Rapid Immunohistological Measurement of Tyrosine Hydrox- ylase in Rat Midbrain by Near-Infrared Instrument-Based De- tection

We present a robust, fresh-frozen approach to immunohistochemistry (IHC), without committing the tissue to IHC via fixation and cryopreservation while maintaining long-term storage, using LiCor-based infrared (IR) quantification for sensitive assessment of TH in immunoreacted midbrain sections for quantitative comparison across studies. In fresh-frozen tissue stored up to 1 year prior to IHC reaction, we found our method to be highly sensitive to rotenone treatment in 3-month-old Sprague-Dawley rats, and correlated with a significant decline in rotarod latencyto-fall measurement by approximately 2.5 fold. The measured midbrain region revealed a 31% lower TH signal when compared to control (p<0.01 by t test, n=5). Bivariate analysis of integrated TH counts versus rotarod latency-to-fall indicates a positive slope and modest but significant correlation of R2=0.68 (p<0.05, n=10). These results indicate this rapid, instrument-based quantification method by IR detection successfully quantifies TH levels in rat brain tissue, while taking only 5 days from euthanasia to data output. This approach also allows for the identification of multiple targets by IHC with the simultaneous performance of downstream molecular analysis within the same animal tissue, allowing for the use of fewer animals per study.

Our aim was to study a rodent model of parkinsonian using a sensitive and automated quantification method. The method presented here is robust because of the nature of IR detection, having a wide linear range, and is fast and impartial due to the data collection being performed by an instrument [36]. and Safety Guidelines due to rotenone toxicity. Chemicals not specifically stated were at least molecular biology grade and purchased from Sigma (St Louis, MO).
Statistical comparisons were performed using SPSS version 25 (64-bit edition) (IBM, Armonk, NY, USA). Unless otherwise noted, one-tailed unpaired Student's t tests were performed to determine statistical significance between the CT and RO groups, defined by a p-value < 0.05. 3. Perform euthanasia using 5% isoflurane overdose and decapitation by guillotine. 4. Open the skull by inserting the tips of the Adson 2/1 forceps into the foramen magnum and removing the parietal bones laterally. 5. Cut the cranial nerves on the ventral side of the brain. 6. Place the brain in 4°C child Hank's Balanced Salt Solution for washing. 7. Place the brain supine in a pre-cooled 4°C stainless steel brain coronal matrix mold with slots separated at 1 mm intervals to normalize brain slices with razor blades. 8.

Materials
Locate the cerebral peduncles in the ventral view and take a 2 mm section beginning at their anterior extent (corresponding approximately to Bregma -5.2 to -5.6 mm). 9.
CRITICAL STEP Remove the SN of the right hemisphere for Western blotting (refer to "3.3. Tissue Preparations for Western Immunoblot" below) and place in 200 μL ice cold homogenization buffer (refer to "3.1. Prepare Solutions" above). 10. Prepare to section the remaining brain tissue by placing it on cork, covering with OCT (Thermofisher Scientific, Waltham, MA, USA; 23-730-571), snap-freezing in isopentane (Thermofisher Scientific, Waltham, MA, USA; 02-002-082), then store it at -80°C until use (refer to section "3.5. Brain Tissue Sectioning" below).

Tissue Preparation for Western Immunoblot. Time for Completion: 45 Minutes
11. Homogenize the tissue removed SN from section "3.2" with 20 strokes in a frosted glass Tenbroek tissue grinder until visibly homogeneous. 12. Centrifuge the homogenate to remove cellular debris. 13. Quantify crude protein homogenates for equal loading using Pierce Rapid Gold Assay (ThermoFisher A53225, Waltham, MA) following the manufacturer's instructions. 14. Resolve 8 μg of homogenate on a 7.5 % TGX gel (BioRAD, 456-1023), transfer to PVDF (polyvinylidene fluoride), and immunoblotted (as described in section "3.4. Western Immunoblot").  19. Wash three times in TBS with gentle agitation on a rocker for 10 minutes each at room temperature. 20. Image the membrane using an Odyssey imaging system (LI-COR Biosciences, Lincoln, NE, USA; Model 9120).

2.3.5.
Brain Tissue Sectioning. Time for Completion: 1:00 Hours per brain, depending on regions of interest.
21. Remove the brain tissue from storage at -80°C and store in dry ice prior to sectioning. 22. Section on cryostat at 20 μm at -15°C and place on Superfrost +/plus slides. a. OPTIONAL STEP Place sections from four separate animals on the same slide for direct comparison. b.
CRITICAL STEP A solvent barrier is required around the border of the slides in the staining section ("3.6. TH Immunohistological Staining" below), so be careful not to place tissue sections close to the edge. 23. Store slides in a sealed container with Drierite dessicant in a -80°C freezer until Immunohistochemical Staining (refer to section "3.6. TH Immunohistological Staining").  from Bregma according to your protocol. 43. OPTIONAL STEP Store slides at 4°C for re-imaging as needed. In our hands, sections minimal signal decay up to 1 year in storage under these conditions.

Results
Motor coordination assessment by rotarod is illustrated in Figure 1. Latency-to-fall measurements indicate rotenone treatment (RO) results in a 2.5 fold lower rotarod performance when compared with control (CT). (61.0 ± 8.2 vs. 173 ± 15 seconds, *p<0.01, n=5, mean ± SEM). No significant differences were found in untreated vs. vehicle-treated rats, (data not shown) thus they were pooled as CT. Rotenone treatment suggests a loss of tyrosine hydroxylase (TH) in the midbrain as shown ( Figure 2). Homogenates of the SN were assessed by TH immunoblot suggesting a loss of TH in rotenone-treated animals (RO) when compared with control (CT), (representative data, Figure 2B) (CT = 564333±117282, RO = 140500±26600, (mean±SE), p = 0.0360 by one-tailed t test, n=3 right-hemispheres/group). The resultant quantification of TH-immunosignal in the midbrain indicates lower TH signal in the midbrain of RO animals when compared with CTs (representative images, 3 animals per group, Figure 2C). Antibody specificity is suggested by human recombinant TH block (TH Block) and reactivity with secondary antibody alone (2°) ( Figure 2C). The region of interest (ROI, yellow, upper panel of Figure 2C) defined the area of measurement for IR signal ( Figure 2D) performed at a digital magnification of 1.5x. Numerical counts of the TH IR signal indicate 31% lower TH signal in RO animals when compared to controls (3.32 ± 0.13 vs. 4.78 ± 0.34 x 10 6 , *p<0.01, n=5, Figure 2E; scale bars = 2 mm).  Bivariate analysis of TH IR 800nm signal was performed against rotarod latency-tofall ( Figure 3). Linear regression analysis of the data set yields a slope of 12120 and yintercept of 3 x 10 6 indicating a modestly positive, significant correlation of the data set with the estimated regression line (R 2 = 0.68, F=15.5, p<0.05, by two-way ANOVA, n=10).

Discussion
Several reports have quantified relative tyrosine hydroxylase levels using near-infrared detection as part of a larger study [38][39][40][41][42]. However, the lesions were all produced differently from this report. The most common inducer was 6-OHDA injected directly into a region of the nigrostriatal pathway [38][39][40]. One study employed rAAV overexpression of α-synuclein [42] and another report performed a functional block/agonist approach of the histamine H1 receptor [41]. While we provide functional correlation between motor coordination and TH levels, these studies did not report correlative data on physiological or functional outcomes with TH levels. Here, we describe, for the first time, IR detection of the quantity of TH after rotenone induction.
Many important differences between our approach and previous studies involve how the brain tissue is processed. The studies cited above applied near-IR detection to traditional pre-processing involving transcardial perfusion fixation followed by a long step of immersion fixation (taking up to 7 days [40]). Fixation is followed by cryopreservation in sucrose (for a period of days, often reported as complete when the specimen sinks) prior to producing thick (40-50 uM) free-floating sections for staining [38,42]. These steps commit the tissue to histological visualization only, precluding homogenization and molecular approaches. The method we report here accomplishes the same task without fixation, making it distinct in the efficiency and flexibility it provides. In fewer steps, we present in this study an additional tool to assess TH in a method resulting in less time to complete experiments and without concerns of error or subjectivityfrom intra-and interobserver variability. Notably, the method presented here using near-IR detection is distinct from other sources because the processing prior to staining is fast and flexible. Furthermore, valuable tissue is saved using 20 μm sections as opposed to thicker sections required by the free-floating section approach. Also, we did not perform transcardial perfusion fixation, allowing us to pursue further molecular studies as demonstrated by the TH Western blot ( Figure 2B) performed on the same animal in which we performed immunohistochemical analysis, providing verification of the potential for two different methods from the same animal as a proof-of-concept. However, the gross anatomical dissection we performed for Western blot analysis is not equivalent to the precise ROI-tracing possible in the LiCor quantification. Because of the gross nature of this dissection for Western, please note that these results are not directly comparable. Western was intended merely as a validation of the potential for analysis of multiple endpoints through different methodologies in the same animal.
Our present method circumvents the weeks of post-processing by using our fresh frozen approach, where the specimen remains frozen continuously until the section contacts the slide at room temperature, at which time it desiccates within about 20-30 seconds. This rapid desiccation serves some of the same purposes as the traditional sucrose-mediated cryopreservation step. During long-term storage at -80°C, we found that we could still detect robust differences in TH levels when comparing RO to CT up to 1 year after harvest from tissue blocks and 6-9 months from fresh frozen mounted slides. The data from tissue treated in this manner indicate that our method sufficiently preserves the specimens to detect differences in TH signal.
We have presented near-IR detection of TH derived from a gross, anatomy-traced format, which may be considered a limitation of the method. Our quantitation included the anatomically adjacent ventral tegmental area, but researchers may choose to limit their anatomical analysis to different regions or nuclei using the same approach. Future advances to this methodology might employ a second target for multiple outcome measures or a control protein that might provide a measure for ratiometric normalization. Conceptually comparing our method to DAB detection of TH in the SNpc and Str, near-IR detection has a greater linear range of detection compared DAB detection due to the enzymatic, exponential nature of DAB, suggesting that the presented near-IR methodology is more capable of detecting subtle changes in TH levels, such as at the early stages of cellular stress; however, future studies are needed to address which detection methods are more statistically sensitive and appropriate for specific study endpoints. Our use of the outbred Sprague-Dawley rats was intentional, as the inter-animal variability represents the variability in the human population. With our technique, we report significant differences with an n=5 per group in this out-bred rat model. Refinement of the molecular target and anatomical regions and nuclei for future studies could benefit different study aims. The effects of long-term storage and tissue handling reported here on the detection of other targets are also not quantified. We have explored the detection of other targets with some success detecting CD11b, a pan microglial marker, in these same tissues using a fluorescence detection approach (data not published). Remarkably, we have also had some success detecting microglial markers by fluorescence approaches on tissues that have already been processed for near-IR TH studies (data not published). If successful in these re-staining approaches, it will be possible to use tissue previously stained and detected by near-IR for re-staining with immunofluorescent antibodies by epifluorescence microscopy. With this advance in the technique, researchers have the capability to study a multitude of targets dependent only upon the number of near-IR and fluorescent probes available for each tissue per slide, allowing researchers to address more questions with fewer tissue sections and possibly with smaller cohorts of animals.

Conclusions
The methods present in the present provide an automated method of quantifying expressed TH levels in sectioned rat brain tissue. We report that the presence of TH signal correlates with motor coordination and show that immunohistochemical detection of TH enzyme by IR-linked secondary antibody may provide a useful measure. We achieve statistical significance with only 5 animals per group, indicating the statistical power of the method using a method without subjective assessment. Our data is presented as its raw output, without normalization or transformation, and so can be directly compared with other studies that use the technique. The presented fresh frozen section approach allows for multiple methodologies in the same tissue, and the time from animal euthanasia to data output can be completed in roughly 5 working days. Furthermore, the fresh frozen tissues presented in this study have been under long-term storage conditions for up to one year prior to immunoreaction and quantification. Once scanned, the saved images can be presented as raw data that can be quantified independently using reported settings to mitigate irreproducibility in reported immunohistological endpoint determination. Using this method also allowed us to reduce the number of animals used while measuring