Real-Time Detection of Central Nervous System Tissues on Bovine Carcasses Using Fluorescence Spectroscopy

Bovine spongiform encephalopathy (BSE) is a fatal, neurodegenerative transmissible spongiform encephalopathy (TSE) in cattle that is thought to be the cause of variant Creutzfeldt-Jakob disease (vCJD) in humans.1 The oral route of infection is considered to be the most likely mode of transmission of BSE to humans.2,3 In this context, removal of specified risk materials (SRM), for example, brain and spinal cord, is of fundamental importance to protect consumers from BSE infection.4,5 Brain and spinal cord from BSE infected cattle have been shown to contain the highest infectivity titer of the causative agent of BSE, the abnormal prion protein PrPBSE.4 Consequently, most countries have banned bovine central nervous system (CNS) tissues from meat products.

The real-time detection and removal of CNS tissues from the human food chain have attracted considerable attention in the scientific community, but most of the currently available methods, namely, immunochemical detection and enzyme-linked immunosorbent assays (ELISAs),6,7GC-MS,8 Western blot,9,10 immunohistochemical methods,4,11 and real-time polymerase chain reaction (PCR) assays,12,13 are mainly restricted to the laboratory and hence are unsuitable for real-time detection of CNS tissues on carcasses during slaughter.

Fluorescence spectroscopy has wide application in various fields primarily due to its very high sensitivity and specificity. The authors previously demonstrated the feasibility of detecting fecal contamination on the surface of meat during slaughter by exploiting the fluorescence of chlorophyll degradation products.14–16 Based on previous studies,14–16 the authors aim to use the fluorescence of lipofuscin for real-time detection of CNS tissues during slaughter. Lipofuscin is a heterogeneous, high-molecular-weight fluorescent material that has been shown to be enriched in high concentrations in neuronal tissues.17 Its exact chemical composition is controversial.17 It is believed to result from oxidative stress of many biomolecules, particularly polyunsaturated fatty acids.

This article summarizes and reviews progress toward the goal of developing a method for CNS tissue detection during slaughter. The authors have investigated spectral shapes and intensities of the two most important SRM tissues, i.e., brain and spinal cord, and compared spectra with those from various non-CNS tissues. To mimic slaughterhouse conditions, they further addressed the detection sensitivity of bovine spinal cord in the presence of three non-CNS tissues, namely, skeletal muscle, fat, and vertebrae.

Materials and methods

A detailed description of the collection of bovine tissue samples and experimental procedure can be found in Refs. 18 and 19. Briefly, bovine CNS and non-CNS tissue samples were collected from healthy cattle with a variation in sex, breed, and age. Brain and spinal cord were investigated as CNS tissues. The non-CNS  tissues were adrenal gland, bone (vertebrae), bone marrow, dorsal root ganglia, fat, heart, kidney, liver, lung, lymph node (e.g., iliofemoral lymph node), peripheral nerve (e.g., sciatic nerve), skeletal muscle, and spleen. Solid tissue samples were placed on a microscopic slide and fluorescence spectra were collected on a SPEX Fluoromax-2 spectrofluorometer (ISA Jobin-Yvon, Edison, NJ) in a front-faced orientation with an excitation wavelength of 470 nm and collecting the emission at wavelengths greater than 505 nm. The detection sensitivity of spinal cord was assessed in the presence of skeletal muscle, abdominal fat, and vertebrae of cattle as background tissues. In order to examine the sensitivity of detection, a small piece of spinal cord was placed in or on the skeletal muscle and fat, which were chopped into small pieces, whereas vertebrae were kept as small pieces. Each background tissue was spread on a microscopic slide in such a way that it covered the entire area exposed to the exciting light; the size of the spinal cord was adjusted to represent 5 and 10% fractions of the illuminated area.

Results and discussion

BSE is a zoonotic disease causally associated with vCJD in humans. The incidence of BSE in various parts of the world had a severe impact on livestock and the meat trade. Even 20 years after first identification of the disease, the United Kingdom estimated that costs for 2005 and 2006 for its BSE control measures were approximately £265 million, including cohort culls and offspring cull compensation. Additional administrative costs for inspection and enforcement were estimated in 1999–2000 to be £64 million.20 In the U.S., after the confirmation of only one BSE case, beef exports decreased by over 80% in 2004 and even now persist below pre-2004 levels.21 Estimated total U.S. beef industry losses arising from decreased beef and offal exports just during 2004 ranged from $3.2 billion to $4.7 billion.22 Resumption of beef exports depended on several restrictions made by the important trade partners, particularly Japan and South Korea. These restrictions included age specifications; removal of SRM (including brain and spinal cord); and, in the case of South Korea, import of deboned product only.21

The materials identified as bovine SRM are the brain, skull, eyes, trigeminal ganglia, spinal cord, vertebral column (excluding the vertebrae of the tail, the transverse processes, and the wings of the sacrum), dorsal root ganglia (DRG) from cattle 30 months of age and older, and the distal ileum of the small intestine and the tonsils from all cattle.23 The 30-month-and-older age classification for SRMs has been accepted by the U.S. and several other countries and is based on World Organization for Animal Health requirements.

In the present work, samples were excited at four wavelengths—350, 410, 470, and 520 nm. Very low emission signal intensity was observed while the samples were excited at 350 and 520 nm. Excitation at 410 nm produced broad, featureless spectra, while an excitation wavelength of 470 nm yielded structured fluorescence spectra and hence was considered a suitable excitation wavelength for all of the bovine tissue samples. This is consistent with previous observations.24