalnanoLCcolumnwasapulledfusedsilicacapillary, 75μm id, in-housepacked toa lengthof10 cmwith3μmC18silica particles fromDr.Maisch (Entringen, Germany). Fourmicroliters ofthepeptidemixtureswasloadedat500nL/mindirectlyontotheanalyticalcolumn. A binary gradientwas used for peptide elution.Mobile phase Awas0.1% formicacid,2%acetonitrile,whereasmobilephaseBwas0.1%formic acid, 80% acetonitrile. For both types of analysis, that of extractsandthatofpellets,gradientelutionwasachievedat350nL/minflowrate,and ramped from 8 to 35% B in 60 min, and from 30 to 100% B inadditional8min;after5minat100%B,thecolumnwasre-equilibratedat0%Bfor2minbeforethefollowinginjection.MSdetectionwasperformedon a quadrupole-orbitrap mass spectrometer Q-Exactive (Thermo FisherScientific, Bremen, Germany) operating in positive ionmode, with nano-electrospray(nESI)potentialat1800Vappliedonthecolumnfrontendviaateepiece.Data-dependentacquisitionwasperformedbyusingatop-12methodwithresolution(FWHM),AGCtarget,andmaximuminjectiontime(ms) for full MS and MS/MS of, respectively, 70 000/17 500, 106/105,50/60. Mass window for precursor ion isolation was 1.6 m/z, whereasnormalized collision energy was 25. Ion threshold for triggering MS/MSeventswas2×104.Dynamicexclusionwas30s.DatawasprocessedusingProteomeDiscoverer1.3(ThermoFisherScientific),usingSequestassearchengine, and the Swiss Prot database accessed on February 2013 assequence database (3 123 840 sequences for Metazoataxonomy). Thefollowing search parameters were used: MS tolerance 15 ppm; MS/MStolerance0.02Da;fixedmodificationscarbamidomethylcysteine;enzymetrypsin;maximummissed cleavages1; taxonomyMetazoa. Search resultswere filtered by q values using Percolator integrated in ProteomeDiscoverer,toachieveapeptidelevelFDRoflessthan1%.
Chapter3
37
RelativeproteinQuantificationAlabel-freeapproachwasadoptedforrelativequantificationofallergens,using five unique peptides for arginine kinase and three unique peptidesfor tropomyosin. Peak areas for each peptide were calculated usingextracted ion chromatograms (XICs) via the Xcalibur software (ThermoFisher Scientific). Peak areas for each peptide were subsequentlynormalized using the total peptide-spectrum matches (TPSM) of thecorrespondingLC–MS/MSanalysis.TheTris freeze-driedsample,onewiththe highest TPSM, was chosen to confirm linearity between injectedamount and TPSM. Triplicatemeasurements of peak areawere averagedforeachpeptideandexpressedasrelativevaluecomparedtotheaveragearea of the same peptide in the Tris unprocessed sample. Relativequantification at theprotein levelwas achieved for all proteinsby takingthemedianvalueofallassociatedpeptides.Basophilactivationtest(BAT)usingshrimpallergicpatientserumBATwas performed as described byMeulenbroek et al. [21] withminormodifications.Cellswere incubatedwithadilution series (1:107–1:102)ofprocessed andunprocessedmealwormextracts (combinedTRIS andureaextracts (5mg/mL)andSDS/DTTextracts (noconcentrationdetermined)).Shrimp extract (ALK), 2mg/mL, and shrimp tropomyosin Pen a 1 (IndoorBiotechnologies), 1mg/mL,wereused aspositive controls. CD63, CD123,andCD203cexpressionwasanalyzedbyflowcytometryusingFACSCantoIIandFACSDivasoftware(BDBioscience,USA).Theresultswereexpressedas a percentage of CD63+ basophils. Basophils of two patients did notrespond in repetition to any of the extracts, nor to the positive control.Basophils of a third patient showed spontaneous release of CD63 on thenegativecontrol.Thesethreepatientswerethereforeexcluded.SkinPricktest(SPT)withprocessedmealwormextractsSPT solutionsof theprocessedandunprocessedmealworms (0.4mg/mL)were kindly provided by ALK (ALK-Abello, Spain). These solutions werepreparedinPBS,whichhasmoreorlessthesameextractioncharacteristicsas the Tris buffer mentioned above. The solutions were applied on theflexor aspect of the forearm using 1 mm tip lancets (ALK). Histaminedihydrochloride10mg/mLandglyceroldiluentwereusedaspositiveandnegative controls, respectively. SPT reactivity was recorded after 15minandmeasuredastheratioofthemeanofthewhealelicitedbythetestedextract and histamine control. When the ratio was 0.5 or greater, the
Chapter3
38
reactionwasregardedaspositive.Nostatisticaltestswereperformedduetothelimitedsizeofthegroup. ResultsHeatprocessingchangessolubilityIt canbeconcluded fromFig.2 thatproteinprofiles significantlychangedafterheatprocessinginalltestedextracts(Tris,urea,andSDS/DTT).BandsofproteinsfromtheTrisextractswithMW<25kDaandat±50kDaweremore intense inallheatprocessedextractscomparedtotheunprocessedextracts (rawandfreeze-dried). Incaseoftheureaextract,proteinbandswith a MW of ±40kDa were more pronounced in unprocessed extracts,whereas bands near 45 and 50 kDa were more pronounced in all heatprocessed extracts. In the SDS/DTT extract the same band at ±45 kDadiminishesafterheating,whileabandappearsat±37kDa.Inaddition,highmolecular weight proteins (70–200 kDa) were detected in the SDS/DTTbuffer extracts after heating. These changes in protein profiles were theresultofchangesinsolubility,asshownbytheLC–MSanalysis.Figure2.
Coomassie stained SDS-PAGE gels of extracts from processed mealworms. TRIS,urea and SDS-DTT extracts were prepared from raw, freeze-dried, blanched,boiled,bakedandfriedmealworm.LC–MS analysis of processed and unprocessed Tris, urea, and SDS/DTTextracts identified a wide range of proteins in mealworm. Putative
Chapter3
39
mealwormallergens(e.g.tropomyosin,argininekinase,myosinlightchain,and triosephosphate isomerase), identified in Tris and urea extract werepreviously reported by us [6]. However, in this study we also identifiedputativeallergensintheSDS/DTTextract(Table1).The most dominant protein in the SDS/DTT extract was arginine kinaseafter heat processing. The concentrations of these putative mealwormallergenswere different in the tested extracts. It can be concluded fromFig.3thatprocessingcausesashiftinsolubilityfromTristoureaandviceversa.Forinstance,argininekinase,whichwasabundantinrawandfreezedriedmealwormTrisextracts,wasalmostundetectable inheatprocessedmealwormTrisextract.However, itbecamedetectable inureaextractsafterheatprocessing.Fortropomyosin,theoppositeeffectwasfound.ThesolubilityoftropomyosininTrisbufferimprovedafterheatprocessing.Thesamephenomenonwasseen forotherallergenssuchasmyosin lightchain,which behaved in a similarmanner to tropomyosin on processing.Triosephosphate isomerase, anotherallergen,behaved similar toargininekinase.However,quantificationwasdifficult,duetodetectionlimits(datanotshown). Inconclusion,solubilityofmealwormallergenschangedafterheatprocessing.HeatprocessingdoesnotobviouslychangeIgEbindingcapacityThe above-mentioned change in allergen solubilitywas confirmed by theimmunoblotofthethreeshrimpallergicpatients(Fig.4).Theimmunoblotsshowed IgE binding tproteins in all testedmealworm extracts, Tris, urea,and SDS/DTT. In the lane of raw and freeze-dried Tris extract a proteinbandat±40kDacanbeseen,whichwaspreviously identifiedasargininekinase [6]. After processing, this band becomes more pronounced at aslightlyhigherMW.ComparingtheseresultswiththeLC–MSanalysis(Fig.3), thisband ismost likely tropomyosin (±37kDa). In the laneof rawandfreeze-driedureaextract,a±37kDaband(previouslyidentifiedafterin-geldigestion, as tropomyosin [6]) was detected. The intensity of this banddiminishedafterprocessing.Thedeclineof tropomyosinband intensity intheureaextractisinaccordancewiththeLC–MSdata(Fig.3).Estimationofthe overall effect of processing on IgE-binding capacity is difficult due tothis shift insolubility.For this reason,apoolofTrisandureaextractwasusedforIgEcross-linkingfunctionalitytesting.HeatprocessingdoesnotchangeIgEcross-linkingfunctionality.
Chapter3
40
Fromthe12usefulBATs,11showedactivationafterincubationwithboththemixedTris/ureaextractsaswellastheSDS/DTTextracts,indicatedbyan elevation in the percentage of CD63+ cells (Fig. 5). Basophils of onepatient reactedsolely toproteins in theSDS/DTTextract.Overall, activityof basophils to processed or unprocessed mealworm proteins was notclearlydifferent.However,basophils from threepatientswere somewhatmore strongly activated by the processed Tris /ureamealworm extracts,thanbytheunprocessedmealwormextracts.Table1:ProteinsidentifiedinSDS/DTTextractusingLC-MS-MS
Top 15 proteins identified by LC-MS/MS in the SDS/DTT fraction in theunprocessed and processed extracts. Arranged on highest mean score of 3measurements. Sequence coverage, Peptides identified and PSM are given as ameanof 3measurements. Identificationwasbasedonhomologywithmetazoanproteins intheSwissProtdatabase.Proteinsarenoted inboldwhenassignedasallergen by the IUIS allergen nomenclature subcommittee. PSM = peptide-spectrum matches, value that represents the number of MS/MS spectra thatmatchedpeptidesequencesassignedtothatparticularprotein.Score=thesumofindividualSequestscoresofalltheidentifiedpeptideswhichwereassignedtotheproteinitself.Thescoreistheprobabilitythattheobservedmatchisnotarandomevent.
Chapter3
42
Figure3.
LC-MS analysis of tropomyosin and arginine kinase in processed mealwormextracts (raw, freeze-dried, blanched, boiled, baked and fried, respectively). Theresults are presented as mean of three LC-MS analyses and calculated as ratiorelativetotheamountintherawTRISextract.Figure4.
Immunoblot of processedmealwormswith serum fromone of the three shrimpallergic patients. Themealwormswere extractedwith a TRIS, urea and SDS-DTTbufferrespectively
Chapter3
43
All15patientsshowedapositiveskinreactiontoshrimp,housedustmite,andmealwormextract. Inadditiontounprocessedmealworm,allshowedapositiveskinreactiontoblanched,boiled,baked,andfriedextracts(Table2).However,someinterindividualdifferenceswereseeninskinreactivity.Two patients had an increased skin reaction to processed extracts. Thewheal size increased from2+ to 3+ fromunprocessed to processed. Skinreaction of one patient decreased by blanching and of one patient byfrying. Overall, SPT reactions were comparable between all processedextractsin13of15patients.Figure5.
Basophil activation test with extracts (left: pooled TRIS and urea extract, right:SDS/DTT extract) from processed mealworms (freeze-dried, fresh, blanched,cooked, baked and fried respectively). Maximum % CD63+ basophils werecalculatedwithrespectto@IgEpositivecontrol.Eachlinerepresentsonepatient.
Chapter3
44
Table 2. Skin prick test results, expressed as a ratio of histamine control, usingextracts from the different processedmealworms (raw, blanched, boiled, bakedandfried)in15shrimpallergicpatients. No. Sex Age(y) Raw Blanched Boiled Baked Fried1 F 46 3+ 3+ 3+ 3+ 2+
2 F 23 2+ 3+ 3+ 3+ 3+
3 M 69 2+ 2+ 3+ 2+ 2+
4 M 45 2+ 2+ 2+ 3+ 2+
5 F 27 3+ 2+ 3+ 3+ 3+
6 M 19 2+ 3+ 2+ 2+ 2+
7 F 60 2+ 2+ 2+ 2+ 2+
8 M 30 2+ 3+ 3+ 3+ 2+
9 M 27 2+ 2+ 2+ 2+ 2+
10 F 47 2+ 2+ 2+ 2+ 2+
11 F 52 0 0 1+ 1+ 1+
12 M 26 2+ 2+ 2+ 2+ 2+
13 M 34 2+ 2+ 3+ 2+ 3+
14 F 23 2+ 2+ 2+ 2+ 2+
15 M 46 2+ 2+ 3+ 3+ 3+
Mmale,Ffemale,MeanSPTasaratioofhistaminecontrol(3+)DiscussionFrom the results obtained in this study it can be concluded that heatprocessinginfluencesproteinsolubility.Someproteinsbecamelesssolublein Tris buffer due to heat-induced denaturation but these proteins couldstill be solubilized in a chaotropic reagent such as urea (arginine kinase).Otherproteins thatundernatural conditionswere insoluble inTrisbufferbecame more soluble after heating (tropomyosin). Furthermore,processing did not lower IgE-binding capacity and IgE cross-linkingfunctionalityofmealwormallergens(e.g.tropomyosin,argininekinase).Arepresentative panel of proteins was assessed due to the use of asequentialextractionmethod.Tothebestofourknowledge,thisisthefirststudytoassesstheeffectofthermalprocessingonmealwormallergenicity.Furthermore, thesequentialproteinextractionmethodused in this studyhas,asfarasweknow,neverbeenusedtoassesstheeffectofprocessingonallergenicity.Heat processing strongly changed the solubility characteristics ofmealworm proteins. Change in allergen solubility, might be caused by
Chapter3
45
changes in 3D structure of the proteins after heat treatment. Someproteinswilllose,irreversibly,theirfunctionalpropertiesandsolubilityandform aggregates due to denaturation, while others may have increasedsolubility. Tropomyosin is amuscle protein, which forms a complexwiththeinsolubleactinandtroponinandisheatstableasaresultof itscoiledcoilhelicalconstruction[22].Theimprovedsolubilityismostprobablyduetobreakageofinteractionswiththeseotherdifficulttosolubilizeproteins.However, no evidence could be found in literature to corroborate this.Anotherpossibility is the formationofsolubleaggregates,whichwasalsodemonstratedfortheJapanesecedarpollenallergenCryj1byAokietal.[23]. Moreover, Usui et al. [24] showed that heat processing of purifiedtropomyosin from shrimp did not induce the formation of insolubleaggregates.However,differenceinsolubilityinPBSbufferbetweenheatedand unheated tropomyosin was not in agreement with our results. Thismightbeduetofactthatheatprocessingoftropomyosinwasnottestedinits natural environment (complex with actin) and thus breakage ofinteractions with other proteins cannot be demonstrated. In contrast,argininekinaseisaglobularprotein,whichtendstounfoldduringheating,exposinghydrophobicaminoacids,whicharenormally insidetheprotein.The exposed hydrophobic amino acids from different molecules willinteract in such a way, that formation of larger protein aggregates willoccur [25]. Cross-linking of arginine kinase may also be caused bypolyphenoloxidase-mediatedcross-linking.Inmostcasestheseaggregatesbecomeinsoluble.Furtheraggregationofglobularproteinsduringheatingisfavoredthroughtheformationofdisulphidebridges.Tosolubilizetheseaggregatesamore stringentbuffer isneeded,which confirmsourfindingthat arginine kinase was not detected in the Tris buffer after heating.Anotherpossibility for the LC–MSdetectionof tropomyosin inTrisbufferafterheatprocessingistheimproveddigestibilityafterheating.Thiseffectwas also seen in a study from Takagi et al. who showed that thermaltreatment markedly increased the digestibility of ovalbumin [26]. This isbecauseovalbuminisaglobularproteinthatunfoldsduringheating,whichexposesaminoacidsequencesthatcanbehydrolyzedbytrypsin.However,for mealworm tropomyosin this improved digestibility after heating wasnot confirmedby the immunoblot. Furthermore, enhancement of trypsindigestion isnotexpected for tropomyosinbecauseof itshelical structurethat upon heating will not suddenly expose different amino acidsequences.Therefore,itismorelikelythatextractabilityandthussolubilityis themain reason for thedifferencebetween theunprocessedandheat
Chapter3
46
processed samples instead of improved digestibility. When testing onlyprotein extracts prepared in Tris buffer, which is the routine procedure,one could wrongly conclude that IgE-binding capacity to tropomyosinwouldbeelevatedduetoheatprocessing[17].However,thecorrectionisthat possibly important allergens are overlooked when using just onebuffer type. This was demonstrated by our immunoblot data and mightalso be the case in some studies [16–18]where induction of IgE bindingafter heat processing was observed. In these studies, only PBS extracts,whichisanondenaturingextractionbuffersimilartotheTrisbuffer,wereused.Solubilityissuesareoftenencounteredwhenproteinsareprocessed.Inmostcasesthisphenomenonisnotrecognizedsincethecompositionoftheproteinextracts isnot identified [27]. IgEbindingon the immunoblotwasdetected inall testedextracts, indicatingthat inmealwormtherearemore allergens present than the ones identified in Tris buffer. Themostdominant IgE-binding proteins in the Tris extract and inthe urea extractwere identified as tropomyosin and arginine kinase, respectively, whichconfirmsourpreviousfindings[6]. IntheSDS/DTTextract,argininekinase(±40kDaband)wasalso identifiedand inaddition IgEbindingtoproteinswithahighermolecularweightwasdetected.ThehighMWproteinsthatwereidentifiedintheSDS/DTTextractsbyLC–MS/MS,weremyosinheavychain, paramyosin, and hemocyanin. It is not clear if these are the sameIgE-binding proteins as detected in the immunoblot. However, myosinheavychain and paramyosin were recently identified as shrimp allergens[28–30].Moreover, paramyosin and hemocyanin are included in the IUISdatabase as arthropod allergens. Sincemealwormand shrimpare closelyrelateditcanbeenvisionedthatparamyosinandmyosinheavychaincouldbe potential mealworm allergens. Another option is that the highmolecular weight proteins are the result of arginine kinase cross-linking.AccordingtoLC–MSidentificationargininekinasewasalsodetectedintheSDS/DTT extracts especially after heat processing. Processing did notchangeIgEfunctionalityinBATandSPT.TheadvantageofBAToverSPTisthatBATallowstestingofextractspreparedwithstringentbufferssuchasurea and SDS/DTT, while due to patient safety SPT only allows proteinextractspreparedaccordingtoclinicalguidelinesinsterilePBSbuffers[31].TheresultsfromtheBATindicate,besidessomeinterindividualvariabilityin basophil activation, no significant effect due to heat processing.Processing showed only induced basophil activation in three patients. InSPT,13of15patients,didnotreactdifferentlytotheprocessedextracts.Only twopatients (other than those inBAT) showeda trendof increased
Chapter3
47
skin reaction. This might be caused by the increased solubility of someallergensinPBSafterheatprocessing,whichmightalsobethecaseinthestudy of Nowak-Wegrzyn et al. [32]. The authors reported that boiledshrimp extract induced larger skin response compared to raw shrimpextractinsomeshrimpallergicpatients.Togethertheseresultsstrengthenthe need for different extraction buffers to assess the allergenicity of abroad representative protein panel. The strength of this study was thecombined use of clinical,ex vivo and in vitro tests in combinationwith asequentialextractionmethodandLC–MSanalysis.Thisallowedinclusionofa broader panel of mealworm proteins in the allergenicity assessment,than usually studied. In conclusion, heat processing did not lower theallergenicity of mealworm proteins, but clearly changed the solubility oftheseproteins.A sequential extractionmethodallowed for inclusionof abroaderproteinpanelintheallergenicityassessmentofmealworm.
Chapter3
48
References1 Tilman D, Balzer C, Hill J, Befort BL. Global food de-mand and the
sustainable intensification of agriculture.Proc.Natl. Acad. Sci. USA2011,108,20260–20264.
5. NwaruBI,HicksteinL,PanesarSS,RobertsG,etal.Prevalenceofcommonfood allergies in Europe: a systematicreview and meta-analysis. Allergy2014,69,992–1007.
7. Schulten V, Lauer I, Scheurer S, Thalhammer T. et al., Afood matrixreduces digestion and absorption of food aller-gens in vivo. Mol. Nutr.FoodRes.2011,55,1484–1491.
13. WormM,HompesS,FiedlerEM,IllnerAK,etal. Im-pactofnative,heat-processed and encapsulated hazelnutson the allergic response inhazelnut-allergicpatients.Clin.Exp.Allergy2009,39,159–166.
14. Nakamura A, Watanabe K, Ojima T, Ahn DH, et al. Effectof maillardreaction on allergenicity of scallop tropomyosin. J. Agric. Food Chem.2005,53,7559–7564.
15. NakamuraA,SasakiF,WatanabeK,OjimaT,etal.Changesinallergenicityand digestibility of squidtropomyosin during the Maillard reaction withribose.J.Agric.FoodChem.2006,54,9529–9534.
Chapter3
49
16. Samson KT, Chen FH,Miura K, Odajima Y, et al. IgEbinding to raw andboiled shrimp proteins in atopic andnonatopic patients with adversereactionstoshrimp.Int.Arch.AllergyImmunol.2004,133,225–232.
17. Carnes J,FerrerA,HuertasAJ,AndreuC,etal.Theuseof raworboiledcrustacean extracts for the diagnosis ofseafood allergic individuals.Ann.AllergyAsthmaImmunol.2007,98,349–354.
18. Liu GM, Cheng H, Nesbit JB, Su WJ, et al. Effectsof boiling on the IgE-binding properties of tropomyosin ofshrimp (Litopenaeus vannamei). J.FoodSci.2010,75,T1–T5.
19. Rappsilber J, Mann M, Ishihama Y. Protocol for micro-purification,enrichment, pre-fractionation and storage ofpeptides for proteomicsusingStageTips.Nat.Protoc.2007,2,1896–1906.
20. WisniewskiJR,OstasiewiczP,MannM.HighrecoveryFASPappliedtotheproteomic analysis of microdissectedformalin fixed paraffin embeddedcancer tissues retrievesknown colon cancer markers. J. Proteome Res.2011,10,3040–3049.
21. MeulenbroekLA,deJong,R.J,denHartogJagerCF,MonsuurHNetal.IgGantibodiesinfoodallergyinflu-enceallergen-antibodycomplexformationand binding to B cells: a role for complement receptors. J. Immunol.2013,191,3526–3533.
22. Hillberg L, Zhao Rathje LS, Nyakern-Meazza M, Helfand B, et al.Tropomyosins are present in lamellipodia ofmotilecells.Eur. J. Cell Biol.2006,85(5),399–409.
24. Usui M, Harada A, Ishimaru T, Sakumichi E, et al. Con-tribution ofstructural reversibility to the heat stability of thetropomyosin shrimpallergen.Biosci.Biotechnol.Biochem.2013,77,948–953.
25. YuHL,RuanWW,CaoMJ,CaiQF,etal.Identifica-tionofphysicochemicalproperties of Scylla paramamosainallergen, arginin kinase. J. Sci. FoodAgric.2013.93,245–253.
26. Takagi K, TeshimaR,OkunukiH, Sawada J. Comparativestudyof in vitrodigestibility of food proteins and effect ofpreheating on the digestion.Biol.PharmBull.2003,26,969–973.
27. Kamath SD, Rahman AM, Voskamp A, Komoda T, et al. Effect of heatprocessing on antibody reactivity toallergen variants and fragments ofblack tiger prawn: a com-prehensive allergenomic approach.Mol. Nutr.FoodRes.2014,58,1144–1155.
28. KhanaruksombatS,SrisomsapC,ChokchaichamnankitD,PunyaritP,etal.Identification of a novel allergen frommuscle and various organs inbanana shrimp (Fennerope-naeus merguiensis). Ann. Allergy AsthmaImmunol.2014,113,301–306.
Chapter3
50
29. SuzukiM, ShimizuK, Kobayashi Y, Ishizaki S, et al. Paramyosin from thediscabaloneHaliotisdiscusdiscus.J.FoodBiochem.2014,38(4),444–451.
30. GiuffridaMG,VillaltaD,MistrelloG,AmatoS,etal.Shrimpallergybeyondtropomyosin in Italy: clinical rele-vance of Arginine Kinase, sarcoplasmiccalcium binding pro-tein and hemocyanin. Eur. Ann. Allergy Clin.Immunol.2014,46,172–177.
31. Dreborg SF, Frew A. Position paper: allergen standard-ization and skintests.Allergy1993,48,49–54.
Majority of shrimp allergic patients are allergic tomealwormBroekmanH.C.H,VerhoeckxK.C.M,denHartogJagerC.F.,KruizingaA.G., Pronk-Kleinjan M., Remington B.C., Bruijnzeel-KoomenC.A.F.M.,HoubenG.F.,KnulstA.C.
(Published as letter to the editor J Allergy Clin Immunol. 2016.137(4):1261-3.)
Chapter4
52
Chapter4
53
Chapter4:Majority of shrimp allergic patients are allergic tomealwormAbstractBackground: The growing world population motivates the exploration ofnew sustainable protein sources to ensure food security. Insects likemealwormarepromisingcandidates,withactiveongoingmarketingeffortswithin America and Europe. This warrants further risk assessment,specifically of potentially allergic responses. On the basis of pilot results,we hypothesized that patients allergic to shrimp and/or house dustmiteareatriskformealwormallergy.Objective:Toinvestigatetheallergicpotentialofmealwormintheshrimpallergicpopulation.Methods: We included fifteen shrimp allergic patients in a double-blindplacebocontrolledchallengetrial,performingdiagnosticImmunoCAP,skinprick test (SPT), Basophil activation test (BAT) and immunoblot in allpatientstocharacterizeourpatientpopulation.Results: 13 out of 15 patients had a positive response to mealworm,starting at doses of 0.1 g of ingested mealworm. Positive ImmunoCAPand/orSPT,confirmedthatallsubjectsweresensitizedtomealworm.Tenpatients recognized tropomyosin, either in combination with or withoutargininekinaseinstandarddiagnostictests.Therangeofallergensinferredfrom the immunoblot assay suggests that known and unknown allergensareinvolved.Conclusion: Themajority of our shrimp allergic population has a DBPCFCproven food allergy to mealworm, reacting to known and unknownarthropodallergensatarelevantdoseofmealworm.BackgroundThedevelopmentofalternativeandsustainablefoodsourcesisconsideredatoppriorityontheinnovationagendaofmanynationalandinternationalbodies[1,2].Variousalternativeproteinsourcessuchasalgaeand insectsarebeinginvestigatedaspossiblecandidates.Tenebriomollitor,thelarvaeof the yellowmealwormbeetle (mealworm)was identified tohave greatpotential for its protein content, sustainability and low maintenance inrearing[3,4].Mealwormasaningredientinburgersordriedcrispysnackis
Chapter4
54
already available through delicatessen shops and major supermarkets inEurope,NorthAmericaandAustralia.Mealworm proteins may induce respiratory allergy [5,6] and one casereport described a systemic reaction after eating mealworm [7].Unfortunately, confirmation of an allergy with a food challenge is stilllacking. Other insects, such as Bombyx mori (silkworm), tettigonid oracridid (grasshopper or cricket) andGonimbrasia belina (mopane worm)havebeenreportedtocausefoodallergy[8-10].Althoughtoxicologicalandmicrobiologicalriskofsuch insect-basedfoodshasbeenassessed[11-12],thepotentiallyallergenicriskshavenotbeensystematicallystudiedyet.Up to 5% of theWestern European population have a food allergy,withshellfish among the tenmost prevalent eliciting sources [13]. Shellfish isthe common name to describe both crustaceans (including shrimp, craband lobster) andmollusks. The proteins tropomyosin and arginine kinaseare the major shellfish allergens, whereas sarcoplasmic calcium-bindingprotein, myosin light chain, hemocyanin, troponin c and alpha-actin areconsidered tobeminorallergens [14-20].Because shellfishbelong to thesame phylum (arthropoda) as insects, a key question is whether cross-reactivity or co-sensitization [21] may lead to clinically relevant allergicresponses. Therefore, the aim of this studywas to assess if amealwormfoodchallengecould triggera foodallergic reaction inpatientsallergic toshrimp.MethodsStudydesignMedical histories were obtained and sensitization to food allergens(shrimp, other shellfish, and wheat), inhalant allergens (house dust mite(HDM),cat,dog,birch-andgrasspollens)andinsectallergens(mealworm,cockroach and silkworm) determined. Patients sensitized to mealwormparticipatedinadouble-blindplacebocontrolledfoodchallenge(DBPCFC)withmealworm. Patients were excludedwhen pregnant, diagnosed withsevere asthma (FEV1 <70%), or using systemic immunosuppressants orbeta-blockers.PatientpopulationAdult patients (n=60) from the University Medical Center Utrecht, alldiagnosed with shrimp allergy on the basis of specialist opinion anddiagnostic testing,were foundeligible.Patientswere rankedaccording tosIgE titers to shrimp and thosewith the highest titers were invited first,
Chapter4
55
since for further serology, higher titerswere preferred. Becausemost oftheseinvitedpatientshadapollenallergy,wemadesurethatsubsequentlypatients without pollen allergy were invited, to avoid selection bias. Allsubjectsgavewritteninformedconsentbeforeparticipation.Thestudywasapprovedbythelocalethicscommittee(NL43731.041.13).SpecificIgESpecific IgE using ImmunoCAP (HDM, shrimp and mealworm) andImmunoCAP ISAC was tested according to the manufacturer’srecommendations (Thermo Fisher Scientific, Uppsala, Sweden), andexpressedinkU/LandISUrespectively.Testswereconsideredpositivewithavalueof0.35orhigher in ImmunoCAPand0.3orhigher in ImmunoCAPISAC.MealwormextractFreshandfreezedriedYellowmealwormsinfinallarvalstage,werekindlyprovidedbyDutch insect farmKreca (Ermelo, theNetherlands). Tris/ureaandSDS/DTTextractswerepreparedasdescribedpreviously[22]andwereusedforimmunoblotandBAT.SPTsolutionofmealworm(0.4mg/mL),waskindlyprovidedbyALK(ALK-Abelló,Spain).ImmunoblotForSDS-PAGE,theCriterionsystemwitha10-20%ReadyGel®Tris-HClgel(Bio-Rad, Hercules, (CA, USA) was used according to the manufacturer’sinstructions. Mealworm, shrimp (ALK) and shrimp tropomyosin Pen a 1,(Indoor Biotechnologies) extracts (5 μg) were loaded on the gel, underreducing conditions (Laemmli buffer). After protein separation,immunoblotwasperformedasdescribedpreviously[22].Basophilactivationtest(BAT)BATwasperformedasdescribedpreviouslywithminormodifications.Cellswereincubatedwithadilutionseries(1:107to1:102)ofmealwormextracts(Tris/urea (5 mg/mL) and SDS/DTT). Shrimp extract (ALK, 2 mg/mL), andshrimptropomyosinPena1(IndoorBiotechnologies,1mg/mL),wereusedas positive controls [22]. The results were expressed as a percentage ofCD63+basophils.SkinPricktest(SPT)Commercial SPT solutions for HDM, shrimp, common inhalant allergensandwheat(ALK),crab,lobster,cockroachandsilkworm(Greer)wereused.
Chapter4
56
SPTwasperformedaccordingto theproceduredescribedpreviously [23].SPTreactivitywasmeasuredastheratioofthemeanofthewhealelicitedbythetestedextractandhistaminecontrol(3+).Whentheratiowas0.5orgreater,thereactionwasregardedaspositive.FoodchallengeBlanchedmealwormwasadded to chickenmeatwith added salt, pepperand 0.3% nutmeg. In the placebo, the mealworm was replaced withchickenmeat.Patientsreceived7portionsoneachchallengeday,startingat2.16mgmealwormprotein(correspondingwith10mgtotalmealworm),followedby21.6mg,216mg,648mg,2.16g,6.48g,and13.0gmealwormprotein.The challenge was discontinued and considered positive in case ofobjective symptoms or if a suggestive moderate to severe subjectivesymptom lasted for > 45 minutes, and scored according to Mueller andSampson[24,25].FourpatientshadanopenfoodchallengetoboiledDutchshrimp,priortoparticipatinginthisstudy,consistingof7dosingsteps(0.1g,0.3g,1g,3g,10g,30gand100gshrimp).AnalysisPopulation size analyses using binominal distribution, indicated that 15patients were needed to be included in the DBPCFC with mealworm, tocalculate the possibility of allergy with a statistical confidence of 90%, if15%ofthesensitizedpatientswouldreacttoDBPCFC.This15%was(basedon the 30%prediction valueof the SPT froma recent shrimp study [26])usedtopreventtoolimitedpower,sincesensitizationtestingmethodsformealworm were used without any knowledge on the positive predictingvaluesandduetotheabsenceofexperiencewiththemealwormextracts.Log-logistic, log-normal, and Weibull eliciting dose (ED) with ED5, ED10,indicating where 5% and 10% of the allergic population is predicted toreact were estimated for mealworm using cumulative dose probabilitydistributionmodels.(supplementarydata)AllresultswereperformedusingSPSSInc,Chicago,version21.0.ResultsClinicalcharacteristicsofshrimpallergicpatientsWeenrolled18patientsinthisstudytoinclude15patientsintheDBPCFC.
Chapter4
57
Thisamountwasbasedonbinominaldistributionforminimal inclusiontofind at least one at risk patient. But three patients were excluded fromfurtheranalysisdue toanexercise inducedallergy inone,one for lackofevidenceformealwormsensitizationandinoneaninconclusivemealwormfood challenge result. Sensitization was 94.4% within the invited shrimpallergicpatients.Intotal,wecompletedourstudywith15patients.Median age of the 15 patientswas 38 years (range 19-69) and 47%wasmale.Symptoms by history, ranged fromoral symptoms to anaphylactic shock.Fourpatientshadapositivediagnosticshrimpchallengeinthepast(Table1). None of the patients knowingly consumed mealworm proteins. ThemajorityhadinhalantallergiestoHDM(11/15)andpollen(11/15,ofwhich7weremild,withnoneedofmedicationduringtheseason))and9patientshad one or more other food allergies. All patients avoided eating othershellfish(e.g.crustaceansandmollusks).SensitizationpatternofshrimpallergicpatientsMean sIgE to shrimpwas 13.0 kU/L (range 0.37 kU/l to 53 kU/L) and allpatients had a positive shrimp SPT (Table 1). ImmunoCap ISAC analysisshowed that 10 patients were sensitized to tropomyosin from Peneusmonodon shrimp (Pen m 1) and only one patient was additionallysensitized toPenm4 (sarcoplasmic calciumbindingprotein). SixpatientsweresensitizedtoPenm2(argininekinase).Threewerenotsensitizedtoanyoftheshrimpcomponents,althoughshrimpImmunoCAPwaspositive.Furthermore, sensitization to tropomyosins from other species (HSM,CockroachandAnisakis)wasfoundandSPTtocrab,lobster,cockroachandsilkwormwerepositive(Table2)inmostpatients.Shrimp allergic patients showed (functional) sIgE to different mealwormproteinsImmunoCAP tomealwormwaspositive (>0.35kU/L) in9patientswithameanof1.8kU/L(range0.66to6.0kU/L),table3.Molecularweightoftherecognizedproteinbandsrangedfrom10to200kDa.Themajority(14/15)recognized tropomyosin and arginine kinase (10/15). Additionally, 4patients recognized bandswith amolecularweight > 60 kDa, including abandat200kDaand4mainly recognizedproteins<25kDa. Interestingly,IgE from one patient bound only to a 200 kDa protein (SupplementaryFigure1).
6 M 30 y/y/y 3 3+ 26 58 0 00.3g(65mg)subj1g(216mg)obj
7 M 46 y/n/y 3 3+ 15 18 01.2
-
8 F 23 n/n/y 3 2+ 4.6 7.6 1.6 0 0.1g(21.6g)subj
9 F 23 y/y/y 3 3+ 53 44 33 0 -
10 F 52 n/n/n 4 1+ 0.37 0.60 0 0 -
11 M 19 n/n/n 4 1+ 6.1 0 0 0-
12 M 45 n/y/y 4 2+ 18 0 0 0-
13 M 26 n/n/n 4 2+ 1.9 3.3 0.60 0 1g(216mg)obj
14 F 27 y/y/y 4 4+ 18 16 4.1 0 -
15 F 47 y/n/y 4 1+ 2.0 0 0 0 -
AA Allergic Asthma, AD Allergic dermatitis, AR allergic rhinitis; n no, y yes;M =male,F=female;SPTinrespecttohistamine3+;Penm1shrimptropomyosin;Penm2shrimpargininekinase;Penm4shrimpSarcoplasmiccalciumbindingprotein;ISU ISAC Standardized Units; Shrimp challenge eliciting dose of shrimp in bold,proteinbetweenbrackets.subjsubjectivesymptoms;objobjectivesymptoms;EDunknown,elicitingdosenotdocumented.
MealwormSPTwaspositive inallpatientsand reactivitywascomparableto shrimpandHDM.MealwormBATwaspositive for all patientswithanapproved test. Three patients were non-responders and one showedspontaneousrelease,therefore,intotal11BATswereinterpretable.(Table3).
Der p 10 house dust mite tropomyosin; Ani s 3 anisakis tropomyosin; Bla g 7cockroachtropomyosin;ISUISACStandardizedUnits;ntnottestedMajority of shrimp allergic patients reacted with allergic symptoms tomealwormchallengeDBPCFC confirmed mealworm allergy in 13 out of 15 shrimp allergicpatients (Table 4). Symptoms ranged from oral allergy (7/13), andsymptoms such as urticaria (6/13), nausea (8/13) abdominal cramping(4/13), vomiting (1/13), to respiratory: dyspnea (2/13). Subjectivesymptomsstartedatadoseof21.6mgmealwormprotein,andobjectivesymptoms at a 10-fold higher dose (216 mg). One patient reacted fourhours after the final dose, showing both subjective and objectivesymptoms.TwopatientsdidnotcompletetheDBPCFCduetotheseverityoftheirreactiononthefirstday.Sincetheybothreactedtotheactivedoseand therefore only the results of the placebo day were missing, it wasdecidedtokeepthemintheanalysis.Patientcharacteristicsfromthetwo
Chapter4
60
patients with a negative challenge, were similar to those that reactedpositive. Mealworm allergy was scored between 0 and 3 according toMueller classification or between 0 and 4 according to Sampsonclassification.(Table4).Table3.Sensitizationpatterns tomealworm (n=15): SPT, ImmunoCAP,BAT andimmunoblot.
Pt.
SPT
mealworm
Mealworm
CAP(kU/l)
CD63
release
mealworm
CD63
release
shrim
p
CD63
release
Pena1
blottrop
.mealworm
blota-k.
mealworm
blotsh
rimp
blotPen
a1
1 2+ 0.16 + + + + - + +
2 2+ 3.1 NR NR NR + + + +
3* 2+ 0.20 + - + + + + -
4 2+ 1.2 + + + + + + +
5 3+ 0.31 + + + + + + +
6 2+ 4.0 SR SR SR + + + +
7 2+ 2.5 + + + + - + +
8 2+ 0.80 NR NR NR + - + -
9 4+ 6.0 + + + + + + +
10 1+ 0.07 NR NR NR + + + +
11 2+ 0.30 + + + + - + -
12 2+ 1.6 + + - - - + -
13 2+ 0.21 + + + + + + +
14 4+ 0.66 + + + + + + +
15 3+ 2.3 + + + + + + -
Pen a 1 = shrimp tropomyosin NR = non-responder (control negative) SR =spontaneous release (spontaneous release on negative control) s.trop = shrimptropomyosinpena1,trop=tropomyosin,a-k=argininekinase*suspectedswitchinBATassayshrimpandtropomyosinresults.
Chapter4
61
Table4.MealwormDBPCFCin15patients.Pt
0.01
g(2.16
mg)
0.1g(2
1.6
mg)
1g(21
6mg)
3g(64
8mg)
10g(2
.16g)
30g(6
.48g)
60g(13.0g)
Mue
llerm
w
Sampson
mw
Diagno
stic
shrim
pchalleng
e
1 - - - - - - - Neg Neg -
2 - - - OA OA - OA 0 1 -
3 - - - P,U - -P,U,NC,F
1 2 10g
4 - - - - OA OA OA 0 1 -
5 - - - - - -OA,N,Sw,U
1 3 -
6 - - - - - - N,AP 2 2 0.3g 1g
7 - OA OA - OA,APOA,AP
ND 2 2 -
8 - - - OA,U OA OAOA,U,AR,Cj
1 2 0.1g
9 - - - -OA,Dy,NC
OA,N,S
OA,N,P,Dy
3 4 -
10 - - - - - - - Neg Neg -
11 - - - - U N,AP ND 2 2 -
12 - - - - -
N,AP,Cj
ND 2 2 -
13 - - - - - - U,N 2 2 1g
14 - - OA - - OA
OA,Cj,P,N,V
2 2 -
15 - - U U,C
U,C,Dy,W,F,N
ND ND 3 4 -
Symptoms duringmealworm challenge. Light grey indicates subjective thresholddoses, dark grey represents objective threshold doses.Mealworm dose in bold,amount of protein between brackets. AP abdominal cramp or pain; AR, allergicrhinitis;C,cough;Cj,conjunctivitis;Dy,dyspnea;F,flushing;NCnasalcongestion;N, nausea;OA, oral allergy; P, pruritus; S, sneezing; Sw, difficulty swallowing;U,urticaria;V,vomiting;W,wheezing.ND=Nodosegiven;mw,mealworm.
Chapter4
62
When comparing themealworm challengeoutcomeof four patients thatalso had a shrimp challenge, eliciting doses (ED5 and ED10) as well asseveritywereinthesamerange.Toplacetheseelicitingdosesinabroaderscope, population threshold dose distribution information for mealwormwas compared with population threshold data for shrimp and otherallergenic foods from the VITAL- panel [27] (Supplementary Table 1 andFigure2and3).Thesewerecomparabletothoseofshrimp.Conclusion/DiscussionThemostimportantconclusionfromourstudyisthatmealwormallergyishighlylikelytobepresentinshrimpallergicpatientswithpotentiallysevereoutcomewhenexposed tomealworm.Outof15 total,14 shrimpallergicpatientsweresensitizedtoeithermealwormtropomyosinand/orargininekinase, and 13 out of 15 shrimp allergic patients reacted positive in aDBPCFC with mealworm with 11 developing moderate to severesymptoms, requiring immediate treatment with anti-histamines,corticosteroids or epinephrine. IgE binding was found not only totropomyosin and arginine kinase,well-known shrimpandHDMallergens,butalsotootherasyetunidentifiedproteins.Wenotethatthethresholdsformealwormwerecomparabletothosefromshrimp.Althoughthesamplesizeof15patientswas limitingpower,theobservedmagnitude of effect is nearly fully penetrant for this particular patientgroup, which is a representation of the shrimp allergic population in atertiarycentre.EventhoughaslightunderrepresentationofthelowersIgEtiters for shrimpwas seen in the study population in comparison to thetotal available shrimp allergic population in our hospital (n=60), thestatistical and graphical comparison of the severity score of the patient-histories showednoclearevidenceofdeviations fromtheshrimpallergicpopulationatourhospital.Therefore,webelievethatthestudygroupwasrepresentative for the shrimp allergic population at our hospital. Thestrength of this studywas the use of DBPCFC, the gold standard in foodallergy diagnosis, in addition to a broad panel of in vitro and in vivosensitization tests. To the best of our knowledge, no other study hasassessedallergenicityofinsect-basedfoodtothisextent.ThemajorityoftheshrimpallergicpatientsthatparticipatedintheDBPCFCwerenotonlysensitizedtomealwormbuttoothercrustaceans(14/15)orinsects (11/12)aswell. Thismightwellbedue to thehomologybetweentropomyosins within the crustacean family and arthropods (sequenceidentity ranging 95-98%) [28]. In addition, these patients were also
Chapter4
63
sensitizedtotropomyosinsfromotherspecies:Anis3(anisakis),Derp10(HDM), Bla g 7 (cockroach). Ten out of 15 patients were sensitized toarginine kinase according to immunoblot,which is confirmed by the factthat 11 out of 12 patients tested had a positive SPT to silkworm, witharginine kinase as the major allergen. The sequence identity betweenshrimp and silkworm (Bombyx mori) arginine kinase was reported to be84%[29].Additionally,manyotherproteinbands,besidestropomyosinandargininekinase,withvaryingmolecularweights,(10kDa->200kDa)wereobserved. These bands could represent other shellfish allergens, such astroponinc,alpha-actinin[19,30],vitellogen,paramyosinorchitinasefromotherinsectsorhousedustmites[31,32][E.WeberIUISdatabase].Findingsuchahighpercentageofshrimpallergicpatientswithaprobablycross-reactive allergy tomealworm (87%) is striking. One open challengestudywithtwoshrimpspecies(P.monodonandM.rosenbergii),foundonlyhalfofpatientsreactingtobothspecies[33].Anotheropenchallengestudyfound that less than 50% of the shrimp allergic patients reacted to twoshrimp species (L.vannamei and P.monodon) [34]. These shrimp specieswere phylogenetically more closely related than shrimp and mealworm.There are no prior DBPCFC studies on cross-reactivity/co-sensitization ofshrimp allergic patients with arthropods, but cross-reactivity withincrustaceanappearedtobeashighas75%betweendifferentspecies,basedonstudiesonsequencehomologyandserology[35].Ahighdegreeofco-sensitization(50-100%)tocrab,crayfishandlobsterusingSPTwasfoundinshrimp allergic patients [36]. Our data indicate a high degree of cross-reaction/co-sensitizationevenbetweenlessrelatedspecies.Two patients had recurrence of symptoms 2 to 5 hours after an initiallyfavorable response upon treatment. In one patient the symptoms werelong-lasting, despite treatment. Another patient started to react only 4hours after the last dose, which is unusual with the exception of meatallergy due to alpha galactose sensitization [37]. Perhaps this delayedresponse can be explained by matrix effects as high-fat or protein-richmatrices were found to lower gastric emptying rates and delay uptake[38,39]. A role for chitin from the mealworm, estimated 19.6 mg/kg byFinkeetal.[5]seemsunlikely,sincethecumulativedosageofchitinduringtheDBPCFCneverexceededthe45grams,whichwasreportedforshellfishchitintobewithoutproblems[40].
Chapter4
64
The severity ofmealworm allergy varied betweenmild (oral allergy) andmoderate (urticaria and gastro-intestinal symptoms) to severe (dyspnea).Fiveofthepatientsshowedoralorskin/mucosalsymptoms(Muellerscore0-1). An open shrimp challenge study reported 80%oral or skin/mucosalsymptoms[34].AsimilarprofilewasfoundinaDBPCFCstudywithshrimpwherethemajority70%experiencedoralorskin/mucosalsymptomswithMueller score 0-1[41]. Together these data indicate thatmealworm is atleast as allergenic as shrimp. Furthermore, it must be mentioned thatsymptomsduringDBPCFCarealwaysanunderestimationofseveritysincethereactionistreatedwhenstoppingcriteriaaremet.Comparison of the preliminary population ED10 estimate for mealwormfrom our study with the ED10 value for shrimp from recent populationthresholddatafromtheEuroPrevallprojectandtheVitalScientificExpertPanel (VSEP) [41,27] indicates a comparable potency for effect elicitationfor mealworm and shrimp protein. More mealworm challenge data isneededtoconfirmthisinitialanalysis.Thesefindingsarehighlyrelevantforconsumption of insect containing foods, as the individual thresholds forobjectivesymptoms(216mgofmealworm)areequaltoorlowerthantheamountofmealwormproteinthatiscurrentlybeingusedininsectsnacks.Forinstance,intheBelgianbrandDamhert,thatisalreadyonthemarketinBelgiumandtheNetherlands,259mgofmealwormproteinisusedina20-gramsnack,whereaservingisaboutthreeofthesesnacks.This indicatesthat shrimp allergic patients are already at risk when consuming theseinsectsnacks.Inconclusion,shrimpallergicpatientsareatriskwheneatingmealwormasan alternate source of dietary protein. Since mealworm containingproducts are increasingly being consumed in America and Europeancountries, it might be prudent to consider notifying shrimp and shellfishallergicpatientsaboutthisrisk.
Supplementary Figure 1. Log-logistic, log-normal, or Weibull probabilitydistribution models of mealworm (expressed as cumulative mg mealwormprotein).Predicteddistributionsandactualchallengedatapoints(•)aredisplayed.
Chapter4
66
Supplementary Figure 2. Log-normal probability distribution models ofmealworm and shrimp (expressed as cumulativemg protein) and corresponding95%confidenceintervals.
4. Bednarova M, Borkovcova M, Komprda T. Purine derivate content andaminoacidprofile in larval stagesof threeedible insects. Journalof thescienceoffoodandagriculture.2014;94(1):71-6.Epub2013/05/02.
5. Bernstein DI, Gallagher JS, Bernstein IL.Mealworm asthma: clinical andimmunologic studies. The Journal of allergy and clinical immunology.1983;72(5Pt1):475-80.Epub1983/11/01.
7. FreyeHB,EschRE,LitwinCM,SorkinL.Anaphylaxistothe ingestionandinhalation of Tenebrio molitor (mealworm) and Zophobas morio(superworm). Allergy and asthma proceedings : the official journal ofregionalandstateallergysocieties.1996;17(4):215-9.Epub1996/07/01.
8. Kung SJ, Steenhoff AP, Gray C. Food allergy in Africa: myth or reality?Clinical reviews in allergy & immunology. 2014;46(3):241-9. Epub2012/11/28.
9. Ji KM, Zhan ZK, Chen JJ, Liu ZG. Anaphylactic shock caused by silkwormpupa consumption in China. Allergy. 2008;63(10):1407-8. Epub2008/09/11.
11. Finke MD. Estimate of chitin in raw whole insects. Zoo biology.2007;26(2):105-15.Epub2007/03/01.
12. KlunderHC,Wolkers-RooijackersJ,KorpelaJM,NoutMJR.Microbiologicalaspects of processing and storage of edible insects. Food Control.2012;26(2):628-31.
16. Ayuso R, Grishina G, Bardina L, Carrillo T, et al.Myosin light chain is anovel shrimp allergen, Lit v 3. The Journal of allergy and clinicalimmunology.2008;122(4):795-802.Epub2008/09/02.
17. Ayuso R, Grishina G, IbanezMD, Blanco C, et al. Sarcoplasmic calcium-binding protein is an EF-hand-type protein identified as a new shrimpallergen.TheJournalofallergyandclinicalimmunology.2009;124(1):114-20.Epub2009/06/16.
18. GiuffridaMG,VillaltaD,MistrelloG,AmatoS,etal.ShrimpallergybeyondTropomyosin in Italy: clinical relevance of Arginine Kinase, SarcoplasmiccalciumbindingproteinandHemocyanin.Europeanannalsofallergyandclinicalimmunology.2014;46(5):172-7.Epub2014/09/17.
19. Kalyanasundaram A, Santiago TC. Identification and characterization ofnew allergen troponin C (Penm 6.0101) from Indian black tiger shrimpPenaeusmonodon.EurFoodResTechnol.2015;240(3):509-15.
21. VerhoeckxKC,vanBroekhovenS,denHartog-JagerCF,GaspariM,etal.Housedustmite(Derp10)andcrustaceanallergicpatientsmayreacttofood containing Yellow mealworm proteins. Food and chemicaltoxicology: an international journal published for the British IndustrialBiologicalResearchAssociation.2014;65:364-73.Epub2014/01/15.
23. Peeters KA, Nordlee JA, Penninks AH, Chen L, et al. Lupine allergy: notsimply cross-reactivity with peanut or soy. The Journal of allergy andclinicalimmunology.2007;120(3):647-53.Epub2007/07/20.
25. Sampson HA, Anaphylaxis and emergency treatment. Pediatrics2003;111:1601-8.
26. YangAC,ArrudaLK,SantosABR,BarbosaMC,etal.Measurementof IgEantibodies to shrimp tropomyosin is superior to skin prick testing withcommercial extract and measurement of IgE to shrimp for predictingclinically relevant allergic reactions after shrimp ingestion. J Allergy ClinImmun.2010;125(4):872-8.
27. TaylorSL,BaumertJL,KruizingaAG,RemmingtonBC,etal.Establishmentof ReferenceDoses for residuesof allergenic foods: report of theVITALExpert Panel. Food and chemical toxicology : an international journal
Chapter4
69
published for the British Industrial Biological Research Association.2014;63:9-17.Epub2013/11/05.
28. ReeseG,AyusoR,LehrerSB.Tropomyosin:aninvertebratepan-allergen.International archives of allergy and immunology. 1999;119(4):247-58.Epub1999/09/04.
29. Liu Z, Xia L,Wu Y, XiaQ, et al. Identification and characterization of anargininekinaseasamajorallergenfromsilkworm(Bombyxmori) larvae.Internationalarchivesofallergyandimmunology.2009;150(1):8-14.Epub2009/04/03.
30. Gamez C, Zafra M, Boquete M, Sanz V, et al. New shrimp IgE-bindingproteins involved in mite-seafood cross-reactivity. Mol Nutr Food Res.2014;58(9):1915-25.Epub2014/07/01.
31. Epton MJ, Malainual N, Smith WA, Thomas WR. Vitellogenin-apoplipophorin likeallergenDerp14 isamajorspecificity inhousedustmitesensitisation.JAllergyClinImmun.2001;107(2):S14-S.
33. JirapongsananurukO, Sripramong C, Pacharn P, Udompunturak S, et al.Specific allergy to Penaeus monodon (seawater shrimp) orMacrobrachium rosenbergii (freshwater shrimp) in shrimp-allergicchildren. Clinical and experimental allergy: journal of the British Societyfor Allergy and Clinical Immunology. 2008;38(6):1038-47. Epub2008/05/24.
34. Thalayasingam M, Gerez IFA, Yap GC, Llanora GV, et al. Clinical andimmunochemicalprofilesoffoodchallengeprovenoranaphylacticshrimpallergy in tropical Singapore. Clinical and Experimental Allergy.2015;45(3):687-97.
35. Sicherer SH. Clinical implications of cross-reactive food allergens. TheJournal of allergy and clinical immunology. 2001;108(6):881-90. Epub2001/12/14.
36. Waring NP, Daul CB, deShazo RD, McCants ML, et al. Hypersensitivityreactionstoingestedcrustacea:clinicalevaluationanddiagnosticstudiesin shrimp-sensitive individuals. The Journal of allergy and clinicalimmunology.1985;76(3):440-5.Epub1985/09/01.
37. ComminsSP,JamesHR,StevensW,PochanSL,etal.Delayedclinicalandex vivo response tomammalianmeat in patientswith IgE to galactose-alpha-1,3-galactose. The Journal of allergy and clinical immunology.2014;134(1):108-15.Epub2014/03/25.
39. Schulten V, Lauer I, Scheurer S, Thalhammer T, et al. A food matrixreducesdigestionandabsorptionoffoodallergensinvivo.MolNutrFoodRes.2011;55(10):1484-91.
40. EFSAPanelondietetricproductsnaa.Scientificopiniononthesafetyog'Chitin-glucan' as a novel food ingredient. EFSA Journal. 2010;8(7):1687-704.
41. Ballmer-Weber BK, Fernandez-RivasM, Beyer K, et al. Howmuch is toomuch? Threshold dose distributions for 5 food allergens. The Journal ofallergyandclinicalimmunology.2015;135(4):964-71.Epub2015/01/16.
Chapter5:Not only shrimp allergic, but possibly all atopicpopulationsareatriskformealwormallergy.AbstractBackground: In previous work, we showed that the majority of shrimpallergicpatientsfromourstudypopulation(n=13/15)werefoodallergictomealworm, probably due to cross-reactivity. Arthropod pan-allergenstropomyosin and arginine kinase were involved. In this study, webroadened the study population, to get more insight in the size of thepopulationsatrisk.Methods:Mealwormsensitizationwasexaminedinshrimpallergicpatients(n=67), HDM allergic rhinitis patients without tropomyosin and argininekinasesensitizationorshrimpallergyorsensitization(n=58),patientswithseasonalrhinitis(n=20),andnon-atopiccontrols(n=62).Thepatientswerefurthercharacterizedusingclinicalhistory,and ImmunoCAP forHDMandshrimp. ImmunoCAP ISAC was performed in the HDM population, tospecify the sensitization to major HDM allergens Der p 1, and 2 and toensuretheabsenceofsensitizationtoDerp10,andpotentiallyshrimpandcockroach allergens (check) to avoid any overlapwith the shrimp allergicpopulation. A new, specifically for this study prepared ImmunoCAP withourmealwormextract,wasusedinthisstudy.ImmunoCAPvaluesequaltoorhigherthan0.35kU/Lwereconsideredpositive.Results:Mealwormsensitizationintheshrimpallergicpopulationwas88%,with a median of 4.8 kU/L. Mealworm sensitization in the HDM allergicsubpopulation was 22%, with a median of 0.62 kU/L, and in seasonalrhinitis patients 15%, with a median of 1.11 kU/L. None of the 62 non-atopic controlswere sensitized tomealworm. Because theHDMpatientswith cross-reactive allergens between shrimp and HDM were excluded,most likely the allergens involved are different from the ones in shrimp-mealwormcross-reactivity.Conclusion: Shrimp allergic patients are most at risk for food allergy tomealworm.However,theremightbeariskalsoinHDMallergicandotheratopic populations. Although the percentages of patients sensitized tomealworm in these latter groups are lower than in the shrimp allergicgroup,onapopulationlevelthesegroupsmayconcernsubstantiallylargeratriskpopulations.
Chapter5
74
BackgroundIntroduction of novel foods may pose risks of development of newallergies.This is illustratedbytheexampleof lupin,whichwasintroducedasanewproteinsourceandreplacementforsoyandcausedfoodallergyinpeanutallergicpatients [1].Mealworm isanother foodrecentlyproposedand introduced as protein source. We previously studied potentialallergenicrisksofmealworm.Shrimpallergicpatientswereshowntobeathigh risk for food allergy to mealworm [2]. Both shrimp and mealwormbelongtothearthropodphylumandweshowedcross-reactivitytoatleasttropomyosin and arginine kinase [2]. Some other studies reported insectsensitization in patientswith shrimp and house dustmite (HDM) allergy,rangingfrom30.7to100%[3,4].SinceHDMalsobelongstothearthropodphylum, HDM allergic patients might also be at risk when eatingmealworm/insects. Sincemanymorepeople suffer fromhousedustmiteallergy, than from shrimp allergy (lifetime prevalence for HDM inducedrhinitis is about 17.1% in Europe and around 0.1% for shellfish allergy[5,6]), this could be an even larger patient population at risk for foodallergy to insect proteins. Within the HDM population, sensitization tomanydifferentHDMallergensisseen,whichdiffersdependingonlocationandage[7].MostrecognizeDerp1(fecalallergen)and/orDerp2(allergenfrom the mite intestine). Smaller parts of the HDM allergic populationrecognize Der p 10, a tropomyosin that originates in the muscle of theHDM, or arginine kinase (Penm2), that derives from shrimpmuscle [8].SomepatientsrecognizeforinstanceDerp23(alsofecal)[9].Tostudythepotential risk for theHDMpopulation,we selectedHDMallergicpatientssensitizedtoHDM(e.g.Derp1and/or2),butnottropomyosin(Derp10,Blag7orPenm1),argininekinase(Penm2)oranyshrimpallergens,tostudythesensitizationtomealwormproteins.Patientswithsensitizationtotropomyosin,argininekinaseorshrimpwereexcludedtoavoidanyoverlapwiththeshrimpallergicgroup.Theshrimpgroupfromourpreviousstudywasexpanded,todeterminemorepreciselythepercentageofmealwormsensitizationinthetotalshrimpallergicgroup.Wefurtherstudiedseasonalrhinitis patientswithoutHDMsensitizationandnon-atopic controls.Withthisstudy,weaimedtoincreaseourinsightintotheriskforshrimpallergicpatientsandtoidentifyotherpotentialat-riskpopulations.MethodsStudydesignandpatientselection
Chapter5
75
Inthisstudyfourpatientgroupswereincluded:1. Shrimp allergic patients (n=67)were diagnosedwith shrimp allergy byspecialist opinion, based on careful medical history and sensitization ineitherskinpricktest(SPT)and/orImmunoCAP.2.HDMallergic rhinitispatients (n=58)werediagnosedwithHDMallergybasedonspecialistopinionbymedicalhistory(perennialorautumn/winterseasonal symptoms of allergic rhino-conjunctivitis) and positive SPT or,when SPT was absent, positive ImmunoCAP HDM. Patients withsensitization to tropomyosin (Der p 10, Pen m 1 and Bla g 7), argininekinase (Penm2) or shrimp, based on data obtained by ImmunoCAP andImmunoCAP ISAC, were excluded to avoid any overlap with the shrimpallergicgroup.ThisHDMallergicsubpopulationwillbefurtherreferredtoasHDMallergicpatients.3.Allergicrhinitispatients(n=20)withseasonalrhino-conjunctivitisduringspring and/or summer and sensitization to inhalant allergens other thanHDMwerea3rdgroupstudied.PatientswithconcomitantHDMorshrimpsensitizationwereexcludedtoavoidanyoverlapwiththeshrimpandHDMallergic group. This 3rd group will be referred to as seasonal rhinitispatientsfromhereon.4. Non-atopic control subjects (n=62) were randomly selected from theEuroPrevall study. These patients had no atopic history and were notsensitizedtoanycommoninhalantorfoodallergens.All serum sampleswere from adult patients from the UniversityMedicalCenterUtrechtwhoprovidedinformedconsent.SpecificIgEHDMandshrimpsensitizationwasdeterminedusingImmunoCAP(ThermoFisher Scientific, Uppsala Sweden). A new, specifically for this studypreparedImmunoCAPwithourmealwormextract,wasusedinthisstudy.ImmunoCAP ISAC (Thermo Fisher Scientific, Uppsala Sweden) wasperformed in the HDM population, to gather information on generalsensitizationisthisgroup.BothImmunoCAP(HDM,shrimpandmealworm)and ImmunoCAP ISAC were tested according to the manufacturer’srecommendations (Thermo Fisher Scientific, Uppsala, Sweden), andexpressedinkU/LandISUrespectively.Testswereconsideredpositivewithavalueof0.35orhigher in ImmunoCAPand0.3orhigher in ImmunoCAPISAC.PreparationofthemealwormextractforImmunoCAP
Chapter5
76
FortheImmunoCAPmealworm,amealwormextractwaspreparedbyextracting 5 grams of freeze-driedmealworms using a sequential proteinextractionmethod[10].First the insectsorshrimpweremixedwith25mL ice-coldTrisbuffer (20mM Tris buffer pH 7.6 containing 1 mM phenylthiocarbamide (SigmaAldrich) and Halt Protease Inhibitor Cocktail (Thermo Scientific).Subsequentlythemealwormsweredisrupted,usinganultraturrax(3x10sec)undercontinuouscooling.Theultraturraxwaswashedwith5mlcoldTrisbufferandthewashliquidwasaddedtothesamplesuspension.Aftercentrifugation(30min,15000xgat4°C),thesupernatantwasrecovered.The insoluble residuewaswashedoncewith5mLTrisbuffer. The30mLand5mLsupernatantwerecombined.25mLwasusedforsamplecleanupandconcentrationusingTCAprecipitation.Secondly, the remainingpelletwasextractedovernightat4°Cwith30mLureabuffer(6Mureain20mMTrisbufferpH7.6containing1mMphenylthiocarbamideandHaltProteaseInhibitor Cocktail). The sample was subsequently centrifuged and thesupernatantwas collected. The pelletwaswashed oncemorewith 5mLureabuffer,centrifugedandthesupernatantwascombinedwiththe30mLureasupernatant.25mLoftheextractwasTCAprecipitated.Trisandureaextractswerecombined(1:1).The Tris/urea mealworm extract was coupled to ImmunoCAP beads byThermoFisherScientific.StatisticalanalysisPrincipal component analysis (PCA) was conducted as general screeningtool. In partial least squares discriminant analysis (PLSDA) the datawereautoscaled (each variable scaled to mean of zero and variance of 1)becausedifferentunitsareinvolvedandwedon’twantthatvariableswithlargevalueswilldominateintheanalysis.PLSDAwasusedtoinvestigateifpositivemealwormCAP/negativemealwormCAPdifferenceswerepresent(classification model including jackknife based variable selection and 10-folddoublecrossvalidation).ResultsThemajorityoftheshrimpallergicpopulationwassensitizedtomealwormIntheshrimpallergicpatientgroup,69%werefemaleandmedianagewas42years.81%hadahistoryof rhinitis,57%ofasthmaand66%ofatopicdermatitis.MedianImmunoCAPshrimpwas4.3kU/L,withaninterquartilerange(IQR)of0.93-17.Thelargemajority,65of67,wasalsosensitizedto
Chapter5
77
HDM (97%), with a median CAP value of 40.3 kU/L and a IQR of 6.9->100kU/L.MealwormIgEwaspositivein59patients(88%).MediansIgEtomealwormforthepositive59patientswas4.8kU/LwithanIQRof1.3-14.2(Table1,Figure 1). In the patients with and without asthma the percentages ofmealwormsensitizationweresimilar,aswerethepercentages inpatientswith or without atopic dermatitis. 15 patients had a DBPCFC withmealworm inourprevious study.Thesepatientsare shown in figure1asblack downward triangles. Filled when the challenge was consideredpositive, open in the twowith a negative outcome. In these 15 patients,73%hadahistoryofrhinitis,53%ofasthmaand53%ofatopicdermatitis.Median ImmunoCAP shrimp was 5.5 kU/L, with an inter quartile range(IQR)of1.9-18.8kU/L.Thelargemajority,14of15,wasalsosensitizedtoHDM (93%), with a median CAP value of 12.3 kU/L and a IQR of 6.9-55kU/L.Median sIgE tomealworm in this shrimp allergic subgroupwas 5.8kU/LwithanIQRof3.5-15.0(Table1).A significant part of the HDM allergic population was sensitized tomealwormIntheHDMallergicpatientgroup(n=58),81%wasfemaleandmedianagewas 40 years. 52% had asthma, 41% had atopic dermatitis and 69% hadfood allergy, but not to shellfish. The median HDM CAP value (in thepatientsthatweretested54/58,duetolimitedserum)was8.5kU/LandanIQRof1.5-34.9kU/L.This issignificantly lowerthan in theshrimpallergicgroup(Table1). Inthepatientswithandwithoutasthmathepercentagesof mealworm sensitization were similar, as were the percentages inpatientswithorwithoutatopicdermatitisandwithorwithoutfoodallergy.Within the HDM population 2 different groups can bemade; 1) patientsrecognizingDerp1and/orDerp2,2)patientsnotrecognizingDerp1,Derp2orDerp10butotherHDMallergensnotonImmunoCAPISAC.ThesIgElevels formealworm in thesegroupsareshown in table1. In theDerp1and/orDerp2group,7patientsshowedsensitizationtomealworm.Inthegroup recognizing other proteins thanDer p 1, 2, and 10, 6 patients hadsIgEformealworm.
All data representmedians and interquartile rangesornumbers andpercentages. * ISACnotperformed inonepatient. #29/6+means29patients recognized components6weremealworm positive, per allergen recognition pattern, number of patients sensitized tomealworm
Chapter5
79
Figure 1: ImmunoCAP (kU/L) mealworm results in log 10 in shrimp, HDM,seasonal rhinitis and non-atopic populations. Sensitization tomealworm 88% ofshrimpallergicpatients (previously inDBPCFC inblack) and22%ofHDMallergicpatients, convincingly higher than 15% in rhinitis and 0% of non-atopic controlsubjects.
Blackdownwardtrianglesaretheshrimpallergicpatientswithmealwormallergy,open black triangles are 2 shrimp allergic patientswith a negative DBPCFCwithmealwormfromourpreviousstudy.Mealworm sIgE was positive in 13 patients (22%). Median sIgE tomealwormforthepositivesubgroupwas0.62kU/LwithanIQRof0.49-1.4In Figure 2a the ISAC results of allergensmostly recognized by the HDMallergic patient group are shown. 50% of these patients recognized theinhalantallergencomponentsPhlp1(grass)andFeld1(cat).Betv1wasrecognizedbyalmost50%ofthepatients.Amongthefoodallergens,Cora1.0401(Betv1relatedhazelnutprotein)andMald1(Betv1relatedappleprotein)werethemostrecognized.Thepercentageofbirch-pollenrelatedfoodallergyinthisgroupisunknown.
Chapter5
80
Principal component analyses showed no clear separation between thegroupswithmealwormpositive ormealwormnegative CAP. However, inPLSDA from the 112 components tested in ImmunoCAP ISAC, fourcomponentsshowedtodiscriminatebetweenHDMpatientswithapositivemealwormandnegativemealwormCAP(Plaa2,Cynd1,Jugr1,MUXF3).The plane tree Pla a 2 showed the best discrimination between the twogroups in the multivariate analysis. Biological explanation for thisdiscriminativefactorcouldnotbefound(Figure2b).The percentage of positivemealworm ImmunoCAP’s in the HDM allergicgroupwas4timeslowerthanthepercentageintheshrimpallergicgroup.MediansIgEtomealwormwas7.7timeslowerintheHDMthantheshrimpallergic group.Almost the samedifference (4.7 times)was seen forHDMsensitizationbetweenthetwogroups.IntheseasonalrhinitispopulationmealwormsensitizationwasfoundIn the seasonal rhinitisgroup78%was female,medianagewas43years,35%hadasthma,25%hadatopicdermatitisand75%hadfoodallergy.Allwere sensitized to inhalant allergens, 90% to pollen and 15% to animaldander(onepatientwassensitizedtoboth).MealwormsIgEwaspositivein3patients(15%).MediansIgEtomealwormforthepositivesubgroupwas1.11kU/Lwitharangefrom0.81-1.97(Table1).Inthepatientswithandwithoutasthma,thepercentagesofmealwormsensitization were similar, as were the percentages in patients with orwithoutatopicdermatitisandwithorwithoutfoodallergy.ThepercentagepositivemealwormImmunoCAP’swas1.5 timeshigher intheHDMgroupcomparedtotheseasonalrhinitisgroup.Nomealwormsensitizationinthenon-atopicpopulationInthenon-atopicsubjectsgroup,55%wasfemaleandmedianagewas49years.AllsIgEtestsincludingmealwormwerenegativeinthe62non-atopicsubjects.
Onthetopleftaremealwormsensitizedpatients.Therowsarefromlefttorighttheaeroallergens,foodallergensand‘others’.DiscussionMealworm sensitization was found in the shrimp allergic (88%), HDMallergic (22%) and seasonal rhinitis population (15%). sIgE levels formealwormwerehigher intheshrimpallergicgroupthantheHDMgroup.In the patients with and without asthma the percentages of mealworm
sensitization were similar, as were the percentages in patients with orwithoutatopicdermatitisandwithorwithoutfoodallergy.Figure2b:SensitisationprofilefromHDMpatientswithapositivemealwormCAPwas comparable with sensitisation pattern of HDM patients with a negativemealwormCAPaccordingtoprinciplecomponentanalysis
We previously found that 87% shrimp-allergic patients that weremealworm sensitizedhad aDBPCFC confirmedmealworm food allergy. Ifwecombine this informationwith thecurrent finding that88%of shrimpallergic patients was mealworm sensitized, it is estimated thatapproximately 75% of the total shrimp allergic population might bemealworm allergic. Given the high degree of cross-reactivity betweenshrimpandothershellfish[11],alsoothershellfishallergicpatients,e.g.forcrabandlobstermighthaveahighriskofmealwormallergy.This sensitization isprobablymostly causedbycross-reactionof thewell-knownarthropodpanallergenstropomyosinandargininekinase,basedontheresultsothersandwehaveshownoncross-reactivitybetweenshrimpand insects [12-14]. We cannot exclude the possibility that also other
Chapter5
83
arthropod allergens play a role. In our previous study,we found besidestropomyosinandargininekinasealsoother IgEbindingproteinbands [2].These allergens could for instance be paramyosin, chitinase or serineprotease[15-17].In theHDMgroup,mealworm sensitizationwaspresent in22%.Recentlyone other study reported on mealworm sensitization in HDM allergicpatients.Mealwormsensitizationwasshown in10/11HDMpatients,4ofwhom were not shrimp sensitized [4]. Rudolf et al. reported 30.7%sensitization,withSPTandintradermaltesttothegrainpestconfusedflourbeetle in HDM allergic patients [18]. From those data, it cannot beextracted if the found reactivity was based on (known) major allergens(tropomyosin and arginine kinase. Since our HDM group consisted ofpatients not sensitized to tropomyosin, arginine kinase or shrimp, HDMallergicpatientsprobablyrecognizeothermealwormallergens.Sensitization tomealworm in HDM patients with Der p 1 sensitization isunlikelytobecausedbycross-reactionwithallergenshomologtoallergensfromtheC1cysteineproteasefamily.Derp1(C1cathepsin)belongstotheC1 cysteine protease family and is for only 33% homologous to C1cathepsin identified in mealworm. This was based on a homologycalculation with the Uniprot database (data not shown). So far, nomealwormhomologueforDerp2wasfound.A large part (42%) of the HDM group, sensitized to mealworm, did notrecognizeDerp1or2(andnotDerp10asthiswasanexclusioncriterion)on the ImmunoCAP ISAC (Table 1). Since all patients were sensitized toHDM by either SPT or ImmunoCAP, this indicates they recognize otherHDM allergens. The variation in recognition of (major) allergens by theHDMpopulationwasalsodescribedbyothers,althoughthepercentageofDerp1and2negativepatientsinourstudywassomewhathigherthanthe30% reported by others [8]. It was suggested that HDM allergic patientswithoutIgEbindingtoDerp1,2or10recognizeaperitrophin-likeprotein,Derp23,possiblychitinbindingprotein[9].Derp23hasshownhomologyto so far not completely characterized allergens from cockroach [19] andmightthereforebeacross-reactiveallergenbetweenHDMandinsects.Tothebestofour knowledge,Derp23 -likeallergenswerenot reportedascross-reactive allergens between shellfish and insects, so these allergensmightbespecificforcrossreactivitybetweenHDMandinsects.
Chapter5
84
Remarkably, the seasonal rhinitis patients without HDM or shrimpsensitization(andthereforewithoutsuspectedcross-reactivityviaallergensinvolved in these allergies) also showed sensitization to mealworm,although at a lower percentage (15%). Insect sensitization in seasonalrhinitispatientswasdescribedincountriessuchasIndiaandIran[20,21].Research has been performed on insect sensitization to mosquitos ormothsinatopicpatients.Sensitizationtothoseinsectspecieswasshowninpercentages that reachorevenexceedthoseofhousedustmite insomeregions[21].Insectsensitizationmayresultfromrespiratoryexposure,butother routes of sensitization might be involved. Exposure by insectscontaminatingourfoodisnotrare[22].Armentiaetal.showedallergytoBruchus lentis, a lentil pest, in 16 patients [23]. Skin prick test (SPT)wasnegativeforpurelentilextract,butpositiveforBruchuslentisandinfestedlentil in all sixteen patients. Bronchial Provocation Test, Double blind,placebo-controlled food challenge (DBPCFC), and immunoblotting werepositiveinthemajorityofpatients.Although the prevalence of sensitization tomealworm ismuch higher intheshrimpallergicpopulation,HDMallergyandallergyotherthanshrimpandHDMaremuchmoreprevalent(lifetimeprevalenceforshellfishallergyisaround0.1%inEuropeandabout17.1%forHDMinducedrhinitis[5,6];allergyprevalence isestimatedaroundthought tobearound10%-20%ofthe general population [24]. Therefore, even though the percentages ofpatients sensitized tomealworm in these latter groups are lower than intheshrimpallergicgroup,onapopulationlevelthesegroupsmayconcernsubstantially larger at risk populations. The clinical relevance of thesensitization to mealworm in the HDM and other allergic populationsremainstobeconfirmed,preferablybydoubleblindfoodchallenges.Conclusion: Not only shrimp allergic patients but also other atopicpopulations are sensitized to mealworm and potentially at risk formealwormallergy.Theallergensinvolvedintheshrimpallergicpopulationvstheotheratopicpopulationsareprobablydifferent.
2 Broekman, H.C, Verhoeckx, K.C., den Hartog Jager, C.F., Kruizinga,A.G., et al., Majority of shrimp allergic patients are allergic tomealworm.JAllergyClinImmunol.2016.137(4):1261-3.
3 PerlmanF.Insectsasinhalantallergens;considerationofaerobiology,biochemistry, preparation of material, and clinical observations. JAllergy.1958Jul;29(4):302-28.
5 Nwaru BI, Hickstein L, Panesar SS, Roberts G, et al. Prevalence ofcommon food allergies in Europe: a systematic review and meta-analysis.Allergy.2014;69(8):992-1007.
6 BlommeK,TomassenP,LapeereH,HuvenneW,etal.Prevalenceofallergic sensitization versus allergic rhinitis symptoms in anunselectedpopulation.IntArchAllergyImmunol.2013;160(2):200-7.
19 WeghoferM,GroteM,ReschY,CassetA,etal:IdentificationofDerp23, a peritrophin-like protein, as a new major Dermatophagoidespteronyssinusallergenassociatedwiththeperitrophicmatrixofmitefecalpellets.JImmunol2013;190:3059-3067.
20 KumarR,GuptaN,KanugaJ,KanugaM.Acomparativestudyofskinprick test versus serum-specific IgEmeasurement in Indian patientswithbronchialasthmaandallergicrhinitis.IndianJChestDisAlliedSci2015;57:81-85
21 Bemanian MH, Korkinejad NA, Shirkhoda S, Nabavi M, et al.Assessment of sensitization to insect aeroallergens among patientswithallergicrhinitisinYazdCity,Iran.IranJAllergyAsthmaImmunol.2012;11(3):253-258
22 U.S. Food and Drug Administration, FDA Food Defect Action LevelsHandbook.2005
23 Armentia A, Lombardero M, Blanco C, Fernández S, et al. Allergichypersensitivity to the lentil pest Bruchus lentis. Allergy. 2006Sep;61(9):1112-6.
PrimaryrespiratoryandfoodallergytomealwormBroekman H.C.H., Knulst A.C., den Hartog Jager C.F., van BilsenJ.H.M., Raymakers F.M.L., Kruizinga A.G., Gaspari M., Gabriele C.,Bruijnzeel-KoomenC.A.F.M.,HoubenG.F.,VerhoeckxK.C.M.(PublishedaslettertotheeditorJAllergyClinImmunol.2017.Mar5.pii:S0091-6749(17)30340-8.)
Chapter6
88
Chapter6
89
Chapter6:PrimaryrespiratoryandfoodallergytomealwormAbstractBackground:Allergenicityofinsectsnewlyintroduced,asfoodisoneofthemajor risks to consider beforemarketing. Co-sensitization/cross-reactivitywithshrimpwaspreviouslydemonstrated,raisingtheissueofwhethertheriskmight be broader and that primary allergy to these insectsmight bepossible.Objective: To elucidate the possibility of de novo/primary mealwormsensitizationandallergy.Methods: We conducted a study in mice to determine the sensitizingcapacity of mealworm. Furthermore, we also conducted a clinical studywith subjects (n=4) having a history of symptoms that started afterexposure to mealworm during domestic or professional breeding.Sensitization was determined by measuring IgE binding to mealwormproteins,usingImmunoCAP,skinpricktest(SPT),immunoblotandBasophilActivation Test (BAT). All four subjects underwent double blind placebocontrolled food challenge (DBPCFC) with mealworm and open foodchallenge with shrimp. Subsequently, using LC-MS, we identifiedmealworm proteins that bound to IgE in sera from mice and humansubjects.Results:MealworminducedIgEagainstmealwormproteinsinmiceandinall 4 human subjects. IgE from mealworm-exposed mice and humansrecognized known mealworm allergens i.e. tropomyosin, arginine kinaseand myosin heavy chain. Two human subjects had food allergy tomealwormconfirmedbyDBPCFCandwerenot shrimpallergic.Theothertwo patients had a suspected respiratory allergy to mealworm.Furthermore,anewputativeallergenwasidentified:larvalcuticleprotein.Conclusion: Exposure to mealworm can lead to primary sensitization inmice and humans. In humans, this can lead to both inhalant and foodallergy.BackgroundGiventheirpredictedworldpopulationof9billionpeopleby2050,theFAOis stimulating the investigation of insects as a new sustainable proteincandidate for feed and food [1]. Mealworm (larvae of the yellowmealworm beetle, (Tenebrio mollitor L.) has great potential, due to its
Chapter6
90
sustainabilityandnutritionalvalueandiscurrentlyintroducedasaproteiningredient in commercially available burgers in a number of Europeancountries [2,3]. Its effect on food allergy prevalence is not known.Depending on age, food allergy prevalence ranges within the Europeanpopulation from 0 to 5.7% [4]. Food allergy to insects is not regularlyreported in Westernized countries. However, some reports fromanaphylaxisdueto insect ingestionareavailable fromAsia,where insectsaremorecommonlyserved[5,6].Forexample,anaphylaxisuponingestionhasbeenreportedforaregularlyeateninsectinAsia,thelarvaeofthesilkworm(Bombyxmori)[6].Althoughprevalenceoffoodallergyto insects isnotdescribed,thesilkwormisestimatedtocauseanaphylacticshockover1000timesayearinChinacomparedtoanestimated1080to30,000casesofanaphylaxistoanyfoodintheUSeachyear[6,7,8].Food allergy to mealworm has so far been reported only once. Thisinvolved a systemic reaction including pruritis, generalized urticaria anddiarrhea [9]. There are also a few reports of allergies to insects in anoccupational setting [10,11]. Occupational allergy to mealworm incombination with waxmoth (Galleria mellonella) and greenbottle (Luciliacaesar) was already described in 1994 in 14 amateur and professionalanglers who used these larvae as live fish bait [12]. The 14 subjectsdescribedmealworm-handlingsymptomsranging fromasthmaand rhino-conjunctivitistocontacturticaria.Thesesymptomsmanifestedafter2.3to6.3yearsofexposure.Inarecentstudy,weshowedthatconsumptionofmealwormformsariskforthemajorityofshrimpallergicpatients[13].Wedemonstratedthatallstudied shrimpallergicpatientswere sensitized/ co-sensitized tomultiplemealworm allergens such as tropomyosin and arginine kinase [13]. Ininsects(e.g.cockroachandcricket)andarthropods(e.g.shrimpandcrab),tropomyosin and arginine kinase have been described asmajor allergenswith high sequence homology, which may contribute to cross-reactivitywithinthearthropodphylum[14-16].Cross-reactivitywasalsoindicatedinthesamestudywhereitshowedthat87%oftheseshrimpallergicpatientsreacted positively tomealworm in aDBPCFC, indicating a food allergy tomealworm.Theriskofprimarysensitizationorallergytoinsectsorproteinsthereof,however,hasnotyetbeencharacterized.Totest thesensitizingpotentialof foodproteins,several rodentandnon-rodent animalmodels have been developed [17-19]. Recentlywe used amodel previously described by Bowman et al. [20] to successfullydistinguishapanelof fiveknownallergenicproteins fromfiveknown low
Chapter6
91
allergenic proteins in C3H/HeOuJ mice (manuscript submitted). In thecurrent study,we used thismodel to determinewhethermealwormandshrimpcaninduceprimarysensitization.Moreover,wesearchedforevidenceofprimarysensitizationandallergyinhumans,studyingdomesticandprofessionalmealwormbreedersandtheirsensitization profiles, the allergens involved and the development ofclinicalallergy.MethodsStudydesignThisstudywasdesignedtoassesswhethermealwormcouldcauseprimarysensitizationandfoodallergy.For the clinical trial, through insect farms and internet blogs, 4 subjectswere located with histories of symptoms after handling, eating or otherexposure to mealworm. During a screening visit, their medical historieswereobtainedand,usingSPT,subjectswereexaminedforsensitizationtofoodallergens (variousshellfishandwheat, the latterbeingan ingredientof mealworm feed), inhalant allergens (HDM, cat, dog, birch- and grasspollens) and insect allergens (mealworm, cockroach and silkworm). Inaddition, BAT andWesternblotwere performedusingmealworm, shrimpand shrimp-tropomyosinextracts. Specific IgE (sIgE) for shrimp,HDMandmealworm was determined using ImmunoCAP, and sIgE to allergencomponents was tested using ImmunoCAP ISAC. Immunoprecipitation incombinationwithLC-MSwasusedtoidentifyproteinsthatboundtoIgEinserumfrommiceandhumansubjects.Allsubjectsparticipatedinadoubleblind,placebo-controlled foodchallenge (DBPCFC)withmealwormand inanopen food challengewithDutch shrimp (Crangon crangon) to excludeshrimpallergyasacross-reactivesourceofthemealwormallergy[21].ShrimpandmealwormextractpreparationFresh and freeze dried yellow mealworm (Tenebrio mollitor L.), in finallarvalstage,werekindlyprovidedbyDutchinsectfarmKreca(Ermelo,theNetherlands). Extracts forhuman trial (SPT fromALK) and in vitro testing(Trisandureaextract for immunoblottingandcombinedTris/ureaextractforBAT)werepreparedasdescribedpreviously[21].Fortheanimalstudy,to compare sensitization with a known and possibly similar allergen,shrimps (Pandalus borealis) were obtained from a local supermarket.Peeledshrimp(10g)werefreeze-driedandtheproteincontentmeasuredusingtheKjeldahlmethod.Ashrimpextract(20%protein)waspreparedby
Chapter6
92
homogenizing the shrimp with water using an Ultra Turrax. Mealwormswere washed several times with demineralized water and using Kjeldahlthe protein content was measured. To inactivate endogenous digestionenzymes,drymealwormswereheatedat95°Cfor5minutes.Aftercarefulremovaloftheskin,theinnerpartofthemealwormwasmixedwithwaterusingUltraTurraxtoobtainamealwormextract(20%protein).Theshrimpandmealwormsuspensionswerefrozenuntilfurtheruse.SensitizationofmicewithmealwormandshrimpextractThemouse study was conducted with female C3H/HeOuJmice obtainedfromacolonymaintainedunderSPFconditionsatCharlesRiver, Sulzfeld,Germany.Atcommencementofthesensitization(day0),themicewere7weeksold.Micewereallowedaccesstofood(cereal-basedVRF1diet;SDSSpecialDietsServices,Whitham,England;certifiedfreeofmealwormsandshrimp) andwater ad libitum.Micewere gavaged two times at aweeklyintervalwithshrimp/mealwormextract inPBS(20mgprotein)with10µgcholera toxin (CT; ListBiological Laboratories, Inc.,Campbell,CA).Controlanimals received PBS and CT only. At day 16, antigen-specific IgG1(immunogenicity) and IgE (allergenicity) was determined as described bySmit et al. (submitted manuscript). The welfare of the animals wasmaintainedinaccordancewiththegeneralprinciplesgoverningtheuseofanimals in experiments of the European Communities (Directive2010/63/EU)andDutchlegislation(TheExperimentsonAnimalsAct,1997).This included approval of the study by the Netherlands Organisation forApplied Scientific Research’s (TNO’s) Animal Experimental Committee(DEC-number3640).PrimarysensitizationtomealworminhumansFour adult domestic or professional mealworm farmers, each of whomdisplayedsymptomsuponoccupationalmealwormexposurebyinhalationor mealworm ingestion, were included in this study. The subjects wereselectedforsuspectedmealwormallergybasedonsuggestivehistoryandsensitizationandgaveinformedconsentbeforeparticipation.Thestudywasapprovedbythelocalethicscommittee(NL43731.041.13).Immunoblot, BAT, SPT (mealworm, ALK), shrimp (ALK, Stallergen andGreer),HDM(ALKandGreer)andImmunoCAPmealworm(kindlypreparedfor us by Thermo Fisher), shrimp and HDM (Thermo Fisher) andImmunoCAP ISAC (Thermo Fisher) tests were performed as describedpreviously[13].
Chapter6
93
ADBPCFCwasperformedinallfourhumanmealwormsensitizedsubjects.This challengewasperformedasdescribedpreviously [13].Anopen foodchallenge with boiled Dutch shrimp (Crangon crangon) was performedaccording to the challengeprotocol for shrimpusedatour clinic, using7servings,dosingfrom1mgofshrimpto100gofshrimp.Both the DBPCFC and the open shrimp challengewere discontinued andconsidered positive in case of objective symptoms or if a suggestivemoderatetoseveresubjectivesymptomlastedfor>45minutes.Thedaysof the DBPCFC were de-blinded, after a panel review by three clinicalexpertsontheoutcome.IdentificationofIgEbindingproteinsFor immunoprecipitation, Dynabeads M-280 Tosylactivated (10 mg,Invitrogen)wereusedaccording to themanufacturer’s instructions.Aftercoating with 0.2 mg Goat anti-Hu IgE (AP175 Upstate, Milipore) theseparatebeadswereincubatedforonehourat37°Cwith1mLhumanormouseserum.Conjugatedbeadswerecross-linkedwith5mMBS³(Pierce)according to themanufacturer’s instructions to ensure reusability of thebeads.Thebeadswerewashedthreetimeswith0.1%Tween20inPBSpH7.4, followed by overnight incubation at 37 °C with 100 µL Tris/ureamealworm mixture diluted with 900 µL PBS. After washing 3 times,proteinswereelutedwith2times100µL0.1MglycineandthepHofthesolution was neutralized using 30 µL of 1M Tris-HCL pH 8.5. Incubationwithmealwormextractwasrepeated3timesandalleluates(800µL)werepooled.Beforeanalysis,sampleswerefreezedried,reconstitutedin250µL0.05% SDS, reducedwith 10mMDTT (1 h, 37°C), alkylatedwith 24mMiodoacetamide (1h, 37 °C) and digested with 600 ng proteomics-gradetrypsin after quenching with 2 mM DTT (20 min, 37 °C). Peptides werepurifiedbystrongcationexchangestagetipsandsubsequentlyinjectedformass spectrometric analysis. Of this peptide mixture, 4 µL was analyzedaccordingtoVerhoeckxetal.[21],withthefollowingminormodifications:Thegradientelutionwasachievedat350nL/minflowrate,rampedfrom8%Bto35%Bin60min,andfrom30%Bto100%Binanadditional8min;after5minat100%B,thecolumnwasre-equilibratedat0%Bfor2minbefore the subsequent injection. MS detection was performed on aquadrupole-orbitrap mass spectrometer Q-Exactive (Thermo FisherScientific, Bremen, Germany) using a top-12 method with resolution(FWHM). Data was processed using Proteome Discoverer 1.3 (ThermoFisher Scientific, Bremen,Germany), using Sequest as search engine, and
Chapter6
94
theSwissProtdatabaseaccessedonFebruary2013assequencedatabase(3,123,840 sequences for Metazoa taxonomy). The following searchparameterswereused:MS tolerance15ppm;MS/MS tolerance0.02Da;fixed modifications carbamidomethyl cysteine; enzyme trypsin; max.missed cleavages 1; taxonomy Metazoa. High confidence peptides(confidence > 99 %) were filtered out, using Percolator, integrated inProteome Discoverer. Protein hits based on two successful peptideidentificationswereconsideredvalid.StatisticalanalysisDatawerecollected in IBMSPSSStatistics21.MicedataarepresentedasmeansandanalyzedusingGraphPadPrismsoftware.Antibodylevelswereanalyzed using MannWhitney U test or by Kruskall- Wallis, followed byDunn’sMultiplecomparisontest.ResultsSensitizationtomealwormandshrimpinthemousemodelMiceweredosedbygavagewithmealwormor shrimpextract inPBS (20mgprotein)with10µgcholeratoxin(CT;ListBiologicalLaboratories, Inc.,Campbell, CA). Control animals received PBS and CT only. Mealwormextractinducedextract-specificIgG1in3/6animalsandextract-specificIgEin 2/6 animals (Figure 1). Shrimp extract led to the induction of extract-specificIgG1andIgEin5/6animals.Togetherthesedatashowthepotencyofbothextractstoinduceprimarysensitization.DevelopmentofmealwormallergyinfourhumansubjectsAll four human subjects (between 22 and 46 years of age, one female,three male) developed mealworm allergy during either professional ordomestic mealworm breeding, potentially caused by a combination ofdermal and respiratory exposure. Two subjects, who worked in a roomwhere mealworms were kept at a professional scale, would develop,minutes after entering the room, symptoms such as conjunctivitis andrhinitis,whichdisappearedgradually,withoutmedication,afterleavingtheroom. Neither subject had any other inhalant or food allergies. Sincestartingtoworkat thefarm,theyhadsporadicallyconsumedmealwormsinsmallamounts(severalwormsatatime)withoutsymptoms.Theothertwo subjects rearedmealworms in a separate room at home on a smallscale.Oneofthetwo(subject4)begantodeveloprhino-conjunctivitisafterabouttwoyearsofexposure,whichprogressedtodyspneaandwheezing
Chapter6
95
after he started thedomestic rearing. Their oral intakeof themealwormwas higher than that consumed by the professional farmers: up to 50grams of mealworm each on around 10 occasions. After having eatenmealwormseveral times,bothdomestic farmersencounteredprogressivesymptoms,frompruritisofthelipstoafeelingofswelling,noanaphylaxisoccurred.Neither subject had a shrimp allergy or any other food allergy.Oneof them (subject 4) hadmild conjunctivitis, rhinitis, nasal congestionand sneezing related to HDM or birch pollen exposure. For a detailedoverviewofthepatientcharacteristicsseeTable1.SensitizationprofileofmealwormallergicsubjectsAll four subjects were sensitized tomealworm according to ImmunoCAP(0.75kU/L,2.32kU/L,2.25kU/Land14.6kU/Lrespectively)andSPT(Table1).Theirsensitizationprofiles,toshrimp(SPT,CAP,Penm1,Penm2andPenm4)andhousedustmite(SPT,CAP,Derp1,Derp2andDerp10),areshown in Table 1. Only subject 2 showed sensitization to shrimp by IgEbinding to tropomyosin and a positive skin reaction to shrimp. However,thissubjectshowsnoclinicalsymptomswheneatingshrimp.Furthermore,all subjectswere sensitized to some common inhalant allergens, such astreeandgrasspollenandanimaldander,althoughtiterswereverylowasshownbyISAC(seesupplementalmaterial).Althoughsensitized,onlyonesubjecthadrhinitissymptoms,possiblycausedbyHDM.As Figure 1 shows, IgE from all subjects bind to more or less thesame proteins in the mealworm extracts, however with differentintensities. Comparing the subjects with a reported respiratoryallergy tomealworm (1 and 2) to the subjectswith food allergy (3and 4), no clear convincing differences were observed. Theexceptionsareaproteinataround15kDaintheTrisextractontheimmunoblotof subject 3 andaproteinof approximately 10 kDa inthe immunoblot of subject 4. The basophils of all 4 subjects wereactivatedbymealwormextracts(Figure2).ThepercentageofCD63+basophils ranged between 18 and 65. No basophil activation wasseenafterincubationwithshrimptropomyosin(Penm1).However,the basophils of subjects 3 and 4 were slightly activated afterincubationwith shrimp extract (%CD63+basophils between 5 and11). These sensitization resultswerenot fully corroboratedbySPT,whereonlysubjects2and3showedaminorreactiontoshrimp.
Chapter6
96
Table1a.Patientcharacteristics
aSex
Age(years)
AA/A
D/AR
Professio
nal/
Domestic
Yearso
fexposure
Direct
inha
lant
expo
sure(h
/w)
Oral
expo
sure
(timesco
nsum
ed)
Averageintake(in
gram
s)
Symptom
sinha
latio
n
Symptom
singestion
HDMsy
mptom
s
Shrim
psymptom
s c
Subject
DBPC
FC
mealworm
1 M 46
n/y/n
Prof
2 0.5 <10 ~1 C,R n n n neg
2 F 22
n/y/n
Prof
5 6 <5 ~1 C,N,R n n n neg
3 M 28
n/n/n
Dom
7 2-4 <10 ~50 n OA,U
n n pos
4 M 32
n/n/y
Dom
10 2-4 5-10 ~50 C,N,S,D,W
OA,U
C,N,S
n pos
Mealworm HDM Shrimp
b
CAP(kU/
L)
SPTAL
K
CAP(kU/
L)
SPTGree
r
SPTAL
K
Derp
1(ISU
)
Derp
2(ISU
)
Derp
10(IS
U)
CAP(kU/
L)
SPTGree
r
SPTAL
K
SPTStallergen
Penm1(ISU
)
Subject
OFCsh
rimp
1 0.75
3+
0 2+
0 0 1.1
0 0 0 0 0 0 neg
2 2.32
3+
0.4
2+
0 0 0 0.6
0.7 2+ 2+ 2+ 0.6 neg
3 2.26
3+
0 2+
0 0 0 0 0 0 0 1+ 0 neg
4 14.6
2+
28
3+
2+
9.7 13 0 0 0 0 0 0 neg
a.Mealwormexposureandclinicalhistorytodirectinhalation(whileworkingwiththeanimal)and ingestionofmealworm,HDMandshrimp.M=male,F= femaleAAallergicasthma;ADatopicdermatitis;ARallergicrhinitis;n=no;y=yes;Prof=professional farmer; Dom = domestic farmer; h/w hours per week; GI = gastro-intestinal; C, conjunctivitis; D, dyspnea; N nasal congestion; OA, oral allergy; R,rhinitis;S,sneezing;U,generalurticaria;W,wheezing.b.SensitizationinSPT,CAPformealworm,HDMandshrimpandISAC(withatleastonesubject<0.3ISU)wereincludedinthetable.Derp1,housedustmitepeptidaseC1;Derp2,housedustmite NPC2 family; Der p 10, house dust mite tropomyosin; Pen m 1, shrimptropomyosinc.Challengeresults.neg=negativepos=positive
Chapter6
97
Figure1.IgEbindingprofileinmiceandhumans
a. Antibody responses (IgE and IgG1) tomealworm extract, shrimp extracts andcontrol (PBS) inmicemodel on day 16. Data represent themean 2log antibodytiter ± SEM. Horizontal dotted lines represent the detection limits of the assays(min/max).
b. Subjects show IgE binding to allergens in both themealwormand the shrimpextract.Nobinding is seen toPena1 (shrimp tropomyosin) IgE frommealwormfood allergic subjects 3 and 4 additionally bind to proteins with low molecularweight.
IgEbindingproteinsidentifiedforsensitizedmiceandhumansubjectsTable 2 shows the top 6 IgE binding mealworm proteins we identifiedbased on highestmean score (which is the probability that the observedmatch isnotarandomevent) formiceandtable3providesthesameforhumans. These tables also show the top 5 known allergens (not in the
Chapter6
98
proteintop6).Forbothmiceandhumans,IgEbindingtoknownallergenssuch as tropomyosin, arginine kinase, myosin light and heavy chain wasshown. Both mouse and human IgE recognized three proteins withconvincinghighscores(5to10timeshighercomparedtotheothers)andwith high sequence coverage (with respect to our previous and currentresults). These were identified as three highly homologous (> 88%sequenceidentity)mealwormlarvalcuticleproteins(LCP).Figure2.FunctionalIgEbinding
Basophil activation tests showing basophil activation to mealworm proteins(Tris/Urea extract) and shrimp (SPT ALK). No reactivity to Pen a 1 (shrimptropomyosin)wasseen.Clinicallysubject1and2showedrespiratorysymptomstomealworm;subjects3and4showedfoodallergicsymptoms.
Chapter6
99
DiscussionIn mice, mealworm extract was able to induce primary sensitization:induction of mealworm specific IgG1 and IgE. Shrimp extract used as acontrol showed IgG1 and IgE to shrimp. The percentage of sensitizationwas higher for shrimp than formealworm,whichmight suggest strongerpotency of shrimp thanmealworm. The usedmousemodel [20] has theabilitytodifferentiatebetweenallergens(e.g.Arah1,beta-lactoglobulin)andlow/non-allergens(e.g.gelatin,beeftropomyosin),butitisnotknownwhetherthemodelcanmeasuredifferencesinsensitizingpotency.Itmightnot, given previous results from a DBPCFC study with mealworm andshrimp,which showed thateliciting thresholds for shrimpandmealwormwerecomparable,suggestingthattheallergenicityissimilar. Mealwormexposure inhumansbydomesticorprofessionalbreedingandingestion resulted in inhalant and/or food allergy tomealworm. Primaryinsect allergies have been documented in case studies in insect raisinglaboratoryworkers,baithandlersorgrainworkersasoccupationalallergies[11,23,24]. The development of insect-based occupational allergies wasdocumented for blowfly (Lucilia cuprina) and other species of adult flies,grasshopper and mealworm [11, 23, 24]. Reports of anaphylaxis due toingestion of the larvae of the silk worm (Bombyx mori), from Asia, lackhistory of prior occupational exposure. These reports disclosed neithershrimpsensitizationnorallergy,thereforeacross-reactiveresultcannotbeexcluded[8].Onlyonecaseoffoodallergytomealwormhasbeenreportedsofar,withasystemicreactionincludingpruritis,generalizedurticariaanddiarrheaaftereatingmealworm,butthiswasnotconfirmedbyDBPCFC.Inthiscase,neitheroccupationalexposurenorshrimpallergywasmentioned[9].Althoughbasedonaverylimitednumberofcases,theresultsofthisstudysuggesttheoccurrenceof3differentphenotypesofprimaryallergyduetoexposure to mealworm allergens. The 2 occupational workers who hadlimitedoral exposuredevelopedoccupational respiratory allergywhereasthedomesticbreederswhoreportedlyhadmorefrequentandhigheroralintakes developed food allergy. One of the two never experiencedrespiratorysymptomsduringtheworkwithmealworm. Thedatasuggestthat the differences in exposure patterns might result in a differentphenotype. Most food related occupational allergens have not beenproventoinducesymptomsafteringestion.
Chapter6
100
Table2. IgEbindingproteinsidentifiedinmiceProtein
(sou
rce)
Accession
Score
Sequ
ence
coverage
(%)
Peptides
iden
tified
PSM
Mass
(kDa
)
BeadsMice
LarvalcuticleproteinA1A(Tenebriomolitor)
P80681
519 97 15 114 17.7
LarvalcuticleproteinA2B(Tenebriomolitor)
P80682 300 97 10 70 12.3
LarvalcuticleproteinA3A(Tenebriomolitor)
P80683 292 97 15 70 14.0
Tropomyosinlike(Triboliumcastaneum)
D6X4X3 85 38 12 23 32.3
Actin-87E(Drosophilamelanogaster)
P10981 60 22 9 14 41.8
TM-E1A=Cuticularprotein(Tenebriomolitor)
Q9TXE4 32 27 3 9 23.2
Otherknownallergensidentified
Chitinbindingprotein(Nasoniavitripennis)
K7IX02 29 8 4 7 23.0
Myosinheavylike(Triboliumcastaneum)
D6WVJ3 27 15 9 10 262.1
TroponinTlike(Triboliumcastaneum)
D6W953 17 14 5 6 45.7
Myosinlightchain2(Triboliumcastaneum)
D6WZU7 12 9 3 4 31.3
Argininekinase(Fragment)(Lasippatiga)
D1L9H9 11 19 2 2 12.0
Proteins identified on beads with IgE from the mealworm sensitized mice afterincubation with Tris/urea extract using LC-MS/MS Top 6 IgE binding proteinsidentifiedusingLC-MS/MSbasedonhighestmeanscore(pooledsensitizedmousesera(n=6))andtop5knownallergen(notinproteintop6).Arrangedonhighestscore.IdentificationwasbasedonhomologywithmetazoanproteinsintheSwiss-Prot database. Known allergens are noted in bold, based on arthropodnomenclature. PSM = peptide-spectrum matches, value that represents thenumber of MS/MS spectra that matched peptide sequences assigned to thatparticularprotein.Score=thesumofindividualSequestscoresofalltheidentified
Proteins identified on beads with IgE from the 4 subjects after incubation withTris/urea extract using LC-MS/MSTop 6 IgE binding proteins identified using LC-MS/MSbasedonhighestmeanscore(4subjects)andtop5knownallergen(notinprotein top 6) Sequence coverage, Peptides identified andPSMare given as themeanofthe4sera.Identificationwasbasedonhomologywithmetazoanproteinsin the Swiss-Prot database. PSM = peptide-spectrum matches, value that
Chapter6
102
represents the number of MS/MS spectra that matched peptide sequencesassignedtothatparticularprotein.Score=thesumofindividualSequestscoresofall the identifiedpeptideswhichwereassigned to theprotein itself. The score istheprobabilitythattheobservedmatchisnotarandomevent.A few isolated caseshavebeen reportedonoccupational allergyand thesubsequent development of food allergy to the same source of protein.These individuals were reported in garlic, buckwheat, shrimp and snowcrabindustry[25-28].Whether this lack of evidence is due to underreporting or due to actualabsenceof foodallergicsymptoms inoccupationalallergy isunclear,dataonintakeofthework-relatedfoodaremissing.In line with our study, insect allergy, developing generally after severalyears of exposure, has been described in various papers [12,29,30]. Themanifestationofoccupationalallergicsymptoms,seenin14subjectsafter2.3to6.3yearsofexposuretomealworm,greenbottleandwaxmoth[12].Allergicsymptoms inanothercasedevelopedafteroccupationalexposuretomealwormfor5.4years[30].Fromthe fourmealworm-sensitizedhumansubjects, twodeveloped foodallergy. These two showed higher sIgE to mealworm, consumed largeramounts ofmealworm (~50 g) and were exposed for a longer period oftime(7-9years)thanthetwowithrespiratoryallergy(~1gmealwormand2-5yearsofexposure).Thismightsuggestthatoccupationalexposureforalonger period of time or oral exposure with high doses are required todevelop food allergy to mealworm. Both food allergic subjects wereexposedtomealwormthroughtherespiratory,dermalandoralrouteandthereforewecannotpinpointtheexactroute(s)ofexposureresponsible.The risk for mealworm allergy in shrimp allergic patients through cross-reactivitywasshown inourpreviousstudy.Thepresentstudyshowsthatprimary respiratory and food allergy to mealworm can develop due todermal, respiratoryand/ororalexposure.Therelativecontributionof theoralexposure routecannotbeestablished,although it is remarkable thatthe 2 subjects that developed food allergy to mealworm had a morefrequent and higher preceding consumption ofmealworm. All 4 subjectsthat developed primary allergy to mealworm were atopic patients, butwithout any food allergy. So far there are no indications that also non-
Chapter6
103
atopic individuals could be at relevant risk for development of primaryallergyduetoexposuretoorconsumptionofmealworms.Ourdatasuggestthat,sincethreeofoursubjectsproducedhigherlevelsofsIgE tomealworm than to any other food or inhalant allergen (includingother arthropods allergens), mealworm might be the primary sensitizer.Although the fourth subject showed slightly higher HDM sensitizationcompared to mealworm, this subject had IgE to Der p 1 and 2 (fecalcomponents),butnottoDerp10(tropomyosin)andargininekinase,whichmakescross-reactivitytoHDMlesslikely.Furthermore,immunoblotsfromshrimpallergicpatientsshoweddifferentbindingpatternscomparedtothefourprimarymealwormallergicsubjects[13].SincethesIgEtoshrimpandHDMfromthefourmealwormallergicsubjectswaslow,inhibitionstudieswere impossible.However,a role for shrimpwasunlikelygiven that foodchallengewithshrimpwasnegativeinallmealwormallergicpatients.In all four human subjects, the mealworm proteins involved were thepreviouslyundiscoveredallergenlarvalcuticleproteinand,toaseeminglylesserextent,thewell-knowntropomyosinandmyosinheavychain.Someof theseproteins (e.g. tropomyosin, arginine kinase,myosinheavy chain)were also involved in cross-reactive food allergy tomealworm in shrimpallergicpatients[22].ThemostabundantIgEbindingproteinsidentifiedinthisstudy, larvalcuticleproteins (LCP),werenot identifiedwiththesamehighscoreandhighsequencecoverage(noteveninthetop20proteins)inshrimpallergicpatientsofourfirststudy[22].ThismightindicatethatLCPis a less dominant allergen in shrimp allergy compared to mealwormallergy.Theselarvalcuticleproteinshavenotbeenpreviouslyidentifiedasallergens in insects or crustaceans, so we have discovered a novelarthropodallergeninmealworm.LCP is a mealworm specific protein, having a conserved domain inarthropod cuticles known as R&R consensus and binds chitin [32]. Thechitin-binding complex links the soft internal tissue to theexoskeletonofthe larvae. The LCPs identified are highly homologous, with a molecularweightvaryingfrom12.3to17.7kDa.Thelackofbandsontheimmunoblotat17kDamightbeaccountedforbymigrationongeltoahighermolecularweight.Thisdiscrepancybetweenmigrationongelandmolecularweightischaracteristic of many cuticle proteins. Aggregated forms of cuticleproteins exist and can appear at higher MW on the immunoblot,dependingonuseofurea,temperatureandpH[28].Giventhedifferences
Chapter6
104
between the shrimp allergic, cross-reactivemealworm allergy (low scoreandsequencecoverage)andthefoursubjectsinthisstudy(highscoreandsequence coverage), we hypothesize that LPC could be the principlesensitizingallergeninprimarymealwormallergy.LCPsfrommealwormaremost likelyboth inhalantandfoodallergens,sincetheywere identified insubjectswithrespiratoryaswellasfoodallergytomealworm.ConclusionExposuretomealwormcanleadtoprimarysensitizationinamousemodeland in humans and can result in food allergy. The risk of primarysensitizationandallergy shouldbe incorporated in the riskassessmentofnewproteins.SupplementaryTableE1.ImmunoCAPISACresults
Allpatientsshowsomesensitizationtobothcommoninhalantandfoodallergens:values (>0.3) expressed in ISU. Only scores that were positive in at least onesubjectareshown.Penm1,2,4fromshrimp;Derp1,2,10fromhousedustmite;Blag7fromcockroach;Anis3fromanisakis;Cynd1fromBermudagrass;Phlp1,2,4,5 from timothy grass; Cry j 1 from Japanese cedar; Cup a 1 from Arizonacypress;Plaa2fromplanetree;Canf5fromdog;Feld1fromCat;Derf1,2fromhouse dust mite; Lep d 2 from storage mite; Jug r 2, 3 from walnut; MUXF3bromelain.
Subj.
Penm
1 Derp
1
Derp
2
Derp
10
Blag7
Anis3
Cynd1
Phlp
1
Phlp
2
Phlp
4
Phlp
5
1 0 0 1.1 0 0 0 0 0 0 0 0
2 0.6 0 0 0.6 0.6 0.8 2.3 0 0 2.0 0
3 0 0 0 0 0 0 2.0 0 0 2.8 2.1
4 0 9.7 13 0 0 0 14 20 1.2 0 3.6
Subj.
Cryj1
Cupa
1 Plaa
2 Canf
5 Feld
1 De
rf
1 Derf
2 Lepd
2
Jugr
2
Jugr
3 MUX
F3
1 0 0 0 0 0 0 2.1 5.4 0 0 0
2 0 0 0.4 0 0 0 0 0 2.4 0 1.7
3 0.5 1.0 1.5 0 0 0 0 0 1.9 0 2.0
4 0 0 0 1.0 6.2 10 24 0 0 0.3 0
Chapter6
105
Figure E1 Three mealworm allergic subjects sensitized to larval cuticle proteinA1Aonimmunoblot
Both food allergic subjects 3 and 4 show IgE binding tomealworm larval cuticleproteinA1A.IgEfromoneinhalantallergicsubjectboundtoLCPA1A.
Chapter6
106
References:1. FAO. FAO Forestry Paper. Edible insects: future prospects for food and
5. Piromrat K., Chinratanapisit S., Trathong S. Anaphylaxis in anemergencydepartment: A 2-year study in a tertiary-care hospital. Asian PacificJournalofAllergyandImmunology,2008.26:121-128.
6. Ji KM, Zhan ZK, Chen JJ, Liu ZG. Anaphylactic shock caused by silkwormpupaconsumptioninChina.Allergy,2008.63(10):1407-1408.
7. Bock SA Incidence of severe adverse reactions to food in Colorado, JAllergyClinImmunol.1992;90:683-685.
8. Sampson HA. Anaphylaxis and emergency treatment. Pediatrics 2003.111(6):1601-1608.
9. FreyeHB,EschRE,LitwinCM,SorkinL.Anaphylaxistothe ingestionandinhalation of Tenebrio molitor (mealworm) and Zophobas morio(superworm).AllergyAsthmaProc,1996.17(4):215-9.
11. Baldo BA, Bellas TE, Tovey ER, Kaufman GL. Occupational allergy in anentomological research centre. II. Identification of IgE-binding proteinsfrom developmental stages of the blowfly Lucilia cuprina and otherspeciesofadultflies.JAllergyClinImmunol1986;77(1):108-13
13. Broekman H, Verhoeckx KC, den Hartog Jager CF, Kruizinga AG, et al.Majority of shrimp allergic patients are allergic to mealworm. J AllergyClinImmunol.2016.137(4):1261-3.
14. Liu Z, Xia L,Wu Y, XiaQ, et al. Identification and characterization of anargininekinaseasamajorallergenfromsilkworm(Bombyxmori) larvae.IntArchAllergyImmunol,2009.150(1):8-14.
15. Binder M, Mahler V, Hayek B, Sperr WR. et al., Molecular andimmunological characterization of arginine kinase from the Indianmealmoth, Plodia interpunctella, a novel cross-reactive invertebrate pan-allergen.JournalofImmunology,2001.167(9):5470-7.
Chapter6
107
16. Gamez C, ZafraMP, BoqueteM, Sanz V. et al., New shrimp IgE-bindingproteins involved inmite-seafoodcross-reactivity.MolecularNutrition&FoodResearch,2014.58(9):1915-25.
17. Berin MC, Mayer L. Immunophysiology of experimental food allergy.MucosalImmunol.2009.2:24-32.
19. Smit JJ, Bol-Schoenmakers M, Hassing I, Fiechter D, et al. The role ofintestinal dendritic cells subsets in the establishment of food allergy.Clin.Exp.Allergy.2011.41:890-898.
20. Bowman CC, Selgrade MK. Differences in allergenic potential of foodextractsfollowingoralexposureinmicereflectdifferencesindigestibility:potential approaches to safety assessment. Toxicol.Sci. 2008. 102:100-109.
24. LybargerJA,GallagherJS,PulverDW,LitwinA,etal.Occupationalasthmainduced by inhalation and ingestion of garlic. J Allergy Clin Immunol1982;69:448-54.
26. Nakamura S, Yamagachi M, Oishi M, Hayama T. Studies on buckwheatallergose report 1: on the case with the buckwheat allergose. AllergImmunol(Leipz)1974;20-21:449-56.
28. Cartier A, Malo JL, Forest F, LafranceM, et al. Occupational asthma insnowcrab-processingworkers.JAllergyClinImmunol1984;74:261-9.
29. Armentia A, Martinez A, Castrodeza R, Matínez J, et al. Occupationalallergicdisease in cerealworkersby storedgrainpests. JAsthma.1997.34(5):369-78.
Chapter7:Ismealwormorshrimpallergyindicativeforfoodallergytoinsects?AbstractScope:Thegrowingworldpopulationisakeydriverfortheexplorationofsustainableproteinsourcestoensurefoodsecurity.Mealwormandotherinsectsarepromisingcandidates.Previouslywefoundthatshrimpallergicpatientsareatriskformealwormallergy,andthatmealwormcaninduceaprimaryallergy[3].This studysetout to investigate theallergenicpotentialofedible insects,suggestedforhumanconsumptionbyagenciessuchasWHO/FAO,inboththe shrimp (potentially cross-reactive) and primary mealworm allergicpopulation.The following insectswerestudied:mealworm,housecricket,giantmealworm, lessermealworm, African grasshopper, large waxmothandblacksoldierfly.Methodsandresults:15shrimp(mealwormsensitizedorallergic)patientsandfourprimarymealwormallergicsubjects,whoparticipatedinpreviousstudies, were included. All shrimp allergic patients were sensitized tomultipleinsectswithsimilarresponseprofilesforallinsectstested.Primarymealworm allergic patients, showed IgE binding to proteins from only afewinsectsonimmunoblot,althoughbasophilactivationtestwaspositiveforalltestedinsects.Conclusion:Shrimpallergicpatientsareatriskoffoodallergytomealwormandotherinsects.Primarymealwormallergicsubjectsaresuspectednottoreacttoallinsects.IntroductionIn order to preserve agricultural land, energy andwater and reduce CO2
footprint, changes to our current systems of food production must bemade. Therefore, on recommendation of the FAO, governments aresanctioning investigations aimed at determining if certain insects can beconsumedasanewsustainableproteincandidateforfeedandfood[1,2].Intworecentstudies,weshowedthatthemajorityofshrimpfoodallergicpatients also hadmealworm food allergy [3]. In addition, we found thatprimary mealworm allergy can develop in professional and hobby insectbreeders[4].
Chapter7
112
PrevalenceoffoodallergyrangesinEuropeupto5.7%dependingonage,withshellfishamongthetenmostprevalentfoods[5].PrevalenceofinsectfoodallergyinEuropeisnotknown.Atpresent insectsarenotcommonlyconsumedinWesterncountries.Therearesomereportsonfoodallergytoinsects in Asia, which are especially focused on the pupal stage of thesilkworm(Bombyxmori),whichisregularlyconsumedinAsia.Asaresultofthe cultural influences of traditional Chinese food andmedicine, Chinesepeople often eat oil-fried, water-boiled or ground pupa powder of thesilkwormandanaphylaxishasbeendescribeduponingestion[6].InChina,it was estimated from literature (1980 -2007) that around 17 % of foodrelatedanaphylaxiswascausedbyeating insects(locust,grasshopperandsilkworm)[7].FoodallergytomealwormwasreportedpreviouslyintheUSbyFryeetal. [8].Fryeetal.,presentedacaseofanaphylaxis inapatientafter ingestion ofmealworm. In a previous study,we demonstrated thatshrimpallergicpatientswerefoodallergic tomealworminadouble-blindplacebo controlled food challenge (DBPCFC). This allergy was based oncross- or co-sensitization to multiple mealworm allergens such astropomyosinandargininekinase[3].Ininsects(e.g.cockroachandcricket)and arthropods (e.g. shrimp and crab) homologous proteins such astropomyosin and arginine kinase have been described, which maycontributetocross-reactivitybetweenshrimp,mealwormandotherinsectspecies[9,10].Thiscouldsuggestthatshrimpandmealwormallergymightbe indicative for allergy to other insects. In our previous study, we alsoshowed that primary mealworm sensitized subjects were not allergic toshrimp.Thestudysuggestedthatprimarysensitizationformealwormwasnot caused by tropomyosin and arginine kinase, but that other proteins,suchaslarvalcuticleprotein(LCP)mightplayarole[4].Thisassertionisatconflict with the shrimp cross-reactivity theory and might indicate thatprimary mealworm sensitization is not indicative for allergy to otherinsects.Beforeinsectscanbecomeasubstantialpartofthediet,potentialrisksoffood allergic reactions should be explored. In this paper, the risk of foodallergytoinsectsotherthanmealwormisaddressed.Seveninsectsfrom4different orders (Coleoptera, Lepidoptera, Diptera and Orthoptera) anddifferentlifestages(larvaeandadult)wereinvestigated.Theseinsectsaredescribed in the EFSA report as possible novel food or feed [11].Sensitization and functional IgE binding to Tenebrio molitor; mealworm(larvae), Acheta domesticus; house cricket (bug), Zophobas morio; giantmealworm (larvae), Alphitobius diaperinus; lesser mealworm (larvae),
Chapter7
113
Locusta migratoria; African grasshopper (bug), which we will callgrasshopper, Galleria mellonella; large wax moth (larvae) and Hermetiaillucens; black soldier fly (larvae) was tested in blood from patients withshrimpallergyoraprimarymealwormallergyfromourpreviousstudies[3,4]. Proteins were identified, using LC-MS-MS, in the different insectextractsforcomparativeanalysis.MaterialsandmethodsStudypopulation15 shrimp allergic patients and four primary mealworm allergic subjectsshowing sensitization to mealworm from our previous studies, wereincludedinthestudy[3,4].13outof15shrimpallergicpatientshadafoodallergy to mealworm as indicated by a positive DBPCFC. Two primarymealworm allergic subjects had a positiveDBPCFC tomealworm and theothertwohadaninhalantallergytomealworm.Allsubjectsgavewritteninformedconsentbeforeparticipation.Thestudywasapprovedbythelocalethicscommittee(NL43731.041.13).SpecificIgEImmunoCAP mealworm was specifically produced for this project byThermoFisherScientific,Uppsala,Swedenandwastestedaccordingtothemanufacturer’s recommendations. IgE is expressed in kU/L. Tests wereconsideredpositivewithavalueof0.35orhigher.InsectextractsFresh Yellowmealworms (Tenebriomolitor), giantmealworms (Zophobasmorio), lesser mealworms (Alphitobius diaperinus), large wax moths(Galleriamellonella) and black soldier flies (Hermetia illucens) all in finallarval stage, and crickets (Acheta domesticus) and grasshoppers (Locustamigratoria migratorioides) in adult form were kindly provided by Dutchinsect farm Kreca (Ermelo, the Netherlands). Dutch shrimp (Crangoncrangon)wereboughtfromalocalstore.Five grams of insects or boiled shrimpwere extracted using a sequentialproteinextractionmethod[12].Firsttheinsectsorshrimpweremixedwith25 mL ice-cold Tris buffer (20 mM Tris buffer pH 7.6 containing 1 mMphenylthiocarbamide (Sigma Aldrich) and Halt Protease Inhibitor Cocktail(Thermo Scientific). Subsequently the insects or shrimp were disrupted,usinganultraturrax(3x10sec)undercontinuouscooling.Theultraturraxwaswashedwith5ml coldTrisbuffer and thewash liquidwasadded to
Chapter7
114
thesamplesuspension.Aftercentrifugation(30min,15000xgat4°C),thesupernatantwasrecovered.Theinsolubleresiduewaswashedoncewith5mL Tris buffer. The 30mL and 5mL supernatantwere combined. 25mLwas used for sample cleanup and concentration using TCA precipitation.Secondly,theremainingpelletwasextractedovernightat4°Cwith30mLurea buffer (6 M urea in 20 mM Tris buffer pH 7.6 containing 1 mMphenylthiocarbamideandHaltProteaseInhibitorCocktail).Thesamplewassubsequently centrifuged and the supernatant was collected. The pelletwas washed once more with 5 mL urea buffer, centrifuged and thesupernatantwascombinedwiththe30mLureasupernatant.25mLoftheextractwasTCAprecipitated.Trisandureaextractswerecombined(1:1).ImmunoblotanddotblotForSDS-PAGE,theCriterionsystemwithanAnykDReadyGel®Tris-HClgel(Bio-Rad, Hercules, CA, USA) was used according to the manufacturer’sinstructions.Insectextracts(5µg)wereloadedonthegel,underreducingconditions (Laemmli buffer) together with 5 µg shrimp control. Afterseparation, proteins were transferred to a polyvinyldifluoridemembraneusing the Criterion Blotter system (Bio-Rad) according to themanufacturer’sinstructions.The membrane was blocked overnight with 5% dried milk powder inphosphate buffered saline with 0.1% Tween 20 (PBST) and incubatedovernightwithserumfromashrimpallergicpatientorprimarymealwormallergic subject (1:50) in 1 % driedmilk powder in PBST. After thoroughwashing,themembraneswereincubatedfor1hourwithHRP-labeledGoatanti human IgE (KPL, Gaithersburg, MD, USA) 1:50.000 in PBST. Afterwashing,thebandswerevisualizedusingachemi-luminescentperoxidasesubstrate kit ECL (Sigma) according to the manufacturer’s instructions.Control blots were performed to exclude a-specific binding. Results areshown in figure E1. Blots were scanned using the Chemidoc XRS+ imagescannerwithImagelabsoftware(Bio-Rad).Dot blots were performed on a PVDF membrane, using a Bio-DotMicrofiltration Apparatus (Bio-Rad, Hercules, CA, USA) according to themanufacturer’s instructions. Insectextracts (5µgofprotein)were loadedon the membrane, under reducing conditions (Laemmli buffer) togetherwith 5 µg shrimp control. Reducing conditionswere used on dot blot tokeep the binding potential similar to the proteins onWesternblot. After
Chapter7
115
loading theproteinon themembrane, the sameproceduredescribed fortheimmunoblotwasperformed.Basophilactivationtest(BAT)BAT’s were performed with blood from shrimp allergic patients andprimarymealwormallergic subjects fromourprevious studies [3, 4]. Fiveshrimp allergic patients were excluded from the BAT because they werenon-responders, showing only IgE binding to freeze dried mealworm orspontaneousrelease inpreviousBAT.Oneshrimpallergicpatientwasnotable to participate as a result of emigration. BAT’s were performed asdescribedpreviouslybyMeulenbroeketal.[13]withminormodifications.Cells were incubated with a dilution series (1:107 to 1:102) of insectextracts (5mg/mL).Mealwormor shrimpextract (5mg/mL)was used aspositivecontrol.CD63,CD123andCD203cexpressionwasanalyzedbyflowcytometryusingFACSCantoIIandFACSDivasoftware(BDBioscience,USA).The resultswereexpressedasapercentageofCD63+basophils. TheBATwasconsideredpositivewhen thepercentageofCD63+cellswasat least5% and no spontaneous expression of CD63 on the cells wasmeasured.Individuals with basophil response of 0-5% CD63, after anti-FcƐRIstimulation, i.e. thepositivecontrol,wereregardedasnon-responders. Intotal13BATswereapproved.NanoLC-MS/MSanalysisandproteinidentificationProteinsfromtheTris/ureamixturesofeach insect (20μg)weredigestedand analyzed according to Verhoeckx et al. [14], with small changes.Chromatography was performed on an Easy LC 1000 nanoscale liquidchromatography (nanoLC) system (Thermo Fisher Scientific, Odense,Denmark).TheanalyticalnanoLCcolumnwasapulledfusedsilicacapillary,75 μm i.d., in-house packed to a length of 10 cm with 3 μm C18 silicaparticles fromDr.Maisch (Entringen,Germany).Trypticpeptides (125ng)were loaded at 500 nL/min directly onto the analytical column. A binarygradient was used for peptide elution.Mobile phase A was 0.1% formicacid,2%acetonitrile,whereasmobilephaseBwas0.1% formicacid,80%acetonitrile. Gradient elutionwas achieved at 350 nL/min flow rate, andramped from 8% B to 35% B in 60 min, and from 30% B to 100% B inadditional8min.After5minat100%B,thecolumnwasre-equilibratedat0%Bfor2minbeforethefollowinginjection.MSdetectionwasperformedon a quadrupole-orbitrap mass spectrometer Q-Exactive (Thermo FisherScientific, Bremen, Germany) operating in positive ion mode, with
Chapter7
116
nanoelectrospray (nESI)potential at1800Vappliedon the column front-endviaateepiece.Data-dependentacquisitionwasperformedbyusingatop-12 method with resolution (FWHM), AGC target and maximuminjectiontime(ms)forfullMSandMS/MSof,respectively,70,000/17,500,10e6/10e5, 50/60.Masswindow for precursor ion isolationwas 1.6m/z,whereas normalized collision energy was 25. Ion threshold for triggeringMS/MSeventswas2e4.Dynamicexclusionwas30s.Datawasprocessedusing Proteome Discoverer 1.3 (Thermo Fisher Scientific, Bremen,Germany), using Sequest as search engine, and the Swiss Prot databaseaccessedonSeptember2016assequencedatabase(8,109,403sequencesforMetazoa taxonomy). The following searchparameterswereused:MStolerance 15 ppm; MS/MS tolerance 0.02 Da; fixed modificationscarbamidomethyl cysteine; enzyme trypsin; max. missed cleavages 1.Search results were filtered by q-values using Percolator integrated inProteome Discoverer, to achieve a peptide-level FDR of less than 1%.Minimumpeptidecountwas2.AnalysisDescriptiveanalyseswereperformedusingSPSSInc,Chicago,version21.0.ResultsShrimpandprimarymealwormallergicpatientsThe 15 shrimp allergic patients, of whom 13 had a food allergy tomealworm,hadamedianageof38years(range19-69)and47%wasmale.SensitizationtoshrimpwasshownbyapositiveImmunoCAP(>0.35kU/L)with a median of 5.5 kU/L (range 0.37 to 53.3 kU/L). The majority hadinhalant allergies to HDM (11/15) and pollen (11/15). 8 had atopicdermatitisand9patientshadoneormoreotherfoodallergies.Noneofthepatients had knowingly consumed mealworm or other insects [3]. Fourprimary mealworm allergic subjects regularly exposed to some of thetested insects were also studied. In addition to mealworm, cricket wasconsumedbysubjects1,3and4,grasshopperbysubjects2,3and4andlessermealwormbysubjects1,2and4.Waxmothwasconsumedonlybysubject4andgiantmealwormonlybysubject3.Thesuperwormandblacksoldier fly had never been consumed by any of the primary mealwormallergic subjects.Oralallergy symptomswere reported forgrasshopper insubjects2and3,andcricketinsubject4;however,thiswasnotconfirmedwithafoodchallenge.
Chapter7
117
SensitizationtoallinsectsinshrimpallergicpatientsSensitization to mealworm was shown by a positive ImmunoCAP tomealworm(>0.35kU/L)inallshrimpallergicpatientswithamedianof5.8kU/L (range 0.64 to 19.1 kU/L), see Table 1. IgE binding on dot blot toextractsofdifferentinsects,wasfoundforallpatients.ThisIgEbindingwascomparable to that ofmealworm extract, except for two patients. ThesepatientsshowedonlyIgEbindingtoproteinsofafewinsects,asshowninTable 1 (representative dot-blots can be found in figure E2). Basophilreactivity to all insect species was highly similar, see Table 2. The BATresults corroborated the dot blot findings, and showed to be somewhatmoresensitivethanCAP.Cross-reactive proteins were identified using immunoblot. Differentprotein profiles were recognized by the shrimp allergic patients. Themajority (9/15) recognizedabandatapproximately40kDa. Identificationof this band in mealworm previously showed that it contained bothtropomyosinandargininekinase.Patient12showedalso IgEbinding toa50kDaproteinsandpatients4,8,and10mainlyrecognizedbandswithamolecularweight>100kDa, includingabandat200kDa.Patients9and11, who had no food allergy to mealworm, recognized proteins withdifferentmolecularweightscompared to theother13patients.Figure1ashows thecoomassie stainedproteingelof the insectextractsand figure1bshowsrepresentativeimmunoblotsofpatients2,4,9and12.InsectsensitizationofprimarymealwormallergicsubjectsTheprimarymealwormallergic patientswere all sensitized tomealwormand subjects 3 and 4 had a food allergy to mealworm. Only subject 2showedminor sensitization to shrimpbut hadno food allergy to shrimp,which was tested in a food challenge [4]. Although all subjects weresensitized to somecommon inhalantallergens (treeandgrasspollenandanimal dander), only subject 4 had mild rhinoconjuctivitis to HDM andbirch pollen. Subjects 1 and 2 had atopic dermatitis and none had anyother food allergy [4]. In contrast to the shrimp allergic group, primarymealwormallergic subjectsdidnot showsensitization toall tested insectextractsondotblot(Table1).However,sensitizationwasshownintheBATto all insects,withdifferences in reactivity. Subjects 2, 3 and4,whohadexperienced clinical symptoms after eating grasshopper and cricket,showed a positive BAT for these insects (Table 2). Dot blot, immunoblotandbasophilreactivityweredifferentforallinsects.Allprimarymealworm
Chapter7
118
allergic subjects recognized proteins between 10-200 kDa depending onthe insect tested. Protein binding was mainly seen for black soldier fly,giantmealworm,cricketandwaxmoth,whichconfirmsthedotblotdata.StrikingisthatIgEfromthesesubjectsdidnotrecognizebandswithanMWof tropomyosin or arginine kinase. See Figure 1 for two representativeimmunoblotsofthefoodallergicsubjects3and4.Table 1. CAP mealworm and insect extract recognition on dot blot by shrimpallergicpatientsandmealwormallergic subjects.Cap IgEvaluesareexpressed inkU/L,+indicatesapositivesignaland-anegativesignalonthedotblot.
Table 2. BAT sensitization patterns to different insects, for shrimp patients andmealwormallergicsubjects.
The results are expressed as a percentage of CD63+ basophils. The grey colorcodes represent the ratio ofmaximum% of CD63 up-regulationwith respect tomealworm.mw=melaworm
<0.25 0.25-0.5 0.5-0.75 0.75-1.25 1.25<
NanoLC-MS/MSanalysisProteinsinthedifferentinsectextractswereidentifiedandanalyzedusingLC-MStoclarifythesensitizationresults tothedifferent insects. Itcanbeconcluded from Table 3 that both tropomyosin and arginine kinasewerepresent inall insectextracts.Awiderangeofotherproteinsandputativeallergens (e.g., myosin light chain and triosephosphate isomerase) werealso identified in the different insect species (Table 3). The previouslydescribed new allergen, mealworm larval cuticle protein (LCP) was also
present in the giantmealwormand cricket extract, butwas not found intheotherinsectspeciestested.DiscussionShrimp allergic patients with food allergy to mealworm showed IgEreactivity(blotandBAT)toallinsectextracts.ThemainIgEbindingproteinsweretropomyosinand/orargininekinase(9/13).Bindingtootherproteinswith a MW of ~50 kDa and MW > 100 KDa was also seen. Primarymealworm allergic subjects showed sensitization to some tested insectextracts.No clear similarities between insect sensitization patterns on theimmunoblot were seen for the primary mealworm allergic subjects.Furthermore,basophilreactivitytotheinsectextractstestedwasdifferentforthesesubjects.Figure1aCoomassiestainedgeloftheinsectextractsusedinimmunoblot
That IgE from shrimp allergic patients binds to proteins from differentinsectsisnotsurprisingbecausecrustaceansandinsectsbothbelongtothesamephylum(Arthropoda).WithinthecladePancrustacea,shrimpbelongto the sub-phylum crustacea and insects belong to the sub-phylumHexapoda[15].Asaresultofthisphylogeneticrelation,homologybetweenproteins of shrimp and different insects can be expected and has beenpreviously documented. Sequence identity between arginine kinase fromshrimp(LitopenaeusVannamei)andsilkworm(Bombyxmori)wasreportedtobe83%[16]andfortropomyosinfromhouseflyandDutchshrimp76%
Chapter7
122
and up to 86% between cockroach and non-biting midge [17, 18]. Highsequence homology is indicative for cross-reactivity, but must beconfirmedbyadditionaltestssuchasIgEbinding(immunoblot),functionalIgE binding (basophil activation test or skin prick test) or by a foodchallenge-thegoldstandardinfoodallergyresearch.Only a few reports show cross reactivity between shrimp and insects.Srinrochetal.usedpooledserumfrom16prawnallergicpatientstoshowcrossreactivity,onimmunoblot,withargininekinasefromdeAfricanfieldcricket (Gryllus bimaculatus) and Lanchester’s freshwater prawn(Macrobrachium lanchesteri) [19].Anotherpaper showedcross-reactivity,also using immunoblotting, between arginine kinase of Bombay locust(Patangasuccincta)andIgEfrompooledprawnallergicpatientsera(n=16)[20].CrossreactivityisalsosupportedbythesimultaneousIgEreactivitytodifferent tropomyosins (e.g. cockroach, HDM, anasakis) on ImmunoCAPISAC, a diagnostic tool used in food allergy diagnosis. To our knowledge,this is the first study to investigate cross-reactivity between IgE fromshrimp and mealworm allergic patients with house cricket, giantmealworm, lesser mealworm, African grasshopper, large wax moth andblacksoldierfly.
Aconsiderablenumberofarthropodallergens (e.g. tropomyosin,argininekinase,myosinlightchainandtriosephosphateisomerase)wereidentifiedintheinsectextractsusingLC-MS/MS,whichcouldalsobeinvolvedinthecross-reactivity. Tropomyosin and arginine kinase were identified in allinsectextracts,andmostpatientsshowedreactivitytotheseallergens.
Details on protein identification such as score, sequence coverage, peptidesidentifiedetc.canbefoundinthesupportinginformation.
Chapter7
124
Thissupportsthepossibilityofcross-reactivitytoinsectsandconfirmsthecomparable IgE binding and BAT reactivity results for the shrimp allergicpatients. We therefore speculate that tropomyosin and arginine kinasewerethemostdominantallergensresponsibleforcrossreactivitybetweenshrimp and insects, but cross-reactivity to the other allergens cannot beexcluded.Incontrasttotheshrimpallergicgroup,wheresimilarreactivity’swere seen between the different insect extracts, primary mealwormsensitized subjects showed variability in the degree and percentage ofsensitization to thedifferent insects.Theyshowed lackof sensitization tosome tested insect extracts (e.g. grasshopper, black soldier fly) inimmunoblot and variable activation intensities in the BAT. Primarymealworm allergic subjects hardly recognized the pan allergenstropomyosinandargininekinase,which is in linewithourprevious studywhere we showed that mealworm larval cuticle proteins instead oftropomyosin and arginine kinase seem to play a principal role in primarymealwormallergy[4].Comparison of sequences from insect cuticle proteins was previouslyperformed by Andersen et al. [21]. Some homology was found betweencuticle proteins from different insect orders, while most have quitedissimilar sequences [21]. This is strengthened by the fact that LCP wasonly identified in two other insects in our study. A common peptidesequence was found in giant mealworm and cricket extracts, which wasalsodocumentedfortheircladesbyAndersenetal.[21].Thiscouldexplainthe possible allergic reaction one of our primary mealworm allergicsubjectsexperiencedwhenconsumingcricketandnotwithother insects.The low homology of inter-species larval cuticle proteins, the differentproteinbindingprofiles seenon immunoblot,andvariation inBAT,mightsuggest primary sensitization to the different insects could be caused bydifferent proteins. This indicates mealworm allergy is not indicative forinsectallergyandsuggests thepossibilityof species-specific insectallergywhenprimarilysensitizedtoinsect-specificproteins.Speciesspecificinsectallergyhasbeendescribedpreviouslyforhousefly(Muscadomestica)andcockroach[22,23].BindingofIgEfromapersonsolelyallergictohouseflywas inhibited by housefly in ELISA, but onlymildly by the closely relatedlesserhousefly.No inhibitionwasseenwithblowfly (Luciliaspp.), fruit fly(Drosophila spp.), horsefly (Haematopota pluvialis) and mosquito (Culexpipiens)[22].SpecificinsectallergywasalsodescribedbyLopataetal,whodemonstrated mono-sensitivity to American cockroach (PeriplanetaAmericana) or German cockroach (Blattella Germanica) in 17 out of 38
Chapter7
125
subjects [23]. Siracusa et al. showed inhibition neither from wax moth(Galleria mellonella) and mealworm (Tenebrio molitor) on greenbottle(Lucilla caesar), nor from greenbottle and mealworm on wax moth, norfrom greenbottle andwaxmoth onmealworm [24]. These results are inline with our data, where the primary sensitized subjects can consumemanyotherinsectswithoutsymptoms.Because of their quite homogenous reaction to all tested insect extracts,combinedwiththeallergicreactiontomealwormin87%ofourpreviouslystudiedshrimpallergicpatients[3],weexpectthatshrimpallergicpatientscouldlikelyhaveaclinicalreactionwheneatingotherinsects.Theclinicalrelevanceofmealwormsensitizationinshrimpallergicpatientswashigh(87%).Informationonpredictivevaluesofinsectsensitizationandfood allergy are scarce. Some information can be found on insectsensitization and inhalant allergies. Siracusa et al. found sensitization in31.6%of76workerswithlivefishbait(e.g.mealwormandwaxmoth)toatleastoneinsect.Workrelatedasthmaandrhinitiswasseenin29%ofthoseworkers [25]. Another study investigating occupational allergy tograsshopper (Locusta migratoria) [26] could not differentiate betweensymptomatic and asymptomatic patients, based on sensitization results.Fromeight sensitized subjects thatworked inabreeding facility, fivehadclinicalsymptomswhenworkinginthegrasshopperfacility.Fromthefourprimary mealworm sensitized subjects in our study, three developedrespiratory symptoms when working with mealworm [4]. No otherresearchhasbeenperformedstudyingfoodallergytoinsects.Therefore,itis difficult to drawany conclusionson the clinical relevanceof the insectsensitization inprimarymealwormallergic subjects.DBPCFC isneeded totranslatesensitizationtoclinicaloutcome.Basedontheinformationpresentedinthispaperweconcludethatshrimpallergicpatientsareprobablyatrisknotonlyforfoodallergytomealwormbut also to other insects. Given the variability in sensitization to otherinsects,primarymealwormallergic subjectsaresuspectednot to react toallinsects.
6. Ji KM, Zhan ZK, Chen JJ, Liu ZG. Anaphylactic shock caused by silkwormpupaconsumptioninChina.Allergy,2008.63(10):1407-1408.
7. JiKM,ChenJ,LiM,LiuZ,etal.Anaphylacticshockandlethalanaphylaxiscaused by food consumption in China. Trends in food science &Thechnology,2009.20:227-231
8. FreyeHB,EschRE,LitwinCM,SorkinL.Anaphylaxistothe ingestionandinhalation of Tenebrio molitor (mealworm) and Zophobas morio(superworm).AllergyAsthmaProc,1996.17(4):215-9.
9. Binder M, Mahler V, Hayek B, Sperr WR, et al. Molecular andimmunological characterization of arginine kinase from the Indianmealmoth, Plodia interpunctella, a novel cross-reactive invertebrate pan-allergen.JournalofImmunology,2001.167(9):p.5470-7.
10. Reese G, Ayuso R, and Lehrer SB. Tropomyosin: an invertebrate pan-allergen.IntArchAllergyImmunol,1999.119(4):p.247-58.
11. EFSA Scientific Committee, Risk profile related to production andconsumptionofinsectsasfoodandfeed.EFSAJournal2015;13(10):4257
13. MeulenbroekL.A,deJongRJ,denHartogJagerCF,MonsuurHN,etal.IgGantibodies infoodallergyinfluenceallergen-antibodycomplexformationand binding to B cells: a role for complement receptors. J. Immunol.2013,191,3526–3533.
15. Regier JC, Shultz JW, Zwick A, Hussey A, et al. Arthropod relationshipsrevealedbyphylogenomic analysisofnuclearprotein-coding sequences.Nature,2010,463(7284):1079–1084
Chapter7
129
16. Liu Z, Xia L,Wu Y, XiaQ, et al. Identification and characterization of anargininekinaseasamajorallergenfromsilkworm(Bombyxmori) larvae.IntArchAllergyImmunol,2009.150(1):8-14.
17. Romero MR, Claydon A.J, Fitches EC, Wakefield ME, et al. Sequencehomology of the fly proteins tropomyosin, arginine kinase and myosinlight chain with known allergens in invertabrates. Journal of insects asfoodandfeed,2016;2(2):69-81.
18. Jeong,K.,Y.,Yum,H-J.,Lee, I.,Y.,Ree,H.,I.,etal.,MolecularCloningandCharacterizationofTropomyosin,aMajorAllergenofChironomuskiiensis,a Dominant Species of Nonbiting Midges in Korea. Clin Diagn LabImmunol.2004Mar;11(2):320–324.
19. Srinroch, C., Srisomsap, C., Chokchaichamnankit, D., Punyarit, P., et al.Identificationofnovelallergeninedibleinsect,Gryllusbimaculatusanditscross-reactivity with Macrobrachium spp. allergens. Food Chemistry,2015;184:160-166.
20. Phiriyangkul P, Srinroch C, Srisomsap C, Chokchaichamnankit D, et al.Effect of food thermal processing on allergenicity proteins in BombayLocust (Patanga Succincta). International Journal of Food Engineering2015;1(1)23-28
21. Andersen SO, Rafn K, Roepstorff P. Sequence studies of proteins fromlarval and pupal cuticle of the Yellow mealworm. Insect Biochem.Molec.Biol,1997;27(2):121-131.
22. FockeM,HemmerW,WöhrlS,GötzM,etal.Specificsensitizationtothecommon housefly (Musca domestica) not related to insect panallergy.Allergy.2003May;58(5):448-51.
23. Lopata AL, Horner WE, Menon PK, Oliver J, et al. Cockroach allergenicactivity:analysisofcommercialcockroachanddustextracts.JAllergyClinImmunol1991;88:895-901.
Reference Doses for residues of allergenic foods: report of the VITAL
ExpertPanel.FoodChemToxicol,2014:63,pp.9–17
Chapter8
146
Nederlandsesamenvatting
147
Nederlandsesamenvatting:
Nederlandsesamenvatting
148
Nederlandsesamenvatting
149
NederlandsesamenvattingNieuwe duurzame eiwitbronnen zijn nodig om de groeiendewereldbevolking te voeden en te zorgen voor zominmogelijke belastingvan hetmilieu.Wanneer nieuwe voedingsmiddelen op demarktwordengebrachtvoorconsumptiedoordemens,moetenzevoldoenaandiverseveiligheidseisen.Eenvandeveiligheidseisenisdatmogelijkeallergeniciteitonderzocht moet worden. Dit staat in de richtlijnen voor nieuwevoedingsmiddelen, geregeld onder de zgn. ‘Novel food law’.Voedingsmiddelen vallen onder deze regelingwanneer ze voormei 1997nognietinsubstantiëlemateinEuropawerdengenuttigd.Richtlijnenoverhoedebeoordelingvanallergeniciteituitgevoerdmoetwordenontbrakenechter. Daarom is als onderdeel van dit proefschrift een gestructureerdeaanpak opgezet die vervolgens is toegepast op een potentiële nieuweeiwitbron, nl. meelworm (Tenebrio molitor). Meelworm behoortfylogenetisch tot de klasse insecten en zijn een voedzame en, invergelijkingmetvlees,meerduurzameeiwitbron.
De eerste stap van de voorgestelde, gestructureerde aanpak voorallergeniciteit beoordeling bestaat uit het verzamelen van algemeneinformatie over de (potentiele) eiwitbron: de eventuele geschiedenis vanblootstelling van demens aan het product (in Nederland of daarbuiten).Daarnaastwordtmogelijke fylogenetischeverwantschap (bv. zelfde fylumof klasse) met bekende allergeen bronnen in kaart gebracht en vindtidentificatievandeeiwittenendevormvantoekomstiggebruik (zoalsdemanier van verwerken of bereiding en de hoeveelheid product die zalwordengenuttigd)plaats.Uithetonderzoekbleekdatmeelwormeiwittenbevat die veel overeenkomst vertonen met allergenen in o.a. garnaal,andere schaaldieren en huisstofmijt (HSM). Ook werd duidelijk datmeelworminmeerderevormengeconsumeerdzoukunnengaanworden:gefrituurd,gebakken,maarookgeblancheerd.
Voor de vervolgstappen waren deze bevindingen van belang, omdatdaaruit kon worden afgeleid wat de evt. risico populaties (garnaal/HSMallergische mensen) zijn. Dit hielp om de belangrijkste studie populatiesvoor het onderzoek naar allergeniciteit vast te stellen. Daarnaast konworden afgeleid aan welke vorm de mensen blootgesteld zullen gaanworden,watbelangrijkisomdetestextractentedefiniëren(mogelijkzijndeverschillendeeiwittennietiniederebufferoplosbaar).
Nederlandsesamenvatting
150
Inditproefschriftisheteffectvanverschillendevormenvanbereidingvanmeelworm op allergeniciteit onderzocht. Het onderzoek liet zien dat demate van oplosbaarheid van verschillende belangrijke eiwitten (zoalstropomyosineenargininekinase),doordebereidingswijzewerdbeïnvloed,maardeallergeniciteitniet.Uithetonderzoekkwamduidelijknaarvorendat verschillende buffers nodig waren om zoveel mogelijk eiwitten endaarmeepotentieleallergeneninoplossingtebrengenomdezevervolgenste kunnen bestuderen. Het gebruik van alleen een waterige buffer kanleidentotverkeerdeconclusies,omdatnaprocessingeiwittenonoplosbaarkunnenworden,maarnogsteedsallergeenzoudenkunnenzijn.
De volgende stap in de gestructureerde aanpak is het bepalen vanmogelijke kruisreactiviteit. Omdat meelworm en garnaal fylogenetischnauw verwant zijn en analogen van bekende allergenen in garnaal(tropomyosineenargininekinase)ook inmeelwormwerdenaangetoond,lag het voor de hand om de mogelijke allergeniciteit van meelworm teonderzoekenbijgarnaalallergischepatiënten.
Ookhierbijwerdeenstapsgewijzeaanpakgevolgd.Eerstwerdnagegaanofgarnaalallergische patiënten gesensibiliseerd waren voor meelworm, datwil zeggen of er IgE antistoffen die meelwormeiwitten herkenden,aanwezig waren in hun bloed. Dit werd gedaan met 4 verschillendetestmethoden (BAT, blot, ISAC, ImmunoCAP). Bij de meestegarnaalallergischepatiënten(88%van60patiënten)werdenIgEantistoffentegen meelworm aangetoond. Om te onderzoeken of deze sensibilisatieklinisch relevant was, werd als volgende stap een dubbelblindevoedselprovocatie met meelworm uitgevoerd. Bij dezevoedselprovocatietest vertoonde de meerderheid (13 van 15) van depatiëntenmet IgE tegenmeelworm, allergische klachtenbij het eten vanmeelworm.Dezeklachtenvarieerdenvanoraleallergieklachtentotgastro-intestinaleklachtenenbenauwdheidenwarenteclassificerenalsmildtoternstig.
Behalvegarnaalallergischepatiënten,lopenookHSMallergischepatiënteneen mogelijk risico op meelworm allergie, vanwege de eerdergenoemdefylogenetische verwantschap. Een deel van de HSM allergische patiëntenreageertopallergenenvanHSM(tropomyosineenargininekinase)dieookdoor garnaalallergischepatiëntenwordenherkend.Deze groeppatiëntenwas ook geïncludeerd in de garnaalallergische groep. Het overgrote deelvandeHSMallergischegroep(~90%)herkentechtergeentropomyosineof
Nederlandsesamenvatting
151
argininekinase,maarandereallergenenvanHSM.IndezeHSMallergischepatiëntengroep werd bij 22% eveneens meelwormsensibilisatieaangetoond. Tevens zijn ook patiënten met seizoensgebondenneusklachten, maar zonder sensibilisatie of allergie voor huisstofmijtgetest. Ook daarbij werd in 16 % van de patiënten sensibilisatie voormeelworm gevonden. In een controlegroep van mensen zonder aanlegvoor allergische ziekten werd geen sensibilisatie voor meelwormaangetoond.
Dezebevindingengevenaandatdegroepdie risico looptopeenallergievoor meelworm mogelijk breder is dan alleen de garnaal (schaaldier)-allergischepopulatie.Echter,deklinischeimplicatiesvandezebevindingenmoetennogverderonderzochtworden.
Behalve een risico op sensibilisatie/allergie voormeelworm bij patiëntenmetsensibilisatieofallergievoorvergelijkbareallergenen(co-sensibilisatieofkruisrectie),moetookrekeningwordengehoudenmetdemogelijkheiddatnieuweeiwitteneennieuwe(totnutoenietbekende)allergiekunnenveroorzaken. Dit wordt ook wel een ‘primaire allergie’ genoemd. Ofmeelworm een nieuwe (voedsel)allergie zou kunnen veroorzaken, werdonderzocht in een volgende stap. Mensen die t.g.v. hun hobby ofberoepsmatig in aanraking komen met meelworm kunnen mogelijkgesensibiliseerdrakenendusallergischworden.Inditonderzoekisdaaromnagegaan of professionele en hobbymatige meelwormkwekers mogelijkallergische klachten hebben ontwikkeld. Vier personen die klachtenhadden bij contact met en/of het eten van meelworm, waarbij ook IgEantistoffentegenmeelworminhetbloedwarenaangetoond,zijnaaneennader onderzoek onderworpen. Bij twee personen werd eenvoedselallergie voor meelworm aangetoond. De andere twee waren opbasis van anamneseen serologie zeer verdacht voor een inhalatieallergievoor meelworm. Geen van deze vier personen had een allergie voorgarnaal en slechts één van de vier had een HSM allergie, zodatkruisreactiviteit als verklaring hiervoor onwaarschijnlijk was. Dit werdbevestigd door het ontbreken van sensibilisatie tegen de bekendeallergenen tropomyosine of arginine kinase in drie van de vier personen(waaronder ook de HSM allergische persoon). Een andermeelwormallergeen, nl. het Larval Cuticle Protein (LCP), lijkt deels ofvolledigverantwoordelijkvoordezeprimairemeelwormallergie.
Nederlandsesamenvatting
152
Omdat de personen met een primaire meelwormallergie op meerderemanieren(via inhalatie,huidenvoeding)blootgesteldzijnaanmeelwormkan er geen conclusie getrokken worden over de risico’s op primaireallergie wanneer insecten als voedingsbron gebruikt gaan worden in dealgemenebevolking.Hetrisicohieropisinelkgevalnietuitgesloten.
Omdat naast meelworm, ook andere insecten als mogelijke nieuweeiwitbron gebruikt kunnenworden, is in dit project ook de allergeniciteitvan enkele andere insecten bestudeerd. Hiervoor is serum van dezelfdepatiëntengroepen (garnaal- en primair meelwormallergische patiënten)gebruikt.
De garnaalallergische patiënten bleken naast Tenebrio molitor;(meelworm), voor het overgrote deel gesensibiliseerd voor Achetadomesticus; (huiskrekel), Zophobas morio; (morio- of super-worm),Alphitobiusdiaperinus; (buffalo-of lesser-meelworm),Locustamigratoria;(Afrikaanse treksprinkhaan), Galleria mellonella; (grote wasmot) enHermetia illucens; (zwarte soldatenvlieg). Hoewel de klinische relevantievan deze resultaten (nog) niet onderzocht is d.m.v. dubbelblindeprovocatie, geven deze resultaten aan dat er een risico is dat ook dezeinsecten allergene risico’s met zich meebrengen voor garnaal- enwaarschijnlijkalleschaaldier-allergischepatiënten.
De mensen met een primaire meelwormallergie lieten wisselendepatronenvansensibilisatievoordeverschillendeandereinsectenzien.Ditwijsteropdathetrisicoopallergievoorinsectenbijdezepersonenandersisdanbijdegarnaal-ofschaaldierallergischepersonen.Andereallergenendantropomyosineenargininekinaselijkenhierbijbetrokken.
Samengevat blijkt de gestructureerde stapsgewijze aanpak een goedemethodeomderisico’svannieuweeiwitbronneninkaarttebrengen.Hetop demarkt brengen van insecten brengt een duidelijk risico op allergiemetzichmeevoorgarnaalallergischepatiëntenenwaarschijnlijkvooralleschaaldier allergische patiënten. Een risico voor patiënten met andereallergieënkannietwordenuitgesloten. Insectenkunnenookeenprimaireallergieveroorzaken.Onderzoek,zoalsinditproefschriftbeschrevenisvanbelang om de veiligheid van nieuwe voedingsmiddelen te kunnenwaarborgen.Ook zal dit besluitvorming ten aanzien van de toelating vannieuwe voedingsmiddelen en/of de eventueel te nemenrisicomanagementmaatregelenondersteunen.
Contributingauthors:JolandaH.M.vanBilsenTheNetherlandsOrganizationforAppliedScientificResearch(TNO),Zeist,theNetherlandsCarlaA.F.MBruijnzeel-KoomenDepartmentofDermatology/Allergology,UniversityMedicalCenterUtrecht,theNetherlandsCaterinaGabrieleDepartment of Experimental and Clinical Medicine, Magna GraeciaUniversityofCatanzaro,ItalyMarcoGaspariDepartment of Experimental and Clinical Medicine, Magna GraeciaUniversityofCatanzaro,ItalyConstanceF.denHartogJagerDepartmentofDermatology/Allergology,UniversityMedicalCenterUtrecht,theNetherlandsGeertF.HoubenTheNetherlandsOrganizationforAppliedScientificResearch(TNO),Zeist,theNetherlandsGovardusdeJongTheNetherlandsOrganizationforAppliedScientificResearch(TNO),Zeist,theNetherlandsFrancescaMonteleoneDepartment of Experimental and Clinical Medicine, Magna GraeciaUniversityofCatanzaro,ItalyAndréC.KnulstDepartmentofDermatology/Allergology,UniversityMedicalCenterUtrecht,theNetherlands
Acknowledgements(Dankwoord)Zonderdehulpenhetenthousiasmevanallepatiëntenwasditproefschrifternietgeweest.Zijhebbenhetmogelijkgemaaktdoorhunenormebijdragedatwemooieresultatenhebbenwetenvastteleggen.Daarbij kunnen ook niet ontbreken alle werknemers van Kreca en danspeciaalookdehelefamilieCalisdieenormgeholpenhebbenbijhetlerenkennenvandeinsectenindustrie.VanVenik:MarianPeters,dooronsmetdeinsectenindustriekennistelatenmaken.Envoorhetmakenvanallemeelwormsnacks:Ruig.HierbijhebbenEelcoBotterenJanRuigonserggeholpenmeteenlekkeresnackomdubbelblindmeetekunnentesten.BinnenhetUMC:Ongelooflijkleukdatnaarheteindetoe,iktweepromotorenhebgekregen:AndréenCarla.IkbenbijzondervereerddatikjulliebeidealsvoorbeeldhebgehadenhetvoorrechthebdezetweegrotenamenindeDermatologieenAllergologieopmijnproefschrifttemogenhebbenstaan.Stans, Judith, Lieneke, Florine,Mary-Ann, Floor,Wouter,Rob, Jorienmijnbenchmaatjeenalmijnfijnekamergenootjes,metjulliezouikzoopnieuwzo’ntrajectstarten,hetisdusmaargoeddatwemeteengrootdeelverdermogenenmakkelijkevensamenkunnenlunchenmetStans!Ookvoeldeikmealtijdgesteundophetlab,speciaaldoorEdward,Simone,Jan, Helma en Laura. Laura enHelma zijn ook doorgegroeid naar anderelocaties,maarzullenvoormij tochophet labhoren.Simone,zetmop jelaatstejaartje!Ischa,bedanktvooraljegoedezorgen!VerderwashetzonderAns,Astrid,Jos, Ellen, Marieke, Mignon, Mark, Francine, Els, Laury, Henny en groepOtten,Rowena,Sharrida,Astrid,Nienke,Barbara,Thuy-My,Margaret,AnnaenHannah nooit zo’n geslaagde periode geweest, door al jullie hulp aanaanmoedigingenzijndejarenvoorbijgevlogen!
162
Opdepoli:Elly,Linda,Doreen,Hayat,julliehulpenenthousiasmeheeftmealtijdvrolijkgemaakt.Gelukkigzieikjullienoglekkerveel!Driesuperdiëtistes:Dieuwke,AnouskaenAnne,altijdfijnommetjullietemogensamenwerkenenaltijdmeteengoedgevoelallepatiëntendoortekunnensturenof tekunnenbespreken. Jullie zijnbijzonderenwaardevolvooronzeafdeling,fijndatikdatvanzodichtbijhebmogenzien!SamenwerkingmethetAMI-labwaszeerfijndoorCoby,WillyenSaskia.Ikmochtaltijdweereennieuweportiemeelwormserumaanleveren,hoefijnjulliealtijdreageerdenenhoesnelallesaltijdkonwordengedaan,ideaal!NatuurlijkzijnKittyenGeertookUMC,maarerzittochmeerTNOindebasis.Datheeftvoorhelefijndiscussiesgezorgdenhetheleboekaangescherpttot wat het is geworden. Hopelijk zullen er nog veel van deze mooieprojectenwordengedaan.HeelveeldankvoornogmeerZeistgevestigdeTNO-ers. Zonder Aard, Astrid, Ben, Jolanda,Marty, Emmeke, Sabina wasalles minder mooi geworden. Zij hebben allen erg geholpen en mij veelgeleerd.Van de Nvwa: fijne samenwerking met Jaqueline Castenmiller en HubNoteborn,hartelijkbedankthiervoor.SpecialthankstoJonasLidholm,MånsÖsterbergWiderstrand,JuciaJimenoNogales,MarcoGaspari,CaterinaGabriele,FrancescaMonteleonefortheirmuch-appreciatedhelpwithextracts,immunoCAPandLC-MS/MS.FromthecollaborationinImpARAS,it’sbeenanhonourtohavegottentoknowandworkwithAnnabelleCapt,AnneConstable,RenéW.R.Crevel,IvanDimitrov, Christiane K. Faeste, Rocío Fernandez-Canton, Stavroula Giavi,KevinC.Glenn,CharlotteB.Madsen.And the specialwork togetherwithKatrineL.Bøgh,JuliaUptonandThomasEiwegger.Mijnschoonfamilievooraldesteundieikvoeldeendeheerlijkedinertjesommeevenafteleiden.Mijnouders,MieenIWen,VesperenFletcherennatuurlijkspeciaalPipvoorde mooie pen die ik begin dit jaar kreeg om dit boek mee te kunnen
163
schrijven, bedankt voor het er zijn en af en toe klankbord voor m’ngedachtenwillenzijn.Maart,omdatjebent.
CurriculunvitaeHenrikeBroekmanwasbornonthe28thofNovember1980inUtrecht.Shegraduated secondary school in 1999 fromUtrechts Stedelijk Gymnasium.Duringmedicalschool,shedevelopedaninterestinresearchwhenshedidaresearchprojectinthechildren’shospitalinBoston.Theinterestforallergybecame apparent when studying peanut allergy in her final year at thedepartmentofDermatologyandAllergologyattheUMCUtrecht.Aftergraduation,shestartedaPhDprogramunderProf.Bruijnzeel-KoomenandProfKnulst,leadingtothisthesis.Atthestartof2017shestartedtheDermalologyrecidencyprogramattheUMCUtrecht.LivesinUtrechtwithherhusbandMaartenHustinx
