Ex situ and in situ demonstration of amyloid fibrils for confirmation of amyloidosis using transmission electron microscopy

Abstract

Confirmation of extracellular amyloid deposition across various animal species and tissue types has been a long-standing challenge in veterinary diagnostic pathology. Transmission electron microscopy (TEM) has historically been used to advance the understanding of amyloid fibril morphology and confirm amyloid fibril deposition when histologic methods provide unclear results. We assessed the feasibility of utilizing TEM for routine confirmation of amyloidosis as an addition to histology. We analyzed ex situ amyloid fibrils with direct, negative-contrast TEM and in situ amyloid fibrils with aldehyde-fixed, plastic-embedding TEM to confirm amyloidosis in a variety of cases in which amorphous extracellular amyloid deposits had been identified by H&E and Congo red staining. We compared the 2 TEM methods and documented amyloid fibril morphology and morphometry in 7 species (goat, guinea pig, kudu, fox, sheep, flamingo, and duck). Ex situ fibrils had helical morphology and widths of 15–18 nm across all species. Fibril crossover distances had more interspecies variation of 60–130 nm, and species could be grouped based on pitch (twist size). Twisting patterns of in situ fibrils could not be visualized, but in situ widths of 10 nm were measured across all species. In 4 different chicken cases, fibrils differing morphologically from amyloid were consistently detected via both TEM methods, suggesting the possibility of a non-amyloid deposit that is commonly diagnosed as amyloidosis based on its histologic appearance. When available, we recommend routine confirmation of amyloid fibril deposition by TEM.

Keywords

amyloid fibrils amyloidosis animal avian electron microscopy mammals

Amyloids are aggregates of misfolded proteins that accumulate in the extracellular matrices (ECMs) of affected tissues to cause organ dysfunction in a group of diseases collectively known as amyloidoses.^6,31,43 These proteins have a characteristic cross β-pleated sheet secondary structure and a helical morphology.^6,11,28 Precursor proteins misfold and self-assemble into insoluble amyloid fibrils that are highly resistant to degradation, resulting in intercellular aggregation.^3,31 In AL amyloidosis, these precursor proteins stem from antibody light chains produced in excess by defective plasma cells, which is commonly diagnosed in humans and less frequently diagnosed in animals.⁸

According to the Nomenclature Committee of the International Society of Amyloidosis, 42 amyloid fibril types have been biochemically identified in humans as of 2022, compared to 11 in animals.⁶ AA amyloidosis is one of the most common forms observed in veterinary medicine. AA amyloid fibrils are derived from serum amyloid A (SAA) protein, which comes from the apoSAA precursor protein.⁴⁸ AA amyloidosis usually occurs as a result of chronic inflammation or infection and affects a wide range of species and tissues.^6,33

The clinical diagnosis of amyloidosis is heavily reliant on histologic staining techniques, beginning with the identification of amorphous eosinophilic deposits within the ECM on H&E staining.⁷ Congo red staining is the most popular histologic method of amyloid detection because it is relatively quick, inexpensive, and widely available, allowing amyloid identification by red, yellow, orange, or green birefringence when exposed to polarized light under a light microscope.^20,49 Under ideal staining conditions and with the optimal observer, Congo red staining is considered a very accurate technique. Despite this, in daily practice, Congo red staining is very sensitive to external factors, such as variability in tissue thickness,^12,20,46 tissue type,^5,10 equipment quality,¹⁰ and inter-observer interpretation.^18,20 Additionally, the stain may have an affinity for non-amyloid components of the ECM, such as hyaline,¹³ lipid proteinosis deposits,² cytoskeletal proteins,²⁴ and collagen.^18,46 Due to its aforementioned benefits, Congo red staining should and will remain a primary method of amyloid identification; however, there is a demonstrated need for additional verification in the diagnosis of amyloidosis.^{24,36,42,47,49} Further histologic confirmation, such as thioflavin staining or immunohistochemistry (IHC), can be used to identify amyloid with relatively high specificity.^4,49 Even in these methods, the presence of other physiologic components, such as keratin for thioflavin³⁷ or plasma proteins for IHC,²⁷ may affect the accuracy of results. Additionally, the use of IHC may be limited by the commercial availability of amyloid-specific antibodies.⁴⁵

Transmission electron microscopy (TEM) can be used in conjunction with histologic methods to confirm the presence of amyloid through high-magnification visualization of amyloid fibril deposition.^16,32,49 Among other methods, TEM has played a large role in shaping the understanding of amyloid fibril structure.^16,21 Aldehyde-fixed plastic embedding (AFPE) is a well-established method of TEM sample preparation that allows the study of amorphous intercellular fibrous deposition in situ through the maintenance of cellular ultrastructure.⁴⁴ Negative contrasting (aka “negative staining”) is another method of TEM sample preparation that is essential for characterizing the morphology of ex situ amyloid fibrils.¹⁶ The technique involves coating the grid in a thick, electron-dense stain, which results in strong contrast between the darker background and the lighter particle of interest.³⁵ Negative staining allows for the rapid, direct visualization of macromolecules and viruses in a purified state for diagnostic purposes.³⁹ Yet, amyloid fibril purification techniques for negative-contrast TEM have seen little change since the standard technique was developed in the 1960s,³⁸ leaving the method too time-consuming for routine diagnostic use.

Similar to negative-contrast TEM, cryogenic TEM uses purification methods that provide clear depictions of the morphologic features of amyloid peptide tertiary structure, allowing in-depth ex situ analysis of the fibrils by minimizing unwanted osmotic changes in proteins through cryofixation.^32,44 Although highly informative, all TEM techniques are often limited in practice by their accessibility.⁴⁹ Beyond confirmation of amyloidosis by TEM lie methods that provide even greater detail in analyzing molecular and structural aspects of the fibrils, such as x-ray crystallography.^6,21 Advanced data on amyloid fibrils provided by these methods are essential for driving our understanding of amyloidosis as a disease but are not necessary for routine diagnostic use given the relative frequency of the disease in veterinary pathology.

We explored the feasibility of using negative-contrast and AFPE TEM for ex situ and in situ confirmation of amyloid deposition, in addition to histologic methods. In doing so, we attempted to optimize the efficiency of an established negative-contrast TEM technique³⁸ for rapid detection of amyloid in a diagnostic setting. Providing a standardized amyloid extraction technique suitable for routine diagnostic use would facilitate expansion of the morphometric characterization of amyloid fibrils. The development of this morphometric database is tightly intertwined with the process of diagnosing amyloidosis with high specificity based on fibril morphometry. We analyzed and cataloged the morphometric dimensions of amyloid fibrils in 7 animal species to be used for future study and reference in diagnostic contexts.

Materials and methods

Tissue samples

We included in our study tissues from 2 goats (Capra hircus), 2 guinea pigs (Cavia porcellus), 2 kudus (Tragelaphus strepsiceros), 2 foxes (Urocyon littoralis), 1 sheep (Ovis aries), 1 flamingo (Phoenicopterus ruber), 1 duck (Netta rufina), and 4 chickens (Gallus gallus domesticus; Table 1). Amyloidosis in these animals was confirmed histologically by submitting pathologists pre-study. The sheep spleen was an exception, as it had amorphous extracellular deposits but was initially determined to be negative with Congo red staining pre-study. For consistency, all samples had H&E and Congo red histologic stains repeated following internal standard operating procedures and interpreted by at least 2 ACVP-certified pathologists at the California Animal Health and Food Safety laboratory, Davis branch (Davis, CA, USA; Suppl. Table 1). H&E-stained slides for each tissue had amorphous extracellular deposits, and Congo red stains displayed colorful birefringence when exposed to polarized light, ostensibly indicating the presence of amyloid fibril deposition. The sheep spleen was positive for amyloid deposition on the standardized Congo red stain. Pre-study Congo red staining results were determined either positive or negative in a binary manner; the standardized Congo red distribution and severity were semi-quantitatively assessed as diffuse, multifocal, focal, negative (–), minimal (+), mild +, mild-to-moderate +(+), moderate ++, moderate-to-severe ++(+), or severe +++.

Table 1.

Animal species, tissue types, and causes of suspected amyloidosis in samples used in our study.

Species	n	Tissue(s)	Cause of amyloidosis
Chicken	4	Liver	Chronic inflammation; lymphoproliferative disease
Goat	3	Uterine caruncle	Endometritis
Fox	2	Tongue	Undetermined
Guinea pig	2	Kidney, liver	Chronic inflammation (pododermatitis)
Kudu	2	Abomasum, kidney, liver, spleen	Chronic inflammation (osteomyelitis)
Duck	1	Liver	Chronic inflammation (mycobacteriosis)
Flamingo	1	Kidney, spleen	Chronic inflammation (pododermatitis)
Sheep	1	Kidney, liver, spleen	Chronic inflammation (abscesses)

Transmission electron microscopy

For TEM, both formalin-fixed tissue and unfixed tissue frozen at −80°C were obtained from each case. One of the goats and both foxes were used as positive controls, based on previous confirmations of amyloidosis in the specimens. In addition to histologic evidence of amyloid deposition in the goat (H&E, Congo red, and IHC; Suppl. Fig. 1A–C), SAA3 amyloid deposition was confirmed using mass spectrometry, quantitative real-time PCR, and genetic sequencing methods.¹⁴ Amyloidosis had been confirmed in the foxes by IHC and negative-contrast TEM.¹⁵

Preparation of fibrils for ex situ examination

Baseline method

Validation of the modified (rapid) method of an amyloid concentration technique was accomplished by comparing preliminary experimental results to a reliable baseline method similar to the published technique.³⁸ This validation was performed using the positive controls noted. A goat spleen and uterine caruncle without histologic evidence of amyloidosis were used as negative controls. All control samples were frozen at −80°C and thawed immediately before processing. Each test was repeated 5 times.

Throughout our study, homogenization was performed with a bead mill homogenizer (Bead Ruptor 12; Omni International) set at 5 m/s for 45 s. All ultracentrifugation was performed in an Optima L-80 XP (Beckman Coulter). Samples were viewed on a transmission electron microscope (JEM-1400Plus; JEOL). The baseline method is described below.

Tissue was weighed (1 g), minced with a scalpel, and homogenized. The homogenate was then suspended in 10 mL of 0.9% NaCl solution (saline) and ultracentrifuged at 12,100 × g for 30 min at 4°C. The supernatant was discarded, and the pellet was resuspended in saline, homogenized, and ultracentrifuged again under the same conditions. This washing process was repeated for 6 saline washes with homogenization occurring between each wash. After the sixth and final saline wash in the ultracentrifuge, the pellet was once again collected and homogenized. This time, it was resuspended in 10 mL of double-distilled water (DDW). The suspension was then ultracentrifuged at 12,100 × g for 30 min at 4°C. The supernatant (“S1”) was discarded, and the pellet was resuspended in DDW, homogenized, and ultracentrifuged again. The supernatant (“S2”) was collected and stored. The pellet was re-suspended, homogenized, and ultracentrifuged twice more, collecting the supernatants from each run (“S3” and “S4”). After the fourth DDW wash, S2, S3, and S4 were combined, and the resulting mixture was ultracentrifuged at 125,000 × g for 60 min at 4°C. The supernatant was discarded, and the pellet was resuspended in 400 μL of boiled DDW.

Modified method

The washing technique from the baseline method was streamlined and used as a platform for the modified version. The number of washing steps was reduced from 6 saline washes and 4 water washes to a single saline wash and a single water wash. Samples of the same positive and negative control tissues from the baseline method were used in the same manner: frozen at −80°C, thawed immediately before processing, and tested with 5 repeats. All other cases were also processed with the modified method, repeated 3 times each. The duck liver was an exception; it was processed using an earlier, developmental version of the modified method, but it was not re-processed using the exact modified method described in our study due to limited sample availability. The duck liver was included for amyloid fibril measurement documentation purposes; its method of preparation differed from the modified method only in its g-force measurements (1,200 × g washes). The modified method is described below.

Tissues were weighed (1 g), minced with a scalpel, and homogenized before being suspended in 10 mL of saline. This suspension was ultracentrifuged at 12,100 × g for 30 min at 4°C. The supernatant was removed, and the pellet was re-suspended in 10 mL of DDW and homogenized once more. This new suspension was ultracentrifuged at 12,100 × g for 30 min at 4°C. The resulting supernatant was ultracentrifuged at 125,000 × g for 60 min. The supernatant from this spin was discarded, and the pellet was re-suspended in 400 μL of boiled DDW.

For both methods, the pellet from the final ultracentrifugation spin was resuspended in 400 μL of boiled DDW and contrasted using a drop-to-grid method.¹ Briefly, a 25-μL drop of alcian blue stain was applied to the surface of a formvar-copper–coated mesh grid (Ted Pella) as a coating agent. The grid was then rinsed and blotted dry before a 10-μL aliquot of the suspension was placed on the grid and incubated for 30 min at room temperature (23°C). The grid was once again washed and blotted, and a drop of either 1% phosphotungstic acid or 1% uranyl acetate was applied to the grid for 1 min as a negative contrast.

Phosphotungstic acid and uranyl acetate are 2 commonly used negative contrasts. Using the positive control goat uterine caruncle with the modified method, 3 grids of each contrast were compared to assess discrepancies between the 2 contrasting options.

Preparation of fibrils for in situ examination

Analysis of in situ fibrils expanding intercellular space were performed on formalin-fixed tissues, as prepared by methods described elsewhere.¹ In summary, 1-mm³ tissue fragments were postfixed with 2.5% glutaraldehyde in 0.1 M sodium cacodylate buffer before further post-fixation with 1% osmium tetroxide in 0.1 M sodium cacodylate buffer. Samples were then washed 3 times in DDW followed by dehydration with a 25–100% ethyl alcohol gradient. Next, tissues were infiltrated with 1:1 ethanol:resin (EMBed-812; Electron Microscopy Services) for 1 h, transferred to a 1:3 ethanol:resin (EMBed-812) mixture for 1 h, and finally infiltrated with 100% resin before being embedded and incubated at 58°C for 48 h. Following polymerization of the resin, samples were trimmed and sectioned on an ultramicrotome (UC6; Leica Microsystems) at 60–70 nm. These thin sections were collected on a 100-mesh nickel grid and contrasted with 5% uranyl acetate for 2 min, followed by ready-to-use lead citrate for 6 min. All reagents were from Electron Microscopy Sciences.

Screening and characterization of amyloid

For ex situ amyloid fibril measurements under negative-contrast TEM, the width of each of 30 fibrils was measured at the center (point of greatest width) of a complete twist on an individual fibril to ensure consistency in morphometric documentation in and across species. Conversely, ex situ fibril crossover distance or “pitch” (i.e., the distance of a single, complete twist) was measured between 2 consecutive points on the fibril with the narrowest widths (n = 15). Amyloid fibril ridges visible under negative-contrast TEM were interpreted as protofilaments^6,22,23 and measured as well (n = 30; Fig. 1). Under AFPE TEM, in situ fibrils could only be measured in width (n = 30).

Figure 1.

Amyloid fibrils in a sheep spleen. Transmission electron microscopy, negatively contrasted with 1% uranyl acetate. Width and crossover distance defined on amyloid fibril diagram. Crossover distance measured between 2 adjacent arrowheads on ex vivo amyloid fibril. Amyloid fibril width shown by black bar and protofilament width shown by white bar on magnified ex vivo amyloid fibril.

Results

Ex situ fibrils

Unbranched, twisting fibrils morphologically consistent with AA amyloid were extracted and visualized in both the baseline method and the modified method (Suppl. Fig. 1D). Grids from both methods were primarily coated by larger clumps of clustered fibrils with individual fibrils scattered around the periphery of clusters. Protofilaments were prominent, resembling pairs of intertwined cord-like structures. No fibrils were detected in the negative control samples. Amyloid fibrils were likewise visualized in the guinea pigs, kudus, foxes, sheep (Fig. 2), and flamingo using the modified method, as well as in the duck liver. In addition to amyloid, collagen fibrils were also visualized in the goats (Suppl. Fig. 2).

Figure 2.

Amyloid fibrils. A. Guinea pig kidney. B. Kudu abomasum. C. Fox tongue. D. Sheep spleen. Transmission electron microscopy, negatively contrasted with 1% uranyl acetate. Bars = 200 nm.

The dimensions of visualized amyloid fibrils were obtained from the 7 species of animals by negative-contrast TEM. Fibril length varied widely within each sample, likely due to breakage or cleavage during sample preparation.²⁹ The negative contrast results for amyloid fibril width were strikingly similar across all species (15–18 nm; Table 2). Protofilament widths were 7–10 nm. Crossover distance varied more between species. The goats, duck, and guinea pigs all had smaller crossover distances of 65.8 ± 3.0, 70.1 ± 8.4, and 75.8 ± 2.5 nm, respectively. Amyloid fibrils in the foxes had a midsize crossover distance of 110.0 ± 3.0 nm; the sheep, kudus, and flamingo had larger crossover distances of 125.1 ± 2.4, 127.7 ± 2.1, and 127.9 ± 2.5 nm, respectively. Amyloid morphometry was very similar intraspecifically and across different tissue types from the same individual (Suppl. Table 2). Protofilament width was unable to be obtained in the duck, as its developmental preparation method did not provide clear images of defined protofilaments within the fibrils.

Table 2.

Amyloid fibril width, crossover distance, and protofilament width obtained by negative-contrast transmission electron microscopy (TEM), as well as amyloid fibril width obtained by aldehyde-fixed, plastic-embedding (AFPE) TEM, in 7 animal species.

Species	Tissue	Negative-contrast TEM			AFPE TEM
		Fibril		Protofilament	Fibril
		Width, nm	Crossover, nm	Width, nm	Width, nm
Duck (n = 1)	Liver	16.0 ± 2.1	70.1 ± 8.4	UTO	8.3 ± 1.2
Flamingo (n = 1)	Kidney, spleen	15.4 ± 1.4	127.9 ± 2.5	7.8 ± 0.9	9.5 ± 1.4
Fox (n = 2)	Tongue	17.7 ± 1.1	110.0 ± 3.0	8.5 ± 0.8	UTO
Goat (n = 1)	Uterine caruncle	16.4 ± 1.2	65.8 ± 3.0	8.3 ± 0.9	9.8 ± 1.8
Guinea pig (n = 2)	Kidney, liver	15.9 ± 1.1	75.8 ± 2.5	8.5 ± 1.5	7.3 ± 2.0
Kudu (n = 2)	Abomasum, kidney, liver, spleen	17.7 ± 1.2	127.7 ± 2.1	8.9 ± 1.1	9.6 ± 2.0
Sheep (n = 1)	Kidney, liver, spleen	17.6 ± 0.9	125.1 ± 2.4	8.6 ± 0.8	11.3 ± 1.7

UTO = unable to obtain.

Interestingly, all 4 chicken livers contained fibrils that did not have periodic twisting patterns (Fig. 3). These fibrils were relatively scarce on the grid compared to the copious amounts of amyloid recovered from other cases with similar amounts of extracellular deposits, as determined by histology. Additionally, the chicken fibrils were observed individually, in non-intertwined pairs, and in parallel stacks and clusters of up to 6 fibrils, differing markedly from the amyloid fibrils observed in their avian counterparts, the flamingo and duck. Because twisting is a fundamental attribute of amyloid fibril structure,⁶ we concluded that the fibrils observed in the chicken livers were likely not amyloid based on their morphologic characteristics. These fibrils were ~10-nm wide.

Figure 3.

Non-amyloid fibrils in a chicken liver. A, B. Black arrows highlight non-intertwined pairs of fibrils. White arrow in (B) highlights a parallel stacked bundle of fibrils. Transmission electron microscopy, negatively contrasted with 1% uranyl acetate. Bars = 200 nm.

In situ fibrils

Under AFPE TEM, fibrils were observed expanding the ECM in all species. In the 7 species that were positive for amyloid deposition by negative-contrast TEM, fibrils were observed to be much thinner than on negative contrast, ranging from 7–12-nm wide across all 7 species (Fig. 4A –C; Suppl. Fig. 1E). Details such as fibril twisting or protofilaments could not be observed in situ; these amyloid fibrils were arranged haphazardly in the ECM with no apparent organization or directional orientation. Collagen fibrils were additionally present in the goats, though not as distinguishable from amyloid as the ex situ preparation (Suppl. Fig. 2).

Figure 4.

Transmission electron microscopy, aldehyde-fixed plastic embedding. A. Amyloid deposition in the extracellular matrix of a duck liver. Double-headed arrows indicate expansion of the interstitium. Cap = capillary; He = hepatocyte. Bar = 5 µm. B, C. Close-up of the white and black boxes in (A), respectively. Note the smaller size and the denser, haphazard arrangement of these amyloid fibrils compared to the non-amyloid fibrils in the chicken liver. Bars = 500 nm. D. Amorphous extracellular deposits in a chicken liver. Bar = 5 µm. E. Close-up of the white box in (D). Note the size, density, and uniform directionality of the non-amyloid fibrils. Bar = 500 nm. F. Close-up of the black box in (D). Black arrow indicates a cluster of non-amyloid fibrils. Note the lack of density in the amorphous matrix behind the fibrils. Bar = 500 nm.

Similar to negative-contrast TEM, fibrils observed in the ECM of chicken livers were ~10-nm wide under AFPE TEM (Fig. 4D –F). Unlike in situ amyloid fibrils observed in the other species, the fibrils observed in the chicken livers were less dense, with most space being covered by an amorphous matrix background. Additionally, the fibrils were oriented parallel to each other as if aligned in a specific direction (Fig. 4E). Because this in situ presentation differed so starkly from the denser, disorganized ECM deposition of amyloid fibrils, the fibrils in the chicken liver under AFPE TEM were unlikely to be amyloid fibrils.

These AFPE TEM findings are consistent with subtle discrepancies between the Congo red presentations of the chicken livers and tissues with confirmed amyloidosis (Suppl. Fig. 3). In situ visualization confirms a general directional trend in these chicken fibrils that is noticeable in Congo red stains. When stained with Congo red, the chicken livers also had a faint-red color without a polarization filter (Suppl. Fig. 3B), whereas amyloid-laden tissues stain with a denser orange-red color in the controls and confirmed cases of our study (Suppl. Fig. 3C). Upon polarization, the chicken livers had spotty, unidirectional streaks of light-green birefringence that were not necessarily correlated with areas of red staining. Directionality is additionally reflected in their affinity for parallel stacking under negative-contrast TEM. This Congo red presentation differs from amyloid-laden tissues that consistently display a deeper orange-red coloration pre-polarization (Suppl. Fig. 3E), which is correlated with a multidirectional birefringence upon polarization (Suppl. Fig. 3F). The haphazard arrangement of amyloid fibrils is not only reflected in this multidirectional birefringence but in both AFPE and negative-contrast TEM presentations, as well (Figs. 2, 4A –C).

Discussion

We demonstrated that AFPE and negative-contrast TEM can be used in addition to histologic staining for the confirmation of amyloid fibril deposition. Further, our modified amyloid extraction technique can be utilized as a rapid tool to confirm amyloidosis with high specificity when paired with negative-contrast TEM. Using this adaptation of a standard method,³⁸ results are obtained within 3–4 h of tissue reception. The sensitivity of the modified negative stain TEM method appears to be at least equivalent to both the established baseline method and AFPE TEM. The modified method recovered similar quantities of amyloid fibrils from tissue compared to the baseline method. On the microscopic level, both techniques proved equally effective at cleansing the sample of debris. The relative ineffectiveness of extra lavages in the baseline method is likely attributable to the efficacy of the bead mill homogenizer, which effectively frees intercellular contents through thorough membrane destruction.⁴¹ This allows for the swift removal of most of the cellular debris in a single saline wash, rendering subsequent washes unproductive. The modified method eliminates the need for additional washes and reduces centrifugation time from 480 min over 11 runs in the baseline method to 120 min over 3 runs. For reference, comparable published versions of this amyloid extraction technique that utilize sonication as the method of homogenization require 570 min over 10 runs³⁸ or 695 min over 14 runs.²⁹

The optimal amyloid fibril extraction technique is one in which the precipitation gradient is established to effectively cleanse the sample of debris and salt while keeping the amyloid suspended in water before the final sedimentation spin.³⁸ The baseline method that we employed was a combination of amyloid extraction and ultracentrifugation principles established by previous studies. A g-force of 12,100 × g was selected for the lavage steps based on methods of saline lavage.⁹ Similar to the standard method, 6 saline washes and 4 water washes were performed at 30 min each, homogenizing the sample between each.³⁸ Unlike the standard method, however, the ultracentrifugal g-force was kept the same for the saline and water washing steps, as done by others,²⁹ to prevent prematurely precipitating the amyloid at higher speeds. Also modeled after the alternative method was the final 125,000 × g ultracentrifugation spin for 60 min to precipitate the amyloid completely,²⁹ which adheres to the finding that 100,000 × g for 1 h is sufficient to completely sediment amyloid.³⁸

One possible drawback to this amyloid purification technique for ex situ visualization is that tissues with considerable amounts of cellular debris may obscure individual amyloid fibrils; however, visualization and morphometric analysis of amyloid fibrils in the goat uterine caruncles was easily achieved despite large quantities of cellular debris. Further research to determine the exact specifications of amyloid precipitation gradients could improve this TEM method. Other studies have successfully explored the use of proteases for enhancing amyloid identifiability without diminishing amyloid quantity or quality.⁹ The implementation of protein-targeting immunogold techniques or technologically advanced TEM programs, such as 3D tomography, might also increase the sensitivity of this process.

Background and definition are 2 of the most important aspects of TEM when it comes to identifying amyloid fibrils in a negative-contrast preparation. A good background facilitates amyloid detection, whereas good definition allows for accurate measurements of the fibrils’ dimensions. Contrast type can affect these 2 properties. A comparison of phosphotungstic acid and uranyl acetate demonstrated that variation in these properties was more contingent on external circumstances than negative-contrast type. Because phosphotungstic acid and uranyl acetate are rather similar in terms of background and contrast, the benefit of either contrast lies in the convenience and personal preference of the microscope technician.

Liberation of intercellular contents via bead mill homogenization followed by the removal of soluble proteins in ex situ preparation enables analysis of each fibril as an individual entity, leading to easier differentiation of various intercellular materials in comparison with AFPE TEM methods. Negative contrasting also better preserves the morphologic integrity of these fibrils, allowing for accurate depiction at higher magnifications with minimal artifacts. The smaller, more variable fibril widths observed in AFPE tissues in our study are likely artifacts of the fixation process.^40,47 Similar to cryogenic TEM, negative-contrast TEM measurements of AA amyloid fibrils obtained from tissue by direct purification accurately reflect fibril dimensions in vivo through the avoidance of fixatives.^32,44 Further, AFPE TEM methods do not utilize homogenization and leave amyloid fibrils confined within intercellular spaces. This creates an overabundance of amorphous deposits in the area of interest that can present difficulty in distinguishing amyloid fibrils from other structural, extracellular proteins.

One important observation was the perceived groupings of fibril crossover distances across species. This has been reported with phylogenetic analyses of SAA gene sequences confirming observed interspecific correlations between crossover size and amyloid genetics, although distinctive SAA subtypes (i.e., SAA1, SAA2, etc.) were not specified.²⁹ Similar to the previous study, amyloid fibril morphology in our study may be interpreted as falling into 3 groups based on crossover size: 60–80, 105–115, and 120–130 nm. This would imply that the goats, duck, and guinea pigs comprise the smallest group, the foxes are alone in the mid-sized group, and the sheep, kudus, and flamingo comprise the large group. Of these animals, only the SAA identity of one goat (SAA3) is confirmed.¹⁴ Without molecular analysis, these groupings are rather arbitrary, although future documentation and research in this domain may allow morphometry-based identification of specific amyloid types (i.e., AA, AL, ATTR) and/or SAA subtypes. Further research into the correlation of width and crossover distance is also suggested.

Diagnosis of amyloidosis without further specification of type or SAA subtype is generally sufficient in veterinary practices because the investigation focuses on the underlying cause, given the lack of amyloidosis treatment strategies. Although branching patterns have been observed in mice and camels,²⁹ it is important to note that none of the definitionally nonbranching amyloid fibrils observed in the 7 species that we studied displayed any sort of branching. Any possible appearance of branching in our study was interpreted as a superposition of multiple, distinct fibrils. It is worth noting that the dimensions of goat amyloid fibrils measured in the previous study (21.2 ± 2.0-nm wide, 159.5 ± 2.7-nm crossover distance)²⁹ are significantly larger than those measured in our study (16.4 ± 1.2-nm wide, 65.8 ± 3.0-nm crossover distance); however, the SAA identity may differ between affected goats and drive this morphometric divergence.

Moreover, visualization of negatively contrasted amyloid fibrils appeared to be composed of 2 protofilaments resembling intertwined “cord-like” strands,^6,22,23 each averaging ~7–10-nm diameter. Documented 3D reconstruction of amyloid fibril morphology has shown these to be the fibrils’ β-sheet twists.²⁵ Coating these mere peptide stacks in negative contrast gives them a more prominent, rounded morphology that garners an appreciation for the high resolution at near-molecular levels of TEM magnification.

The use of TEM discerned that the fibrils in the chicken livers were likely not amyloid. Three of the chicken livers were examined 3 times to verify results. Although polymorphism has been shown to exist in AA amyloid fibrils, helical twisting is a constant, defining trait among all morphologic variants.³⁰ Combined with amorphous extracellular deposition on H&E slides, the mere presence of green birefringence in all 4 chicken Congo red stains may sway pathologists toward a positive diagnosis. The sheer amount of birefringence in these chicken livers suggests that more fibrils would be recovered with the modified negative-contrast TEM technique if this were truly amyloid. The relatively low number of fibrils that were recovered indicates that either the molecules highlighted by polarization are soluble, non-amyloid proteins removed via saline and/or water lavages or that the source of the birefringence is caused by some other component in the amorphous ECM. One possible explanation for the presence of the spotty birefringence is simply low amounts of amyloid deposition that were not visualized on TEM; however, the amount of amorphous material in the ECMs of the chicken livers is consistent with other challenging cases in which we extracted large amounts of amyloid. Another possibility is that these observed “non-amyloid” fibrils themselves are the source of the birefringence,¹⁸ and this remains completely unknown. AA amyloidosis is the only type of amyloidosis that has been found in birds,¹⁷ and, although multiple origins of the disease exist,^19,26,34 the ECM deposit observed in these chickens does not appear to be one of them. Although further investigation lies beyond the scope of our study, there is a demonstrated need for the characterization of these fibrils and this possibly non-amyloidosis condition causing extracellular accumulation in chickens. Methods such as x-ray diffraction or immunolabeling would be helpful in providing a definitive diagnosis.

The systemic problem of inconsistency in Congo red stain interpretation is seen across many vertebrate species and is not restricted to chickens. The case of these chickens exemplifies the need for additional confirmatory methods when histologic stains, such as Congo red, provide unclear results. When available, we recommend the use of AFPE and/or negative-contrast TEM for high-specificity confirmation of amyloid fibril deposition via visualization of fibrils in situ or ex situ, respectively. TEM has been a driving force in the advancement of understanding amyloidosis for decades, and these techniques allow for an unbiased diagnosis of amyloidosis based on fibril morphometry across all species in the animal kingdom. Continuous expansion of the amyloid fibril morphometric database is the most necessary future step to bolster both existing knowledge about amyloidosis and the accuracy of these TEM techniques.

Supplemental Material

sj-pdf-1-vdi-10.1177_10406387251321415 – Supplemental material for Ex situ and in situ demonstration of amyloid fibrils for confirmation of amyloidosis using transmission electron microscopy

Supplemental material, sj-pdf-1-vdi-10.1177_10406387251321415 for Ex situ and in situ demonstration of amyloid fibrils for confirmation of amyloidosis using transmission electron microscopy by Robert Polon, Christina R. Heard, Omar Gonzales-Viera, Melissa Macías-Rioseco, Aslı Mete, Katherine Watson, Leslie W. Woods and Aníbal G. Armién in Journal of Veterinary Diagnostic Investigation

Footnotes

Acknowledgements

We thank the California Animal Health and Food Safety (CAHFS) Laboratory and American Association of Veterinary Laboratory Diagnosticians (AAVLD) for supporting this project, as well as Mrs. Rosa Lynn Mañalac for outstanding electron microscopy preparations and photographs. We additionally thank Drs. Arno Wuschmann, John Adaska, and Regina Zavodovskaya for the case contributions.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Melissa Macías-Rioseco

Aslı Mete

Katherine Watson

Aníbal G. Armién

Supplemental material

Supplemental material for this article is available online.

References

Armién

, et al Molecular and biological characterization of a cervidpoxvirus isolated from moose with necrotizing dermatitis. Vet Pathol 2020;57:296–310.

Bayer-Garner

Smoller

BR.

AL amyloidosis is not present as an incidental finding in cutaneous biopsies of patients with multiple myeloma. Clin Exp Dermatol 2002;27:240–242.

Berthelot

, et al

What does make an amyloid toxic: morphology, structure or interaction with membrane?

Biochimie 2013;95:12–19.

Biancalana

Koide

Molecular mechanism of thioflavin-T binding to amyloid fibrils. Biochim Biophys Acta 2010;1804:1405–1412.

Bowen

, et al AL-amyloidosis presenting with negative Congo red staining in the setting of high clinical suspicion: a case report. Case Rep Nephrol 2012;2012:593460.

Buxbaum

, et al Amyloid nomenclature 2022: update, novel proteins, and recommendations by the International Society of Amyloidosis (ISA) Nomenclature Committee. Amyloid 2022;29:213–219.

Chan

JK.

The wonderful colors of the hematoxylin-eosin stain in diagnostic surgical pathology. Int J Surg Pathol 2014;22:12–32.

Chuang

, et al Amyloid assembly and disassembly. J Cell Sci 2018;131:189928.

Cohen

Calkins

The isolation of amyloid fibrils and a study of the effect of collagenase and hyaluronidase. J Cell Biol 1964;21:481–486.

10.

Devata

, et al Detection of amyloid in abdominal fat pad aspirates in early amyloidosis: role of electron microscopy and Congo red stained cell block sections. Cytojournal 2011;8:11.

11.

Eanes

Glenner

GG.

X-ray diffraction studies on amyloid filaments. J Histochem Cytochem 1968;16:673–677.

12.

Elghetany

Saleem

Methods for staining amyloid in tissues: a review. Stain Technol 1988;63:201–212.

13.

Fernandez-Flores

A review of amyloid staining: methods and artifacts. Biotech Histochem 2011;86:293–301.

14.

Gaffney

, et al Protein profiling of isolated uterine AA amyloidosis causing fetal death in goats. FASEB J 2015;29:911–919.

15.

Gaffney

, et al Systemic amyloid A amyloidosis in island foxes (Urocyon littoralis): severity and risk factors. Vet Pathol 2016;53:637–647.

16.

Gras

, et al Transmission electron microscopy of amyloid fibrils. Methods Mol Biol 2011;752:197–214.

17.

Guo

, et al Characterization of serum amyloid A protein mRNA expression and secondary amyloidosis in the domestic duck. Proc Natl Acad Sci U S A 1996;93:14548–14553.

18.

Guy

Jones

CK.

Abdominal fat pad aspiration biopsy for tissue confirmation of systemic amyloidosis: specificity, positive predictive value, and diagnostic pitfalls. Diagn Cytopathol 2001;24:181–185.

19.

Habibi

, et al Experimental induction of chicken amyloid A amyloidosis in white layer chickens by inoculation with inactivated vaccines. Avian Pathol 2017;46:497–505.

20.

Howie

AJ.

Origins of a pervasive, erroneous idea: the “green birefringence” of Congo red-stained amyloid. Int J Exp Pathol 2019;100:208–221.

21.

Iadanza

, et al A new era for understanding amyloid structures and disease. Nat Rev Mol Cell Biol 2018;19:755–773.

22.

Jahn

Radford

SE.

Folding versus aggregation: polypeptide conformations on competing pathways. Arch Biochem Biophys 2008;469:100–117.

23.

Jiménez

, et al The protofilament structure of insulin amyloid fibrils. Proc Natl Acad Sci U S A 2002;99:9196–9201.

24.

Khurana

, et al

Is Congo red an amyloid-specific dye?

J Biol Chem 2001;276:22715–22721.

25.

Kollmer

, et al Cryo-EM structure and polymorphism of Aβ amyloid fibrils purified from Alzheimer’s brain tissue. Nat Commun 2019;10:4760.

26.

Landman

WJ.

Amyloid arthropathy in chickens. Vet Q 1999;21:78–82.

27.

Lavatelli

Vrana

JA.

Proteomic typing of amyloid deposits in systemic amyloidoses. Amyloid 2011;18:177–182.

28.

, et al Amyloid fibril structure of α-synuclein determined by cryo-electron microscopy. Cell Res 2018;28:897–903.

29.

Lin

, et al Comparison of AA amyloid fibril morphology and serum amyloid A gene sequence in 5 animal species. Vet Pathol 2021;58:369–375.

30.

Makowski

The structural basis of amyloid strains in Alzheimer’s disease. ACS Biomater Sci Eng 2020;6:2498–2505.

31.

Merlini

Bellotti

Molecular mechanisms of amyloidosis. N Engl J Med 2003;349:583–596.

32.

Mollee

, et al How to diagnose amyloidosis. Intern Med J 2014;44:7–17.

33.

Murakami

, et al Transmission of systemic AA amyloidosis in animals. Vet Pathol 2014;51:363–371.

34.

Nakayama

, et al Spontaneous, experimentally induced, and transmissible AA amyloidosis in Japanese quail (Coturnix japonica). Vet Pathol 2017;54:912–921.

35.

Ohi

, et al Negative staining and image classification—powerful tools in modern electron microscopy. Biol Proced Online 2004;6:23–34.

36.

Ouchi

, et al Clinical significance of Congo red test. Tohoku J Exp Med 1976;118(Suppl):191–198.

37.

Penticoff

, et al Survey of amyloidosis cases among different free-living wild and zoo animals. Amyloid 2021;28:145–152.

38.

Pras

, et al The characterization of soluble amyloid prepared in water. J Clin Invest 1968;47:924–933.

39.

Scarff

, et al Variations on negative stain electron microscopy methods: tools for tackling challenging systems [Erratum appears in J Vis Exp. 2020 Jan 30;(155)]. J Vis Exp 2018;132:57199.

40.

Shirahama

Cohen

AS.

High-resolution electron microscopic analysis of the amyloid fibril. J Cell Biol 1967;33:679–708.

41.

Simpson

RJ.

Homogenization of mammalian tissue. Cold Spring Harb Protoc 2010;2010:pdb.prot5455.

42.

Stemmerman

Auerbach

OF.

The value and limitations of the Congo red test for amyloidosis. Am J Med Sci 1944;208:305.

43.

Tennent

GA.

Isolation and characterization of amyloid fibrils from tissue. Methods Enzymol 1999;309:26–47.

44.

Tizro

, et al Sample preparation for transmission electron microscopy. Methods Mol Biol 2019;1897:417–424.

45.

Westermark

, et al Development of mouse monoclonal antibodies against human amyloid fibril proteins for diagnostic and research purposes. Methods Mol Biol 2018;1779:401–414.

46.

Westermark

, et al Staining methods for identification of amyloid in tissue. Methods Enzymol 1999;309:3–25.

47.

Winey

, et al Conventional transmission electron microscopy. Mol Biol Cell 2014;25:319–323.

48.

Woldemeskel

A concise review of amyloidosis in animals. Vet Med Int 2012;2012:427296.

49.

Yakupova

, et al Congo red and amyloids: history and relationship. Biosci Rep 2019;39:BSR20181415.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

1.95 MB