Rapid assessment of bovine spongiform encephalopathy prion inactivation by heat treatment in yellow grease produced in the industrial manufacturing process of meat and bone meals

Background Prions, infectious agents associated with transmissible spongiform encephalopathy, are primarily composed of the misfolded and pathogenic form (PrPSc) of the host-encoded prion protein. Because PrPSc retains infectivity after undergoing routine sterilizing processes, the cause of bovine spongiform encephalopathy (BSE) outbreaks are suspected to be feeding cattle meat and bone meals (MBMs) contaminated with the prion. To assess the validity of prion inactivation by heat treatment in yellow grease, which is produced in the industrial manufacturing process of MBMs, we pooled, homogenized, and heat treated the spinal cords of BSE-infected cows under various experimental conditions. Results Prion inactivation was analyzed quantitatively in terms of the infectivity and PrPSc of the treated samples. Following treatment at 140°C for 1 h, infectivity was reduced to 1/35 of that of the untreated samples. Treatment at 180°C for 3 h was required to reduce infectivity. However, PrPSc was detected in all heat-treated samples by using the protein misfolding cyclic amplification (PMCA) technique, which amplifies PrPScin vitro. Quantitative analysis of the inactivation efficiency of BSE PrPSc was possible with the introduction of the PMCA50, which is the dilution ratio of 10% homogenate needed to yield 50% positivity for PrPSc in amplified samples. Conclusions Log PMCA50 exhibited a strong linear correlation with the transmission rate in the bioassay; infectivity was no longer detected when the log PMCA50 of the inoculated sample was reduced to 1.75. The quantitative PMCA assay may be useful for safety evaluation for recycling and effective utilization of MBMs as an organic resource.


Background
Transmissible spongiform encephalopathies (TSEs), including scrapie in sheep and goats, chronic wasting disease (CWD) in deer and elk, bovine spongiform encephalopathy (BSE) in cattle, and Creutzfeldt-Jakob disease (CJD) in humans, are infectious and fatal neurodegenerative diseases [1]. Proteinaceous infectious agents called prions are thought to be responsible for TSEs, which are characterized by the accumulation of the pathogenic form of prion protein (PrP Sc ) in the nervous tissues of infected subjects [2,3]. PrP Sc is a conformational isoform of the normal cellular prion protein (PrP C ), which is rich in beta-sheet structures, insoluble in mild detergents, and resistant to protease digestion [4,5].
Because prions retain infectivity after undergoing routine sterilization processes [6], contaminated meat and bone meals (MBMs) are suspected to be the source of BSE infection [7,8]. MBMs are manufactured through a multi-step process involving the crushing of carcasses in a pre-breaker, heating at 120°C-140°C in yellow grease (lower-quality grades of tallow) in a cooker, and degreasing from solid material by an oil separator. To determine BSE prion inactivation during the manufacturing process of MBMs, industrial processes were replicated on a pilot scale by using BSE-infected brains, and the infectivity of processed materials in each step was investigated in detail [9]. However, in reality, processing conditions for MBMs differ among rendering houses producing commercial MBMs. Since the efficiency of prion inactivation could be influenced by various factors such as treatment temperature, time, steam pressure in the cooker, size, water and fat contents of carcasses [10][11][12][13], it is difficult to identify the risks attributable to specific processing conditions. Furthermore, PrP Sc retained in the manufacturing process of MBMs remains to be elucidated.
The governments of many countries prohibited the feeding of bovine MBMs following the feeding ban on MBMs in the United Kingdom. A prion detection method with high sensitivity and high accuracy must be developed so that MBMs can be used safely in the future. In addition, BSE prion is more resistant to physical and chemical treatments than are scrapie and CJD prions [14]. Therefore, experiments using BSE-infected materials are essential for the assessment of BSE prion inactivation as they can be considered a worst case among prions. In recent years, it has become possible to perform in vitro amplification of PrP Sc derived from various animals [15][16][17][18][19][20][21] by using protein misfolding cyclic amplification (PMCA) [22]. We developed an ultrasensitive method for BSE PrP Sc detection using potassium dextran sulfate (DSP) [20]. The PMCA technique can also be used to quantitatively assess scrapie PrP Sc [23][24][25], and our PMCA method can be applied as an effective test for the assessment of prion inactivation by monitoring residual BSE PrP Sc [26]. In the present study, we investigated efficiency of BSE prion inactivation following heat treatment in yellow grease by bioassay and quantitative PMCA.

Infectivity of heat-treated homogenates
Long-term follow-up confirmed infectivity in the mice intracerebrally inoculated with up to a 10 -5 dilution of the 10% homogenate of the pooled spinal cords (Table 1). PrP Sc accumulation was confirmed in the brains of the diseased mice by western blotting and histopathological analysis (data not shown). The infectious titer of the homogenate was estimated to be 10 6.7 LD 50 per gram. A strong linear correlation (r = 0.99) between the incubation times and dilution ratios of the inoculated homogenate was observed in mice inoculated with up to a 10 -3 dilution. Some mice inoculated with 10 -4 and 10 -5 diluted samples developed the disease after similar prolonged survival times (735 or 736 days). In the extreme dilution range, lower rate of transmission and prolonged incubation time are generally observed in the mice intracerebrally inoculated with prion-infected brain homogenates. Since PrP Sc tends to aggregate, these phenomena may be due to the near-absence of PrP Sc which would have been almost completely diluted out. Table 2 shows the effect of various heat treatments in yellow grease on the BSE-infected spinal cord homogenates. All mice inoculated with samples treated at 140°C for 1 h died after an average of 304 days. The infectivity was reduced to approximately 1/35 (log reduction = 1.54) following the heat treatment. When the samples subjected to temperatures above 140°C were used, 100% (180°C for 1 h) and 67% (160°C for 1 h) of the mice developed the disease after prolonged average survival times. Regarding the treatments for 3 h, infectivity was still detected in some mice inoculated with the samples treated at 140°C or 160°C. Because the incubation times of these diseased mice were beyond the range of application of the regression line obtained using the titrated BSE-infected homogenates, the log reduction of infectivity in each sample was estimated to be more than 3.0. Meanwhile, mice inoculated with samples treated at 180°C for 3 h did not exhibit disease onset 790 days after inoculation.
PrP Sc detection by PMCA Figure 1a illustrates the results of the amplification of the samples subjected to the grease-heating method. No PrP Sc signals were detected in the heat-treated samples by western blotting before amplification (data not shown). After one round of amplification, PrP Sc signals were detected in the samples treated at 140°C-180°C for 1 h and at 140°C for 3 h. PrP Sc signals were also detected in both duplicate samples treated at 160°C and 180°C for 3 h after two or three rounds of amplification. In the samples treated at 180°C for 3 h, trace amounts of PrP Sc remained after the treatment, although infectivity was not detected in the bioassay. Figure 1b shows the results of the amplification of each diluted sample of untreated BSE-infected spinal cord homogenate. PrP Sc present in 10 -7 dilution of the infected homogenate was detected in all tubes after three rounds of amplification. PrP Sc signals were detected in three of the 10 -8 and one of the 10 -9 dilutions after three rounds of amplification. However, no additional tubes became positive for PrP Sc in these dilutions after four rounds of amplification. No signals were detected in the more extreme dilution ranges even after four rounds of amplification. Thus, the PMCA 50 of the 10% homogenate was calculated to be 10 8.5 units on the basis of the results obtained from the fourth round of amplification.

Quantitative analysis of PrP Sc
To evaluate the PrP Sc inactivation efficiency of each heat treatment, we estimated the PMCA 50 from the results obtained at the fourth round of amplification of serial 10fold dilutions of heat-treated samples. Serial PMCA was sufficiently sensitive to detect PrP Sc in these diluted samples ( Figure 2). The log reduction of PMCA 50 values of the heat-treated samples are shown in Table 2. Regarding the treatments for 1 h, PrP Sc inactivation appeared to be most efficient in the samples treated at 160°C. This finding is concordant with the observations of partial transmission of infectivity (67%) in the mice inoculated with this sample and prolonged incubation times of the diseased mice. Log PMCA 50 decreased with extended heat treatment time: although 180°C for 3 h was the most effective treatment, it was unable to completely inactivate the proportion of PrP Sc that is amplifiable by serial PMCA. Figure 3 shows the relationships between the transmission rate in the bioassay and the log PMCA 50 values of the inoculated samples. A strong linear correlation (r = 0.97) was observed between the log PMCA 50 values and transmission rate. When the log PMCA 50 exceeded 5.25, the transmission rate in the bioassay reached 100% as observed in the mice inoculated with samples treated at 140°C or 180°h for 1 h. Infectivity was not detected in the mice when the log PMCA 50 of the inoculated sample was reduced to 1.75.

Discussion
In this study, BSE prion inactivation was analyzed quantitatively in terms of infectivity and the PrP Sc contents of the samples after heat treatment in yellow grease. Following treatment at 140°C for 1 h, which is the heat treatment condition generally used for carcasses in rendering houses in Japan, the infectivity of the BSEinfected spinal cord homogenate was reduced to at most 1/35 of that of the untreated control samples; furthermore, PrP Sc retained its capability for in vitro propagation. Because carcasses are usually heat treated in closed cookers, prions are affected by the steam pressure from the water contained in the carcass. If a sufficient amount of water is present in the carcass, prion inactivation may proceed more efficiently in the cooker than under atmospheric pressure. However, some degree of BSE infectivity was still detected after autoclaving at 133°C in spiked raw materials with high infectivity levels [26]. Furthermore, the precise effects of high-pressure steam on carcasses submerged in yellow grease are not known. Therefore, high-risk materials such as brains and spinal cord should be excluded from the rendering process for effective inactivation of BSE prion.  We previously examined residual infectivity and PrP Sc after heat treatment of scrapie-infected hamster brains under various experimental conditions [27]. The PMCA results were concordant with bioassay results. However, BSE PrP Sc was detected in the samples treated at 180°C for 3 h, although infectivity was not detected in the bioassay. There are several possible explanations for this discrepancy between infectivity and PrP Sc occurrence. For example, BSE PrP Sc might contain various forms of PrP Sc with different amplification properties and infectivity, and a PMCA-compatible form of PrP Sc with low or no infectivity might predominate after heat treatment and be maintained over other forms throughout the amplification process. However, in the present study, the log PMCA 50 values were strongly correlated with the transmission rate in the bioassay (Figure 3), suggesting that such PMCAcompatible but less-infectious PrP Sc was not selectively amplified in vitro.
In our previous paper, we demonstrated our amplification system was highly sensitive and accurate, and no spontaneous generation of PrP Sc was observed in the amplification of various kind of samples derived from  uninfected animals [20]. Determination of PMCA 50 based on quadruplicate amplification was also done in our previous study [26], and we confirmed that similar PMCA 50 values (around 10 11 per gram) were obtained in two independent studies. In the present study, the PMCA 50 of the BSE-infected spinal cords was estimated to be 10 11.6 per gram, which is approximately 80,000fold greater than the corresponding intracerebral LD 50 per gram (10 6.7 ) determined by the bioassay. The PMCA 50 /LD 50 per gram of BSE prion was considerably higher than those of scrapie prion strains (160-4000 fold) [25]. If this ratio reflects the number of PrP Sc particles that compose an infectious unit of prions, more PrP Sc particles might participate in an infectious unit of BSE prions; moreover, such a large mass of PrP Sc particles might be processed into several smaller ones with lower infectivity in vivo.
Alternatively, PrP Sc accumulation might proceed in animals inoculated with PrP Sc when the PrP Sc concentration is below a specific cut-off, but the animals might not develop the disease within their lifetimes. Actually, clinically asymptomatic infections are known as the subclinical infection stage [28][29][30]. In the present study, we examined PrP Sc in brains of asymptomatic mice inoculated with titrated BSE-infected homogenate, and PrP Sc was found at various levels in four of five mice inoculated with 10 -6 dilution of the infected homogenate ( Figure 4). Therefore, pathogenicity might be detected by serial transmission in animals as in the case of serial PMCA. If so, the detection sensitivity of the bioassay used in the present study may not be sufficiently high for proper safety evaluation, because ecycling of BSEinfected bovine tissues possibly augments the concentration of PrP Sc in commercial MBMs if the carcasses contain infinitesimal amounts of prion.
Another aspect of heat treatment in yellow grease is that higher-temperature treatments do not necessarily inactivate BSE prion more effectively. In the case of heat treatment for 1 h, the results of both the bioassay and PMCA indicate that BSE prion inactivation proceeded more effectively with treatment at 160°C rather than at 180°C. Samples treated at 180°C were dark brown, suggesting that the surface was scorched during treatment. In such high-temperature conditions, thermal conduction may be inhibited by scorching of the sample periphery, consequently requiring longer treatment time to reach thermal equilibrium in the sample. Extension of the treatment time to 3 h was actually necessary for the loss of infectivity. However, further studies are needed to confirm the above possibility.

Conclusions
In this study, we demonstrated that heat treatment at 180°C for 3 h is required for the loss of infectivity of BSE prion in grease heating in our experimental conditions. Furthermore, BSE PrP Sc retains amplification ability even after such a treatment. The inactivation efficiency of BSE PrP Sc could be quantitatively analyzed with the introduction of the PMCA 50 , which is strongly correlated with the transmission rate in the bioassay. The serial PMCA technique is more practical and less time consuming than bioassays, and may be applicable for monitoring residual  Detection of PrP Sc in brains of asymptomatic mice. A 10% brain homogenate from five (#1-#5) of six asymptomatic mice inoculated with 10 -6 dilution of infected homogenate (Table 1) was prepared, and amplified by serial PMCA. Quadruplicate samples were analyzed after each round (R1-R4) of amplification by western blotting after digestion with proteinase K. One (#1 and #3), two (#5) and four (#2) of the quadruplicate samples were found to be positive for PrP Sc after three or four rounds of amplification. No PrP Sc signal was detected in #4 mouse. The eight lanes labeled "N" are samples in which only the PrP C substrate was treated in the same manner. Horizontal lines indicate the positions of molecular-weight markers corresponding to 37, 25, 20, and 15 kDa. nt: not tested.
PrP Sc in the other steps of the manufacturing of MBMs and useful for safety evaluation for recycling and effective utilization of MBMs as an organic resource.

Experimental heat treatment procedure
All animal experiments were approved by the Animal Care and Use Committee of the National Institute of Animal Health (approval IDs: 450 and 08-008) in accordance with the Guidelines for Animal Transmissible Spongiform Encephalopathy Experiments of the Ministry of Agriculture, Forestry, and Fisheries of Japan. Spinal cords were obtained from four cows experimentally inoculated with BSE at the terminal stage of the disease. The infected materials were pooled and homogenized using a blender. Pure homogenate (0.5 g) was placed on a strip of aluminum foil (2 cm × 2 cm) and stored at -80°C until further use. For use, the homogenate with the aluminum foil was thawed at room temperature and then immersed in 15 mL yellow grease preheated to 140°C, 160°C, or 180°C in a ceramic crucible by using an electric heating device (ND-M11, Nissin Rika, Tokyo, Japan). The yellow grease used was obtained from a rendering house in Japan. The crucible was covered, and a thermosensor was inserted through a hole in the cover to monitor the temperature of the yellow grease. The yellow grease was, then, kept for 1 or 3 h at the desired temperature. The homogenate sample firmly adhered to the surface of aluminum foil and was not broken into pieces during the heat treatment. After the treatment, the homogenate with the aluminum foil was removed from the yellow grease with tweezers and placed on a paper towel for absorption of the excess yellow grease. The weights of the homogenates were reduced to 60-70% of their original weights. The resultant materials were thoroughly crushed with a mortar, and suspended in PBS at 10% (w/v). Insoluble materials were separated by brief centrifugation, and aqueous fraction was stored at -80°C until further use.

Bioassay
Infectivity titer using transgenic mice overexpressing bovine PrP C is generally 100-1000 times higher than that using cows. Therefore, more accurate estimation of BSE infectivity is able to be conducted by using such mice. The heat-treated samples were injected intracerebrally into six Tg(BoPrP)4092HOZ/Prnp 0/0 (TgBoPrP) transgenic mice (20 μL per mouse) overexpressing bovine PrP C [31]. To determine the infectivity titer, serial 10fold dilutions of the 10% homogenate of the untreated spinal cords were prepared in PBS and injected intracerebrally into five to seven TgBoPrP mice (20 μL per mouse). After inoculation, the mice were evaluated daily for signs of infection. The lethal dose (LD 50 ) was determined according to the 50% endpoint calculation method. Mean incubation times of the diseased mice were analyzed by one-way ANOVA and Tukey's multiple comparison test.

PMCA
Bovine PrP Sc was amplified as described previously [20]. Briefly, the brains of TgBoPrP transgenic mice and PrP knockout (PrP 0/0 ) mice were homogenized separately in PBS containing 1% Triton X-100 and 4 m mol L -1 EDTA. After centrifugation at 4500 × g for 5 min, the supernatants were mixed in PrP 0/0 /TgBoPrP (5:1). A mixture containing 0.5% DSP was used as the PrP C substrate for PMCA.
The 10% homogenates of heat-treated samples were mixed at 1:9 with the PrP C substrate (total volume, 100 μL) in electron beam-irradiated polystyrene tubes. Amplification was performed in duplicate with a fully automatic cross-ultrasonic protein-activating apparatus (Elestein 070-CPR, Elekon Science, Chiba, Japan), which has a capacity to generate high ultrasonic power (700 W). PMCA amplification was performed by 40 cycles of sonication (3-s pulse oscillations repeated 5 times at 1-s intervals), followed by incubation at 37°C for 1 h with agitation. For serial PMCA, 1:5 dilution of the PMCA product and subsequent amplification was repeated twice.
To evaluate the inactivation efficiency of BSE PrP Sc by heat treatment, the PMCA 50 , which is the dilution ratio of the 10% homogenate needed to yield 50% PrP Sc positivity for amplified samples, was determined. Serial 10-fold dilutions of the 10% homogenate of the heat-treated and untreated samples were prepared and mixed 1:9 with the PrP C substrate (total volume, 80 μL) and amplified in electron beam-irradiated eight-strip polystyrene tubes (076-96, Elekon Science). Amplification was performed in quadruplicate using 40 cycles of sonication (pulse oscillation for 5 s, repeated 5 times at 1-s intervals), followed by incubation at 37°C for 1 h with agitation. For serial PMCA, 1:5 dilution of the amplified product and subsequent amplification was repeated 3 times. The PMCA 50 was estimated from the results of the fourth round of amplification by using the 50% endpoint calculation method.

Western blotting
The amplified samples (10 μL) were mixed with 10 μL proteinase K solution (100 μg mL -1 ) and incubated at 37°C for 1 h. The digested samples were mixed with 20 μL 2× SDS sample buffer and incubated at 100°C for 5 min. The samples were separated by SDS-PAGE and transferred onto a polyvinylidene fluoride membrane (Millipore, Bedford, MA). After the membrane was blocked, it was incubated for 30 min with a horseradish peroxidase (HRP)-conjugated T2 monoclonal antibody [32]. After washing, the blotted membrane was developed using the Luminata Forte Western HRP Substrate (Millipore) according to the manufacturer's instructions. Chemiluminescence signals were analyzed with a Light Capture System (Atto, Tokyo Japan).

Histopathological analysis
The left hemispheres of the brains were fixed in 10% buffered formalin for neuropathological analysis. Coronal brain sections were immersed in 98% formic acid to reduce infectivity and embedded in paraffin wax. Sections (4 μm thick) were cut and stained with hematoxylin and eosin, and analyzed immunehistochemically as described previously [20].