Article start

Bird feathers as a biomonitor for environmental pollutants: Prospectsand pitfallsVeerle L.B. Jaspersa,*, Adrian Covacib, Dorte Herzkec, Igor Eulaersd, Marcel EenseaDepartment of Biology, Environmental Toxicology Group, Norwegian University of Science and Technology (NTNU), Høgskoleringen 5, 7491, Trondheim,NorwaybToxicological Center, University of Antwerp, Universiteitsplein 1, 2610, Wilrijk, BelgiumcNorwegian Institute for Air Research (NILU), FRAM Centre, 9007, Tromsø, NorwaydDepartment of Bioscience, Aarhus University, Frederiksborgvej 399, 4000, Roskilde, DenmarkeDepartment of Biology, Behavioural Ecology and Ecophysiology Group, University of Antwerp, Universiteitsplein 1, 2610, Wilrijk, Belgiumarticle infoArticle history:Available online 27 May 2019Keywords:FeathersBiomonitoringBirdsMetalsPOPsPBDEsPFASsOPEsMicroplasticsExternal contaminationabstractDue to increasing amounts of hazardous chemicals released into the environment, there is a high de-mand for developing easy and non-destructive biomonitoring tools. In a recent paper published in Trendsin Analytical Chemistry, Rutkowska et al. (2018) concluded that feathers are a good matrix for bio-monitoring of environmental pollutants in birds. In this commentary, we discuss the general conclusionof this paper. We provide several examples for which this statement is not correct, and we emphasizethat only for legacy persistent organic pollutants (POPs) and mercury there is currently enough scientificevidence to use feathers as a reliable biomonitor, given that appropriate sampling designs and QA/QCprotocols are taken into account. Furthermore, we discuss different pretreatment (e.g. feather washing)and analytical protocols along with specific QA/QC to be considered. In summary, this commentaryprovides an overview of the prospects and pitfalls when using feathers as a biomonitor for environ-mental pollutants.©2019 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-NDlicense (http://creativecommons.org/licenses/by-nc-nd/4.0/).1. IntroductionIn a paper recently published in Trends in Analytical Chemistry,Rutkowska et al.[1]have presented an extensive literature reviewon the use of bird feathers to assess contamination levels of variousenvironmental pollutants. The discussed environmental pollutantsrange from trace metals and legislated persistent organic pollutantsto emerging organic pollutants (such as per- and polyfluoralkylatedsubstances (PFASs), organophosphate esters (OPEs) and phenols)and microplastics.Although we acknowledge the value of such comprehensivereview, we have reservations about the main conclusion presentedby the authors thatFeathers can be successfully used as environ-mental pollution monitors. We will provide several examples forwhich this statement is not correct (as will be detailed furtherbelow), and underline the lack of knowledge of emerging con-taminants in particular. In addition, we wish to make importantclarifications regarding preparation, analysis, and restrictions onthe use of feathers for environmental pollutant analysis. Our mainmessage is that scientists aiming to employ feathers in futureresearch should be well aware of the prospects and pitfalls whenusing feathers as biomonitors for environmental pollution.2. Discussion2.1. Suitability of feathers for different types of pollutantsIn the conclusion section, Rutkowska et al.[1]state thatFeathers can be successfully used as environmental pollutionmonitors because the proportion of contaminants in the body tothe level present in the feathers is relatively constant for eachxenobiotic and there is a high correlation between the levels ofcontaminants in the food of seabirds and those in feathers. Thismay indeed be the case for legislated persistent organic pollutants(POPs), in particular polychlorinated biphenyls andp,p'-dichlor-odiphenyldichloroethylene (p,p-DDE), as indicated by severalpublished articles (reviewed by García-Fernandez et al.[2]). How-ever, this statement is not true for many metals that show external*Corresponding author.E-mail address:veerle.jaspers@ntnu.no(V.L.B. Jaspers).Contents lists available atScienceDirectTrends in Analytical Chemistryjournal homepage:www.elsevier.com/locate/trachttps://doi.org/10.1016/j.trac.2019.05.0190165-9936/©2019 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).Trends in Analytical Chemistry 118 (2019) 223e226
contamination on the feather surface[3e6], and has recently beenquestioned for emerging compounds, such as PFASs and novelflame retardants[7]. Indeed, for many pollutants the correlationsbetween the concentrations in feathers and internal tissues are notknown (Table 1), and there may be differences among feathertypes. There are three different routes of how pollutants can betransferred to feathers; i) during feather growth through theconnection to the blood circulation of the bird, ii) during preeningvia preen oil and iii) via contact with the environment (air, dust andwater). Different chemical groups will be transferred via differentroutes according to their physical-chemical properties, with POP-like lipophilic chemicals mostly being found in the preen oil andpolar chemicals such as PFASs and some OPEs being transferredduring the growth of the feathers and the contact with the envi-ronment. This means that for some chemicals, concentrationsmeasured in feathers may not be related to internal tissue con-centrations but rather originate from external sources.There is, thus, a great need for further studies to examine as-sociations between concentrations in different feather types andinternal tissues for different classes of contaminants, and toinvestigate the importance and extent of external contamination. Inaddition, concentrations in feathers and internal tissues may notalways be correlated when there is a time difference between theformation of the feathers and the sampling of the feathers.Depending on the research questions, it is generally preferred tosample nestling feathers which are recently grown or still con-nected to the blood circulation rather than feathers from adult birds[3,8]. For example, adult migrating bird species may grow theirfeathers on the overwintering grounds, which subsequently maynot show associations to the concentrations in the body of birds onthe breeding grounds [e.g. 9]. The collection of museum samplesmay also present some challenges, such as the preservation,degradation of some contaminants over time, different types offeathers available, contamination during storage (e.g. OPEs) orthrough the use of disinfectants (e.g. arsenic[6]) and the lack ofmetadata on the sampled individuals[8,10]. We refer toTable 1foran overview of the general suitability of feathers for different pol-lutants, and the knowledge gaps that still exist.In addition, the specific moulting sequence can be of concern[11], which is highly species dependent, with some speciesmoulting their sets offlight feathers over several years[12].Therefore, the sampling of body feathers can be an alternative, asthese feathers moult more regularly and evenly throughout theyear and have been shown to correlate well with internal bodyconcentrations of POPs[13,14]and mercury[15]. However, exten-sive variation between body feathers for lead in Northern goshawk(Accipiter gentilis) and tawny owl (Strix aluco) has been recorded[16], and therefore the use of body feathers may not be suitable forall compounds and species. Recently, down feathers have also beenstudied as an alternative biomonitor[17]. It is also possible tocollect moulted feathers from adults at the nest, but then the issueof external contamination is of higher concern than when samplingnestlings. If one is interested in biomonitoring the internal expo-sure of birds of prey for trace metals, it is important to analysefreshly moulted feathers or to have procedures to eliminateexternal contamination [e.g.[18]], which is not easy to achieve anddeserves further (experimental) study. On the other hand, featherscould potentially be used as a passive monitor for environmentalconcentrations of metals and elements[5,18]and other emergingpollutants [e.g.[19,20]], thus not reflecting the actual concentra-tions of contaminants in the bird, but rather a snapshot of thepollutant load in the environment of the bird. This could be the casefor microplastics as well, although Reynolds and Ryan[21][21]indi-cated that their sampling protocol (i.e. using feather brushings andfaecal samples from seven duck species) likely underestimated theoccurrence of micro-plastics in South-African wetlands. This hasbeen misinterpreted by Rutkowska et al.[1]as Reynolds and Ryan[21]did not specifically advocate for the use of feathers as a bio-monitor for microplastics.Lastly, the specific habits of the bird species, from which thefeathers are obtained, are of crucial importance for the studydesign. Some bird species migrate long distances, and thereforemay in general not be suitable to collect feathers from at thebreeding grounds if they were grown on the wintering grounds.This has been misinterpreted by Rutkowska et al.[1]in section 4.2.5on OPEs. The reason for which Svendsen et al.[9]recommendedcollecting feathers from nestlings and not adults, is not a generalstatement for OPEs, but rather a general consideration when usingmigratory birds, irrespective of the nature of pollution investigated.In that sense, knowledge of the bird speciesbiology is of highimportance when planning a biomonitoring study[3,22]. Forinstance, biomonitoring of the local terrestrial environment cannotbe solely done using raptor species, due to their relatively largeforaging areas. Instead passerine bird species, with smaller homeranges, are more appropriate to successfully monitor localcontamination for example around point sources[6,23]. However,it should be noted that only a limited amount of feathers can becollected from small birds without harming them. Therefore, theTable 1Current information regarding the suitability of feathers for biomonitoring of different types of pollutants in birds. If not enough information isavailable to draw any con-clusions, this is indicated with a question mark (?).Reflecting internalconcentrationsExternalcontaminationSuitability asa biomonitorRemarksSelectedreferencesMetals and elementsFor Hg andsome othermetalsImportant issuefor some metalsDependingon the metalExternal contamination is dependent on the location.Different elements may be important in sediment or air[3e6,8,16,18,25,33]PCBsStrongMinor issueYesExternal contamination mostly from preen oil (internal exposure)[2,27][2,27][2,27]DDEStrongMinor issueYesExternal contamination mostly from preen oil (internal exposure)[2,27]PBDEsStrongPotential issueYesExternal contamination mostly from preen oil (internal exposure)[2,27]PFASsWeakeStrongPotential issue forsome compounds?Good correlations found for some compounds, but not for others.External contamination should be further investigated[7,19,34]OPEsNoImportant issue? ?High detection in feathers and low detection in blood could be dueto either external contamination or high metabolism[7,20]PAHs???Only few studies to date[35]Bisphenols and Alkylphenols ???Only 1 study to date[36]Neonicotinoids???Only 1 study to date[37]Microplastics?Yes?Only 1 study to date (analysing external contamination usingfeather brushings)[21]Pharmaceuticals???Only 1 study to date (onfluoxetine)[38]V.L.B. Jaspers et al. / Trends in Analytical Chemistry 118 (2019) 223e226224
required amount of feather material, which is likely to dependlargely on the pollutant studied as well as on the pollutant con-centrations, should be checked in the literature [e.g.[6, 23]] or withthe respective analytical lab.2.2. Collection and storage of feathersAlthough the second part of the conclusion,it is easy to collectfeathers non-invasively and it is possible to store test material formany years[1]is largely correct, it is important to know what ismeant by non-invasively collecting feathers. If the authors are onlyreferring to the sampling of moulted feathers around the nest, thisis indeed very easy to perform. However, this could have severaldrawbacks regarding the suitability of the collected material. Onedoes not know the exact individual from which the moultedfeathers originated, and the moulted feather could have been layingthere for weeks or months, and is thus prone to external contam-ination or degradation. It is important to take this into accountwhen washing the feathers (see next section) and when inter-preting the results. Another non-invasive sampling method wouldbe cutting the feathers from living birds (instead of plucking). Incontrast to collectingflight feathers, the collection of body feathersfrom living birds generally allows for better standardization andminimal harm to the birds[13,14,24]. In respect to storage, also herea strict quality assurance/quality control (QA/QC) protocol needs tobe followed to prevent contamination and protect the integrity ofthe feathers. Therefore, no opportunistic selection of storedfeathers can be recommended for contaminant analysis, if theapplicable QA/QC protocol was not followed. Once collected,feathers can be stored for many years, but they need to be storedunder controlled conditions, protected from potential sources ofexternal contamination and degradation, such as UV radiation andmoisture. Dry feathers can be stored either in paper envelopes oralumina foil, further sealed in polypropylene (PP) zip-lock bags atroom temperature, or dirty and wet feathers can be stored in thefreezer. However, analyses of certain pollutants which can leachfrom PP zip-lock bags such as softeners, UV stabilizers and otheradditives as well as microplastics may not be possible for thesesamples. More information about the collection, transport, storageand pre-treatment of feathers can be found in the sampling andcontaminant monitoring protocol for raptors[24].2.3. Preparation and extraction of feathers for different types ofpollutantsSection2in the paper of Rutkowska et al.[1]describes thepreparation of the feathers. Although this section provides a verygeneral description of the analysis of feathers for traditional POPs,no specification has been made regarding different pollutants andtheir specific preparation. This section may therefore be confusingfor the reader since POPs, metals, PFASs, emergingflame retardantsand other classes of pollutants each require different amounts ofsample for analysis and undergo different preparation and extrac-tion procedures. It is therefore important that relevant literature isconsulted before preparing the feathers (e.g. Ref.[18]for metals[13], for POPs[7], for PFASs and[20]for current-useflame re-tardants). It is also unfortunate that the milestones in Figure 3 arenot reflecting the development of new extraction methods forother pollutants than metals.Regarding the preparation of feathers for metal analysis, thewashing step is very important as many metals are present in highconcentrations on the external surface of the feathers[5,18,25]. Thisis of concern when considering feathers as a potential biomonitor,and this discussion is missing in the section 4.1 of Rutkowska et al.[1]. In addition, it is very important to use non-metallic tools andmetal-free liquids when washing and handling the feathers beforeperforming trace elemental analysis. It should also be noted thatwashing procedures to eliminate external contamination have beenextensively investigated only for metals[3,26]and POPs[27]. Still,we are currently lacking standardised washing procedures formetals and POPs and different authors have used different methods[2,25,27,28]. Current recommendations can be found in the sam-pling and contaminant monitoring protocol for raptors[24].Extensive knowledge on washing procedures for other legacy andemerging contaminants in feathers is currently not available.Variability due to external contamination can be reduced byrestricting analysis to the shaft or rachis as external contaminationtends to be greater for the vane than the shaft and the shaft is moreeasily and effectively cleaned[8]. Indeed, the rachis has been usedin several studies for the monitoring of Pb exposure from ammu-nition sources in birds of prey[4,29,30]. However, García- Seoaneet al.[31]reported high intraspecific variability and analytical dif-ficulties due to the very low levels of arsenic, cadmium and lead inthe rachis, which may limit the suitability of the rachis whenconcentrations are low. Yet, a new methodological approach hasrecently been suggested to limit and correct for external contami-nation with metals and elements on the feather surface, includinganalysing only the rachis of the feather, using specific washingprocedures and the use of a reference element to correct forexternal contamination from sediment[18]. However, it should benoted that currently no reference materials are available specific forfeathers. In addition, the use offield and analytical method blanksor passive air samplers may contribute to control for potentialexternal contamination during bothfield sampling, storage andlaboratory preparation.Regarding the preparation of feathers for POPs and PFASs, thefine cutting or milling of the feathers is crucial for effectiveextraction of the contaminants. However, before performingwashing and/or cutting of feathers, it is important to record theweight and the length of the feathers[32]. As indicated in section3of Rutkowska et al.[1], the deposition rate of contaminants infeathers can potentially be calculated, but this has in practice onlyrarely been done. Although this has been proposed previously[2],its application is limited toflight feathers with clear growth barsthat correspond to daily growth.Lastly, all kind of micro-plastic analyses in feathers needs to becarried out under particle-controlled conditions and follow strictQA/QC protocols includingfield and analytical method blanks andspecific preparation of all materials and chemicals in contact withthe feathers.3. ConclusionsIn conclusion, we do not support the general statement byRutkowska et al.[1]that feathers can successfully be used asmonitors for any environmental contaminant. Feathers can beuseful as a biomonitor for POPs, mercury and several other metalsunder the conditions that appropriate sampling designs and pre-treatment of samples along with QA/QC protocols during storage,preparation and analysis are taken into account. Still, even formercury a recent study has indicated that still several knowledgegaps existed related to variation within feather parts, amongfeather types and between feathers of the same type[33]. Quiteclearly, more research is needed to investigate whether featherscan be useful to monitor the internal concentrations of other classesof contaminants. We hope to have clarified the specific consider-ations that need to be taken into account before investigatingpollutants in feathers. The most important aspects relate to thechoice of the bird species and the possibility of external contami-nation on the feathers, which in turn depends on the compoundsV.L.B. Jaspers et al. / Trends in Analytical Chemistry 118 (2019) 223e226225