Formic acid treatment drastically reduces sporadic Creutzfeldt–Jakob disease and variant Creutzfeldt–Jakob disease infectivity in histological samples as tested in a high-sensitivity mouse bioassay

Abstract

Formic acid treatment is widely used in diagnostic neuropathology to reduce the infectivity of prion-containing tissues; however, quantitative in vivo evidence supporting its effectiveness under routine laboratory conditions remains limited. Here, we assessed the impact of formalin fixation and formic acid treatment on the infectivity of type 1 sporadic Creutzfeldt–Jakob disease (sCJD) and variant CJD (vCJD) prions using highly sensitive transgenic mouse models overexpressing human-PrP M129 (Hu-Tg340) or bovine PrP (Bo-Tg110). Brain tissues were processed under conditions closely resembling standard histopathological workflows and tested as untreated, formalin-fixed or formalin-plus-formic-acid–treated inocula. Untreated samples produced short incubation times and full attack rates, whereas formalin fixation caused only a modest prolongation of incubation times. In contrast, formic acid treatment markedly extended incubation times and reduced attack rates for sCJD. Based on incubation-time interpolation, the estimated infectivity reductions were on the order of 4.4 log₁₀ for vCJD and 5 log₁₀ for sCJD. These estimates indicate a major reduction in infectious titre under the conditions tested, although residual infectivity was still detected. The findings support formic acid treatment as an important risk-reduction step in routine neuropathology workflows for the two prion strains examined.

Impact Statement.

Prion diseases pose major biosafety challenges because infectious prions can persist in tissues even after standard fixation. Although formic acid treatment is widely recommended to reduce prion infectivity in diagnostic laboratories, quantitative in vivo evidence under routine histopathology conditions has been limited. This study fills this gap by directly measuring residual infectivity after formalin fixation followed by the commonly used 1 h formic acid treatment.

Using highly sensitive transgenic mouse models, we show that formic acid treatment resulted in an estimated reduction of infectivity of ~4.4–5 log₁₀ for the sporadic Creutzfeldt–Jakob disease and variant Creutzfeldt–Jakob disease agents tested, whereas formalin treatment alone produces only a modest effect. These results are relevant not only for prion researchers but also for neuropathologists, laboratory technicians, biosafety officers and institutions handling potentially infectious brain tissue.

This work provides a quantitative benchmark for evaluating prion inactivation procedures and strengthens the evidence base supporting current biosafety guidelines. Although formic acid does not eliminate infectivity completely, it removes the vast majority of infectious prions, providing a critical safety margin. This study therefore represents an important step toward validating practical measures that improve laboratory safety when handling prion-infected samples.

Introduction

Prions are infectious agents responsible for transmissible spongiform encephalopathies (TSEs) consisting of a pathogenic conformer (PrP^Sc) of the normal cellular prion protein (PrP^C). PrP^Sc templates the conformational conversion of PrP^C, generating additional disease-associated PrP^Sc molecules. This isoform displays biochemical properties distinct from those of PrP^C, notably a strong tendency to aggregate and partial resistance to proteinase K digestion under controlled conditions [1]. PrP^Sc accumulation underlies the pathogenesis of naturally occurring TSEs identified in several animal species including sheep, goats, cattle, cervids and camels. Some of these species enter the human food chain, raising concerns regarding potential zoonotic transmission. Zoonotic transmission was demonstrated in the case of classical bovine spongiform encephalopathy (BSE), which gave rise to variant Creutzfeldt–Jakob disease (vCJD) in humans [2].

Human TSEs, or prion diseases, are classified into sporadic, genetic and acquired forms. The most common is sporadic Creutzfeldt–Jakob disease (sCJD), with an annual incidence of ~1.4 cases per million population [3]. Genetic prion diseases result from mutations in the prion protein gene (PRNP) and include Gerstmann–Sträussler–Scheinker disease, fatal familial insomnia and genetic Creutzfeldt–Jakob disease. Acquired forms arise from exposure to prion-contaminated materials, such as ingestion of BSE-infected food products leading to vCJD, or via medical procedures involving contaminated grafts or human growth hormone preparations [3]. More than 230 cases of vCJD have been reported, most of them in the UK, following population-level exposure to BSE [4]. Concerns persist regarding the possibility of secondary human-to-human transmission and the emergence of a second wave of vCJD cases due to prolonged incubation periods [5].

Neuropathological examination of prion diseases is typically performed on brain tissues that have been formalin-fixed (FF) and treated with formic acid. These analyses reveal characteristic histopathological features. In sCJD, pathology includes spongiform change, neuronal loss, astrocytic gliosis and PrP^Sc deposits, with variable severity and regional distribution, and encompasses several disease subtypes defined by the PRNP codon 129 genotype and the associated clinicopathological phenotype [6]. Variant CJD is distinguished by widespread spongiform degeneration, numerous florid plaques and extensive PrP^Sc deposition [7]. The occupational risks associated with handling human prion-infected materials remain insufficiently defined, and some studies suggest potential hazards that warrant further clarification [8, 9].

Formic acid treatment is known to reduce prion infectivity [10]. Although prion inactivation has been assessed in both in vivo and in vitro systems [11, 12], the in vivo reduction in infectivity of human prions following formic acid treatment has not been evaluated using the most sensitive transgenic mouse models available. This gap in knowledge is particularly relevant because formic acid treatment is routinely used in the processing of prion-infected tissues for diagnostic purposes.

In this study, we assessed the reduction in infectivity of type 1 sCJD and vCJD prions after formic acid treatment under conditions similar to those routinely used for the histopathological processing of prion-infected samples. To achieve this, we used highly sensitive transgenic mouse models overexpressing human or bovine prion protein, respectively [13, 14].

Methods

Transmission experiments

Hu-Tg340 mice (overexpressing human PrP M129) [13] and Bo-Tg110 mice (overexpressing bovine PrP) [14] were used. Both lines are homozygous for the corresponding PRNP transgene and express PrP at levels fourfold (Hu-Tg340) or eightfold (Bo-Tg110) higher than endogenous expression on a PrP-knockout background. Hu-Tg340 transgenic mice were challenged with type 1 sCJD, whereas Bo-Tg110 mice were challenged with vCJD. The sCJD inoculum consisted of brain from Hu-Tg340 mice [15] previously inoculated with a type 1 sCJD M129M-infected case (0.08.02523_001). The vCJD inoculum was prepared from the brains of Hu-Tg340 mice challenged with a vCJD M129M-infected case (BC1458) [15]. Each mouse brain was divided into three equal-sized pieces without further sectioning and subjected to three different treatments to generate the experimental inocula, selected to reflect the main conditions encountered in diagnostic neuropathology workflows:

Untreated inoculum: homogenized at 10% (w/v) in 5% glucose and frozen.

FF inoculum: processed following routine histological procedures. Each sample was stored in 15 ml of 10% neutral buffered formalin for 3–7 days before execution of the fixation protocol at 37 °C using Leica Biosystems TP1020 tissue processor™: 10% neutral buffered formalin (two steps, 60 min each), 70% ethanol (two steps, 60 min each), 96% ethanol (two steps, 60 min each), absolute ethanol (two steps, 60 min each), xylene (two steps, 60 min each) and finally 10% neutral buffered formalin [16]. The solvent volume in the processor reservoir at each step was ~1400 ml. After the last step, each sample was washed three times in 5% glucose. Finally, the brain was homogenized at 10% (w/v) and frozen.

Formic acid inoculum (FF+FAcid inoculum): processed following routine histological procedures (including 10% neutral buffered formalin) and further incubated for 1 h in 98% formic acid [10, 17], washed three times, homogenized at 10% (w/v) and frozen.

Mice were inoculated intracerebrally with 20 µl of 10% brain homogenate and monitored for clinical signs. Brains were collected at endpoint for PrP^res detection by Western blotting. Incubation time was expressed as mean days post-inoculation among PrP^res-positive mice. Attack rate was defined as the proportion of PrP^res-positive animals.

Western blotting analysis

Brain tissue was homogenized in 5% glucose in distilled water in grinding tubes (Bio-Rad) and adjusted to 10% (w/v) by using a TeSeE™ Precess 48™ homogenizer (Bio-Rad) following the manufacturer’s instructions. To determine the presence of PrP^res in transgenic mouse brains, 100 µl of 10% brain homogenate were analysed by Western Blot as previously described [15]. For immunoblotting, membranes were incubated with Sha31 PrP monoclonal antibody (epitope 145-YEDRYYRE-152 of the human-PrP sequence; RRID: AB_3714424) [18] at a final concentration of 1 µg ml⁻¹. Immunocomplexes were detected with horseradish peroxidase-conjugated anti-mouse IgG (Amersham Pharmacia Biotech) after 1 h incubation, and blots were developed using ECL Select chemiluminescent substrate (GE Healthcare Amersham Biosciences). Images were captured with a ChemiDoc XRS+ system and processed using Image Lab 6.1 software.

Calculation of infectivity

The equivalent infectious titre of the inocula administered to the animals was estimated by linear interpolation between the log₁₀ of the titre and the incubation time, following the standard approach used for titre–incubation curves in murine prion models. ID₅₀/g values are expressed on the same logarithmic scale (10ˣ) as in [19–21] to allow direct comparison. Statistical significance was analysed using the log-rank test implemented in R software (survival and survminer packages).

Results

In this work, sCJD and vCJD inocula were bioassayed in Hu-Tg340 and Bo-Tg110 mice, respectively. These transgenic lines overexpress human or bovine PrP and are therefore considered highly sensitive in vivo models for detecting infectivity from each corresponding prion strain. Three different inoculum conditions were bioassayed: (i) the untreated inoculum, which reflects the native infectivity present in the prion-infected brain; (ii) the FF inoculum, which models routine fixation procedures that do not include steps aimed at reducing infectivity; and (iii) the FF+FAcid inoculum, which reproduces histological workflows incorporating a formic acid treatment step intended to decrease infectivity. Comparison of the untreated inoculum with the treated inocula allows assessment of the infectivity-reducing potential associated with each treatment. Table 1 summarizes the incubation periods observed in animals inoculated with isolates subjected to the different treatments. Mice inoculated with the untreated inocula developed clinical signs after short incubation periods and showed full attack rates, as expected for this type of material. A slight prolongation in the onset of clinical signs, while maintaining full attack rates, was observed in animals inoculated with the FF inocula. In contrast, incubation periods were markedly extended in animals inoculated with the FF+FAcid inocula. The biochemical characteristics of the propagated prion (Western blot profile) were indistinguishable regardless of the treatment applied (Fig. 1). In the case of the sCJD inoculum, this prolongation was accompanied by a reduced attack rate: two animals remained negative for clinical signs of prion disease and showed no PrP^res accumulation in the brain more than 580 days post-inoculation (Fig. 1). Estimation of the infectivity in the assayed inocula, as shown in Fig. 2, revealed a substantial reduction in infectious titre in the FF+FAcid inocula for both prion strains. Because these estimates are derived from small experimental groups (in some cases only four animals), the limited cohort size may introduce some uncertainty in the precise magnitude of the inferred reductions. Estimated infectivity reductions were on the order of 4.4 log₁₀ for the vCJD inoculum and five log₁₀ for the sCJD inoculum, indicating an evident loss of infectivity under these treatment conditions.

Table 1. Infectivity of vCJD and sCJD prions in transgenic mouse bioassays after different treatments.

Inoculum/mouse line	Inoculum treatment	Incubation period* ±SD† (n/n0)‡	Infectious titre§ ID₅₀/g	Infectivity loss due to the treatment¶	Statistical significance of the treatment††
vCJD/Bo-Tg110	Untreated	281±40 (6/6)	105.659; 4.55×105	N/A**	N/A
	FF	355±26 (4/4)	104.209; 1.62×104	≈1.4 log; 96.44%	* (0.0267)
	FF+formic acid	508±66 (6/6)	101.211; 16.25	≈4.4 log; 99.9964%	** (0.0016)
sCJD MM1/Hu-Tg340	Untreated	183±9 (4/4)	106.748; 5.60×106	N/A	N/A
	FF	214±9 (7/7)	105.615; 4.12×105	≈one log; 92.64%	*** (0.00044)
	FF+formic acid	320±87 (8/10)	101.738; 54.7	≈five log; 99.999%	*** (4.9e-05)

*Mean incubation period (days) of PrPres-positive animals.

†SD: standard deviation.

‡n/n₀: attack rate where ‘n’ means the number of PrPres-positive animals by Western blotting and ‘n₀’ means the total number of inoculated animals.

§Infectious titres were estimated by the Spearman–Karber method interpolating the data from [21].

¶Infectivity loss when compared to the untreated inoculum expressed as log₁₀ and percentage.

**N/A: not applicable.

††Statistical significance assessed by log-rank test: *P-value<0.05; **P-value<0.01; ***P-value<0.001.

Fig. 1. Immunoblot of brain proteinase K–resistant prion protein (PrP^res) from (a) Hu-Tg340 mice inoculated with sCJD ‘untreated inoculum’ (lane 1) or sCJD ‘formic acid’ inoculum (lanes 2–11). Lanes 10 and 11 correspond to animals that did not develop clinical disease. (b) Bo-Tg110 mice inoculated with vCJD ‘untreated inoculum’ (lane 1) or vCJD ‘formic acid’ inoculum (lanes 2–7). PrP^res was detected using the Sha31 monoclonal antibody. MW, molecular weight in kilodaltons.

Fig. 2. Infectivity titration according to Douet et al. [21]. Linear regression was derived from incubation-period changes after serial dilution, and infectivity loss was estimated by interpolation in the corresponding regression lines for vCJD (Bo-Tg110, left) and sCJD (Hu-Tg340, right), using the equations shown in each panel.

Discussion

Infectivity was assessed using transgenic mouse models overexpressing either Met129 human PrP or bovine PrP, which are considered among the most sensitive in vivo systems currently available for detecting these prion strains [21]. Because in these bioassays longer incubation periods directly reflect lower infectious titres, extrapolation of incubation-time differences relative to those previously reported for similar inocula titrated in the same models indicates that incubation in formic acid under conditions closely resembling those routinely applied (with minor procedural modifications) for processing prion-infected tissues results in a substantial reduction in infectivity [10, 16, 17]. Although these findings clearly point to a marked loss of infectivity, it should be noted that the reductions reported here are indirect estimates derived from incubation-time interpolation rather than from direct endpoint titration. While this approach is widely used and accepted in prion bioassays, the limited cohort size and the inherent variability associated with incubation-time–based estimations introduce some uncertainty in the precise magnitude of the inferred titre reductions. In comparison, the FF inocula showed only a detectable but limited decrease in infectious titre, consistent with earlier observations [10]. This modest reduction is likely attributable to the combined effect of washing steps and the intrinsic inactivation capacity of formalin, which can chemically modify proteins while preserving their overall structural organization [22]. By contrast, formic acid, acting as a strong protein solvent, is superior to common organic solvents (e.g. glycerol, DMSO or trifluoroacetic acid), solubilizing the polypeptide chain through protonation, disruption of hydrogen-bond networks and interference with hydrophobic interactions, ultimately destabilizing β-sheet architecture characteristic of the PrP^Sc fibrils [23–25]. Significantly, differences may exist between human prions derived directly from human brain tissue and those propagated in humanized transgenic mouse lines [26], which may not fully reproduce all strain-specific biological properties. Moreover, although two distinct human prion strains (sCJD and vCJD) were examined here and yielded broadly similar results, other prion strains may display different susceptibilities to formic acid inactivation procedures, and such potential variability should be considered when extrapolating these findings.

The estimated reduction in infectivity, although not complete, is biologically significant. An estimated decrease of 4.4–5 log₁₀ units corresponds to an approximate 10,000- to 100,000-fold loss of infectious titre, representing the elimination of more than 99.99% of the infectious agent. This percentage is intended solely to convey the magnitude of the reduction and does not imply full inactivation, as residual infectivity remains detectable due to the high sensitivity of the transgenic mouse models employed, consistent with previous titration experiments. These findings have practical implications for the handling of prion-infected tissues: when samples are treated with formic acid under conditions comparable to those described in this study, most of the infectivity is effectively eliminated, and this should be taken into account in routine laboratory workflows.

Overall, these results provide quantitative in vivo evidence supporting the effectiveness of formic acid treatment, under routine histopathological conditions, as a critical biosafety measure in diagnostic neuropathology, leading to a major reduction in the infectivity of prions.

Abbreviations

BSE

bovine spongiform encephalopathy

FAcid

Formic acid

FF

formalin-fixed

PrP

prion protein

sCJD

sporadic Creutzfeldt–Jakob disease

TSE

transmissible spongiform encephalopathies

vCJD

variant Creutzfeldt–Jakob disease