Beyond the transect: An alternative microchemical imaging method for fine scale analysis of trace elements in fish otoliths during early life

Highlights

• Ba and Sr levels in otoliths from P. coelestis were imaged using LA-ICP-MS.

• Longerich's calibration method was applied to quantitative otolith imaging.

• An alternative to single line transects for microchemical analysis is presented.

Abstract

Microchemical analysis of otolith (calcified ‘ear stones’ used for balance and orientation) of fishes is an important tool for studying their environmental history and management. However, the spatial resolution achieved is often too coarse to examine short-term events occurring in early life. Current methods rely on single points or transects across the otolith surface, which may provide a limited view of elemental distributions, a matter that has not previously been investigated. Imaging by laser ablation-inductively coupled plasma-mass spectrometry (LA-ICP-MS) permits microchemical analyses of short-term events in early life with high (< 10 μm) resolution, two-dimensional (2D) visualization of elemental distributions. To demonstrate the potential of this method, we mapped the concentrations of Sr and Ba, two key trace elements, in a small number of juvenile otoliths of neon damselfish (Pomacentrus coelestis) using an 8 μm beam diameter (laser fluence of 13.8 ± 3.5 J cm^− 2). Quantification was performed using the established method by Longerich et al. (1996), which is applied to 2D imaging of a biological matrix here for the first time. Accuracy of > 97% was achieved using a multi-point non matrix-matched calibration of National Institute of Standards and Technology (NIST) 610 and 612 (trace elements in glass) using Longerich's calculation method against the matrix-matched standard FEBS-1 (powdered red snapper [Lutjanus campechanus] otolith). The spatial resolution achieved in the otolith corresponded to a time period of 2 ± 1 days during the larval phase, and 4 ± 1 days during the post-settlement juvenile phase. This method has the potential to improve interpretations of early life-history events at scales corresponding to specific events. While the images showed gradients in Sr and Ba across the larval settlement zone more clearly than single transects, the method proved sample homogeneity throughout the structure; demonstrating that 2D scanning has no significant advantage over line scans.

Introduction

Fish otoliths provide an unparalleled source of demographic and ecological information in the animal kingdom. Otoliths are small (usually < 30 mm diameter) calcified structures used for hearing and balance in the heads of all teleost species (Popper and Fay, 2011; Fig. 1). They grow continuously throughout an individual's lifetime, depositing calcium carbonate (CaCO₃) and proteinaceous tissue (otolin) in alternating bands (increments) that are visible when viewed under magnification (Campana and Neilson, 1985). Increments are deposited chonologically, allowing estimation of age with great precision (Campana, 2001). The width of increments can also be measured to estimate growth rates, as otolith and body growth are proportional (Casselman, 1990). Otolith-based assessments of annual age and growth are used to monitor the status of exploited fish stocks across the globe (Campana and Thorrold, 2001, Cushing, 1975, Maciena et al., 2007), while examination of daily increments in the juvenile region of the otolith is used to determine the growth performance and timing of critical early life stages (e.g. the larval phase) (Brothers et al., 1976, Suthers, 1998).

In addition to information on age and growth, otoliths can provide valuable insights into the connectivity of fish populations. Most reef fishes have a bipartite life cycle, with a highly mobile larval phase followed by a relatively sedentary adult phase (Leis, 1991, Roughgarden et al., 1988). The larval phase lasts days to months depending on the species, during which time ocean currents can transport larvae thousands of kilometers from the location where they were spawned (Cowen et al., 2006). The sources of larvae for populations are therefore rarely known, rendering conservation of important breeding areas a challenging task. Trace elements incorporated into the otolith from ambient water can identify the source locations of individuals, because elemental compositions (microchemical ‘signatures’) are unique to particular locations (Campana et al., 2000). Microchemical analysis can be used to compare signatures from the larval core of the otolith (source signature) to a range of potential source locations until a match is found. If a large enough sample of the population is examined, the extent of larval exchange between adjacent populations can be estimated. Analysis of the otolith region corresponding to settlement onto adult habitat (settlement zone) can also be used to determine the duration of the larval phase (pelagic larval duration, PLD), because water chemistry can differ greatly between pelagic and benthic habitats (Hamilton and Warner, 2009). PLD is an essential parameter in physical models used to predict larval dispersal distances and trajectories (Wilson and McCormick, 1999), and cannot always be obtained by counting daily increments from the time of hatching because the settlement mark is not always visible in otolith microstructure (Hamilton and Warner, 2009, Hoover and Jones, 2013). Microchemical analysis of settlement zones in the otolith may provide PLD estimates for many species for which such information is currently unavailable.

Despite the utility of otolith microchemistry for connectivity investigations, the spatial resolution of commonly used spectrometry techniques is often too coarse to resolve elemental concentrations over time-scales relevant to larval dispersal and settlement events (i.e. days). Spatial profiling of trace elements (e.g. Sr and Ba) in otoliths is usually conducted using laser ablation inductively-coupled plasma mass spectrometry (LA-ICP-MS), where a focused laser beam is directed onto the surface of a sectioned otolith to quantify elemental concentrations in specific growth regions. This technique is typically used to gather data on spots or transects across the otolith surface using beam diameters ranging from 20 to 50 μm (Ashford et al., 2006, Chang et al., 2012, DiMaria et al., 2010, Elsdon et al., 2008, Ferguson et al., 2011, Hamilton and Warner, 2009, Longmore et al., 2010). However, daily growth increments can be < 1 μm wide (generally 1–5 μm) depending on the species and developmental stage (DiMaria et al., 2010, Hamer and Jenkins, 2007). Narrow increments present an obvious limitation: data obtained using laser beam diameters greater than these increment widths means that daily resolution is often unachievable. Given that dispersal and settlement in reef fishes can occur over a time period of days (Hoover and Jones, 2013), such techniques limit the resolution of short-term events in early life, including identification of the settlement zone. Additionally, single spots or transects only provide a limited view of elemental changes throughout early life, because concentrations are known to vary spatially within the otolith (Durrant and Ward, 2005), even within increments of the same chronological period. A method that quantifies elemental concentrations in multiple directions is required to visualize gradients in elemental composition across key developmental boundaries (e.g. larval settlement).

Microchemical imaging by LA-ICP-MS, as opposed to single spot or transect analysis, has the capacity to resolve a greater number of microchemical features, whilst still retaining the ability to observe radial trends in otolith microchemistry. The crater diameter and depth produced by the solid-state Nd:YAG laser beam are dependent on beam energy, pulse frequency, duration and laser power output; and the volatility of the analyte, which determines the volume of material ablated and ultimately detected by the ICP-MS (Lear et al., 2012). Using a smaller laser beam diameter results in somewhat shallower depth penetration, and has the potential to provide more precise information about short-term life-history events. A smaller beam diameter results in higher detection and quantification limits, and decreases the precision of determined analyte concentrations in the specimen. Therefore, a compromise between laser beam diameter and power to achieve optimal imaging parameters is essential.

A further limitation to LA-ICP-MS analysis of materials such as otoliths is the lack of characterized matrix-matched reference materials (Campana, 1999, Hare et al., 2012). The ablation and ionization of particles from the sample surface are characteristic of that sample's chemical and physical composition, and the substitution of a non matrix-matched standard has the potential to generate inaccurate results. A small number of studies have achieved success using partially matrix-matched materials. NIST (National Institute of Standards and Technology) 1486 (bone meal) has been used as a hydroxyapatite matrix for quantifying trace metals in teeth (Arora et al., 2011, Austin et al., 2013, Hare et al., 2011). Two sagittal otolith standards (INMS FEBS-1 and NIES-022) are produced for bulk elemental analysis by solution nebulization ICP-MS (Sturgeon et al., 2005), but have not been characterized as reference materials for solid sampling.

This paper describes a new method for producing fine-scale quantitative images (maps) of trace element distribution in fish otoliths during early life using a matrix matched standard and NIST glasses 610 and 612. We demonstrate how the technique can be used to observe gradients in Sr and Ba concentrations during the early life history of a small reef fish (Pomacentrus coelestis) by correlating high resolution scanning electron microscopy (SEM) with images produced by LA-ICP-MS. We compare the results obtained using microchemical imaging to those achieved using a single ablation transect, which is a common method for microchemical analysis of fish otoliths during early life. The calibration method of Longerich et al. (1996) has been applied to microchemical imaging by LA-ICP-MS of a biological matrix here for the first time to test the suitability of readily available non matrix-matched standards NIST 610 and 612 for external calibration of the LA-ICP-MS system in the analysis of fish otoliths. Paul et al. (2014) recently used a similar approach for imaging multiphase assemblies in geological samples. Limburg et al. (2011) described a synchrotron microprobe-based X-ray florescence microscopy imaging approach, though this technique does not suffer the same quantitative pitfalls of LA-ICP-MS. The approach described here is a potential alternative to single line transects.