Application & Background

Neuropeptides are known to play important roles in almost every aspect of normal physiology, and we are increasingly understanding their varied roles in different disease pathologies (1). An important consideration in studying neuropeptides is their ability to diffuse far from their secretory vesicle and influence cellular chemistry far from the original neuron in which they were produced (1). Mass spectrometry imaging (MSI) is thus an excellent analytical technique with which to study neuropeptide distribution and quantification in both normal and diseased states, as it preserves the spatial location of each individual biomolecular signal, without need for prior knowledge of the analytes of interest. However, while many protocols exist for protein matrix-assisted laser desorption/ionization- (MALDI-) MSI, there are relatively few studies demonstrating the use of this analytical tool for imaging of neuropeptides.

In this study, a novel sample preparation protocol was developed for imaging neuropeptides in crustacean brain tissue via MALDI-MSI. Using an original washing method and optimized matrix spraying protocol developed using the HTX TM-Sprayer, changes in neuropeptide distribution in the crustacean brain were able to be analyzed in response to three different stress conditions compared to a control.

Experimental

experimental design

Female blue crabs (Callinectes sapidus) were obtained from Midway Asian Market (Madison, WI) or LA Crawfish Company (Natchitoches, LA). All crabs were originally housed in sea water at 35 parts per thousand salinity, 17-18⁰C, 80-100% oxygen (O2) saturation, and a pH = 8.3 for over 5 days. Then, the crabs were divided into four experimental groups: (a) control (pH = 8.3, 100% O2 water saturation), (b) severe hypoxia (pH = 8.3, 10% O2 water saturation), (c) mild hypoxia (pH = 8.3, 50% O2 water saturation), and (d) hypercapnia (pH = 7.6−7.8, 50% O2 water saturation). Hypoxic experimental conditions were created by sparging tanks with nitrogen gas until the O2 was at the desired level (approximately 30-40 minutes), and the hypercapnic experimental conditions were created by sparging a tank with CO2 gas until the pH was lowered to the desired level (approximately 5 minutes). For each condition, after the experimental tank was prepared, one crab was placed in the tank for 2 hours. After anesthetization on ice, brains were removed from the crabs and embedded in gelatin. Tissues were sectioned at 12 μm using a Micro HM525 cryostat (Thermo Scientific) and mounted onto plain, glass microscope slides. Samples were stored at -80°C until needed.

Wash protocol optimization

Control brain samples were removed from the freezer and placed in a vacuum chamber for 10-20 minutes until the sample was dry. Samples were exposed to one of 14 different wash conditions using a Tissue-Tek slide staining system. These conditions were the 7 following EtOH:H2O volume:volume (v:v) solutions for either 10 or 30 seconds: 100:0, 85:15, 70:30, 50:50, 30:70, 15:85, and 0:100. Three technical replicates were used to evaluate each washing condition. The osmolarity of the wash solution was analyzed pre- and post-wash using an Advanced Instruments Model 325 Single-Sample Osmometer in order to evaluate whether the wash was removing salts from the tissue. The post-wash solution was also then spotted in a 1:1 ratio with 150 mg/mL DHB in 50:50 MeOH:H2O and 0.1% formic acid (FA) in order to evaluate whether the wash was removing any additional analytes, including some neuropeptides, from the tissue.

sample preparation

After washing, samples were placed under vacuum for 15-20 minutes to dry. DHB matrix was applied using the HTX TM-Sprayer with the parameters displayed in Table 1.

data collection and analysis

All experiments were performed on a MALDI-linear trap quadrupole (LTQ)-Orbitrap XL mass spectrometer (Thermo Fisher). Data were collected from a mass range of m/z 500-2000 at 75 μm spatial resolution and 30,000 mass resolution at m/z 400. Raw data files were exported in imZML format and imported into MSiReader (Version 1.00). Files were normalized in MSiReader to the total ion current (TIC), and images were generated. Accurate mass matching of ± 5 (condition comparisons) or 10 (method development) ppm to a home-build database was used to annotate neuropeptides, and accurate mass matching of ± 5 ppm to the LIPID MAPS online database was used to identify lipids within the images. All images were normalized to the same intensity scale. Using the Coloc2 FIJI plugin, images were overlaid in ImageJ to perform co-localization analysis based on a linearleast-squares regression for the technical replicates. A minimum of five points were used as coordinates for each registration. A Pearson’s correlation coefficient was calculated for each pixel between the intensities of different images to conduct intensity analyses. The Manders’ split coefficient was then calculated for each analyte and used to evaluate how well the distribution of each analyte was conserved between the images of technical replicates. The raw data files were also imported into SCiLS software (Bruker Daltonics), and a student’s t test was used to determine statistically significant changes between the four experimental conditions. A Bonferroni correction was applied in order to account for multiple tests.

results

Wash Protocol Optimization

The different wash protocols were first evaluated based on the number of neuropeptide identifications. It was found that 50:50, 30:70, 15:85 EtOH:H2O yielded the highest number of identifications (~30). Then, average normalized intensity was compared between the three wash solutions, and, while it was found that all the washes increased signal intensity compared to a control of no wash, the 50:50 EtOH:H2O solution gave the highest overall signal intensity (approximately a 5-fold increase) at both the 10 and 30 second duration (Figure 1).

In order to identify the optimized time of washing, a colocalization analysis for five different known neuropeptides was performed. However, it was found that there was not significant difference between the 10- and 30-second washes, so the 10-second wash was used going forward in all experiments simply due to its brevity.

For the optimized wash system of 50:50 EtOH:H2O for 10 seconds, the intensity of 34 neuropeptides were found to be enriched. In comparison, three neuropeptides showed no change in signal intensity, and six neuropeptides actually showed a decrease in signal intensity compared to the control (Figure 2).

The effect of the optimized wash procedure on other analytes was also investigated. Overall, the signal intensity of lipids appeared to increase in the washed tissues compared to the control, which is likely due to the removal of some salts by the washing step. However, there was no significant change in the osmolarity of the washing fluid measured. The washing fluid was also analyzed for neuropeptides. A few were identified in the post-washing fluid, which was expected given the decrease in signal intensity noted for a few neuropeptides in the washed sample compared to the control.