Select Committee on Agriculture Fifth Report

APPENDIX 29

Memorandum submitted by Hellen Fullerton PhD (L37)

  I enclose 20 copies of my submission to the Agriculture Committee for their inquiry into bovine and badger tuberculosis: "Can optimum trace element nutrition activate resistance to tuberculosis and provide a common solution for both cattle and badgers? "

  The nutritional issue was not raised in the Krebs Report but MAFF are now considering it (see MAFF Factsheet C10). I present the scientific arguments supporting the proposal that the immuno-resistance of the cattle to TB can be raised by increasing their trace element intake. Since there are widespread deficiencies of selenium in particular, but also of copper, zinc, cobalt and iodine in UK soils, the simplest and most effective method is to restore these elements to the soil of livestock farms, as has been done for years by R. Goodwin-Jones of Trace Element Services, with dramatic results on the health and performance of cattle and sheep.

  Farming and Livestock Concern have put the proposal to MAFF and also to the South West of England TB and Badger Study Group, chaired by Bill Jordan of Care for the Wild International. Care for the Wild offered to fund a small pilot research project to see whether by raising the nutrient status of the entire food chain, both cattle and badgers could be protected from TB, and if so to persuade MAFF that its advice to farmers, its funding and research should be diverted to that end.

  The accelerating spread of TB is approaching crisis. It is said to be comparable to the BSE disaster, but following on its heels likely to bring about the collapse of dairy and beef farming in vulnerable areas. Hence I believe the Agriculture Committee will wish to go beyond its advertised remit and examine an alternative cause and an alternative solution.

---

CAN OPTIMUM TRACE ELEMENT NUTRITION ACTIVATE RESISTANCE TO TUBERCULOSIS AND PROVIDE A COMMON SOLUTION FOR CATTLE AND BADGERS?

SUMMARY

  Trace elements, particularly selenium, copper, zinc, cobalt and iodine are essential to immune resistance. As co-factors of enzymes and in other roles they activate metabolism and contribute to white blood cell defences. Resistance mechanisms have been largely ignored since the introduction of antibiotics. It is proposed that resistance to M. Bovis could be enhanced by raising the trace element intake of cattle and also of badgers. This can be achieved if trace elements are restored to our depleted soils.

  Selenium deficiency is widespread and often severe in UK soils, as demonstrated by analysis of soils on livestock farms, by the frequency of symptoms of subclinical deficiency in cattle and sheep, and by the low intake of UK citizens from home-grown food and their falling blood levels. There is overwhelming evidence that MAFF's reference values for an adequate selenium intake are too low for cattle and sheep.The only valid test for an adequate intake is the response ie the correction of a malfunction be it infertility, depressed immune system etc. Certain malfunctions may require higher intakes. There is evidence that when selenium supplies are limited, it is distributed to those tissues most vital to the organism, leaving others to go short.

  The role of Se, Cu, Zn, Co and I deficiencies and their interactions in impairing immune resistance are reviewed. Also reviewed is a 1950-60s study of the effect of thyroid hormone (T4 and T3) injections in promoting TB resistance and in reversing an established disease. Nevertheless it is concluded that thyroid-mediated activation of the enzymes required by the macrophage to initiate and control the respiratory burst would be as effectively and more safely achieved by providing trace element levels necessary for thyroid function than by hormone injection.

  Research at the molecular level has found that T3 coordinates and augments glucose metabolism in the production of lipogenic enzymes for fatty acid synthesis, including glucose-6-phosphate dehydrogenase and 6-phosphogluconate dehydrogenase, pentose phosphate pathway enzymes that generate NADPH. It is proposed that the mechanism whereby T3 stimulates resistance against the TB bacilli depends on its stimulation of NADPH generation in biosynthesis of the phagolysosome membrane. In the presence of pathogen the NADPH reduces O2 to superoxide, with initiation of the respiratory burst.

  Other factors than trace element deficiency per se must contribute to suppression of immune resistance. A wet spring at turn-out has been identified as the major risk factor for herd breakdowns in the following year. It is proposed that the disease is transmitted from cattle to cattle via exhalation in the crowded conditions of winter housing and the stress of a cold wet spring at turn-out with low nutrient availability in the grass at that time, and other stressors such as carrying an autumn calf will depress the immune response of those challenged by M. bovis. It could be that the explanation for the fact that herds in SW England have always been especially vulnerable is that the humid conditions and mild winters are conducive to survival of the bacilli, not as has been assumed in the field but in the cubicle house. Transmission between farms and between districts is problematic: bought in cattle play a part and the contribution of imported cattle is unknown. The badger has been incriminated as a source of infection, but other forms of wildlife cannot be discounted, particularly birds with access to feed stores. The major question however is whether cattle can be made resistant to TB. If that is the case, transmission controversies can be settled later.

  Husbandry factors identified as likely to suppress cattle immune resistance include: Fertilisation with ammonium sulphate, where sulphur displaces soil selenium. Silage treatments with certain additives that cause nutrient losses. Failure to feed minerals. Bulls housed with cattle—bulls lose selenium at each ejaculation and are therefore both themselves more vulnerable than cows, and could be a serious reservoir of infection. Spreading of raw slurry with insufficient time given for microbial destruction of pathogens—cattle grazing such pastures face additional challenge. Stress of any sort including intensification, since pushing for high productivity is a stress on the cows' metabolism.

  It is proposed that badger susceptibility to TB is exacerbated when their immune systems are depressed: (a) by selenium deficiency—if soil selenium levels are low, so will be their dietary intake. Samples of badger fur were found severely low in selenium (b) by carrying an internal parasitic load. Vitamin B12 is important in the control of parasites. Badgers most at risk are thus those in areas of cobalt deficiency or low cobalt and zinc availability ie the granite areas of Devon and Cornwall and limestone areas of Somerset and Gloucestershire respectively.

  It is suggested that a retrospective survey of farms where trace element levels have been restored by Goodwin-Jones of Trace Element Services and their incidence of positive reactors compared to neighbouring farms will indicate whether or not similar trace element treatments could protect the nation's herds against TB.

1.  MOLECULAR MECHANISMS OF RESISTANCE

  The three major tubercle bacilli of the genus Mycobacterium are M.tuberculosis, M.bovis and M.avium. Unlike the many free living Mycobacterium that inhabit soil, these are intracellular pathogens that live and replicate within the phagocytic vacuoles of the host's alveolar (lung) macrophages, deriving from them their metabolic requirements and protected against many of the host's defence mechanisms, including the circulating B lymphocytes and their antibodies. Whether there is a resistant or pathogenic outcome depends on the complex interactions between mycobacterial proteins and the immunological response of the macrophage. Resistance depends on a cell-mediated immune response: infected macrophages are transported to lymphatic nodes and activated by T-cells, mainly T helper (TH1) cells to release lysosomal enzymes and the products of the respiratory burst against the intruders. The bacilli depend for their success on an ability to elude and withstand the killing mechanisms of the macrophages by suppressing their activation [1 rev].

  One of the most difficult problems in TB control is the presence of latent bacilli, possibly related to an upregulation that enables them to survive in an anaerobic environment (although long held to be a strict aerobe, it is now known the M.tuberculosis genome has genes important for both aerobic and anaerobic growth [1]). The host protects itself by encapsulating and walling off infected macrophages in granulomatous tissue. Here the inactive bacilli may lie low indefinitely, deriving nourishment from the cells unless or until they are reactivated by an impaired immune system to become pathogenic. Activated macrophages respond with an inflammatory reaction directing free radicals against the bacilli, a fraction of which may escape if the surrounding tissue is damaged or destroyed by the reaction, and exhaled as spores with potential for transmission. Nearly all TB lesions are caused not by the mycobacterium, but by the cell-mediated response. This suggests that if the host was provided with adequate antioxidants to control free radical production the damage to lung and other tissues might be avoided.

  Thirty to forty years ago resistance mechanisms against M.tuberculosis and pathogens in general were intensively investigated. M.Lurie researched resistance to TB, using the rabbit as a model [2]. In humans, the success of antibiotics suppressed interest in resistance mechanisms, and in cattle, TB outbreaks fell following a policy of culling all reactors.

  The roleof badgers is problematic. The discovery in Gloucestershire in 1971 of a badger infected with M.bovis laid on them the finger of suspicion. But a 25 year culling policy has not proved that they provide the reservoir of infection. We do not know whether, as many believe, badgers infect cattle via infected urine, - 300,000 bacilli/ml may be excreted by those with advanced infection [3]—or cattle infect badgers when they root around the cowpats in search of worms and other invertebrate components of their diet. Or whether cattle circulate it among themselves by exhalation. It should be noted that a TB epidemic is rife in the buffalo and antelopes of S. Africa's Kruger National Park, where cattle are the original reservoir. Transmission to wild life weakened by drought is thought to be from infected water holes or vegetation. It is then spread to lions feeding on infected prey [4].

  We propose that once again research should be focussed on resistance mechanisms. In medicine this is already happening as a result of the rapid advances in molecular biology and immunology [1], and the search for vaccines. A vaccine against M.bovis is said to be 10-15 years away. All vaccines against intracellular pathogens face similar difficulties, in that those evoking humoral responses have limited value and those evoking cell-mediated responses carry the risk of immno-pathology. Where risk is minimised and protection mediated by T-cell dependent antibody, vaccine designers face the confounding factors of variation in the chosen subunits of the organism and genetically determined variations in the host response [5]. Our solution however would be for the host itself to mount the immune response. This avoids the problems both of immunopathology and of mutation, for crucial to the response is not the make-up of the organism but the terrain: the condition of the host.

  It is said that M. tuberculosis infection in humans results in active disease in a minority (10 per cent), suggesting that most infected individuals can mobilise an effective immune response [1]—provided not compromised by HIV. One factor imparing host macrophage function is protein malnutrition, and hence in humans TB is a disease of poverty. Protein malnutrition is not likely in UK cattle. Nor is it likely in badgers, according to Vic Simpson of Truro Vet Investigation (VI) centre, who has examined badger carcases and stomach contents over the years. Nutritional deficiency in cattle and in badgers would be in the form of a subclinical trace element deficiency. Subclinical deficiencies in selenium, copper, zinc, iodine and cobalt are widespread in UK herds, particularly selenium. This may also be the case in our wildlife. An analysis of the fur of 36 badgers showed on average selenium level of 0.036 mg/kg, nearly 10 times lower than the fur of domestic mammals and often negligible. Copper levels were adequate (Se and Cu analysis report presented to the Bluebell Sett Group Sept 1997). Analysis of badger fur in selenium-adequate areas in Europe is necessary to find out if in Meles meles fur selenium is always low. There are no records of UK badger blood levels. We await with inerest the Central Veterinary Laboratory (CVL) Se and Cu analyses of their 343 liver samples. Livers are usually definitive indicators of nutrient status.

2. RESTORING TRACE ELEMENTS TO DEPLETED SOILS

  Specifically we ask (i) whether by raising the trace element nutritional status of cattle and badgers their immune resistance can be sufficiently enhanced that they become resistant to TB? (ii) are the positive reactors the result of a trace element deficiency that inhibits macrophage activation and their destruction of the bacilli? (iii) is it a trace element/vitamin E deficiency that accounts for the lack of antioxidant protection and the development of TB lesions?

  Goodwin-Jones of Trace Element Services has been pioneering the restoration of trace elements to depleted soils on livestock farms since 1983, solving problems arising from depressed immunity, such as recurring mastitis, failure to thrive and calf mortality, as well as infertility in male and female, difficult births etc. Treatment of the soil raises the nutritional status of the whole food chain—of the cattle through their pasture, fodder and feed, and of the badgers through the worms, insects and small mammals in the badger diet. If we are correct and by this approach immune systems are sufficiently enhaced to induce TB resistance, the route of transmission will be irrelevant. There will be no need to cull badgers and reactor cattle will no longer appear.

3. EVIDENCE THAT SELENIUM DEFICIENCY IS WIDESPREAD

  In their reply to our preliminary protocol, MAFF scientists raised some queries as to the role of individual trace elements, and the evidence for selenium deficiency in UK soils and for low Se blood levels in cattle and sheep. VI centre mesurements of the seleno-enzyme glutathione peroxidase (GSHPx) had shown these to be adequate. The evidence for the statement that "selenium deficiency is widespread in our soils" and hence in our farm animals rests upon the following:

1.   Soil analysis

Goodwin-Jones of Trace Element Services has had soils analysed on over 2,000 livestock farms throughout England and Wales, and a few in Scotland and N. Ireland. Samples are taken from each soil type and terrain, if necessary from individual fields. Total selenium levels are rarely higher than 0.3 ppm, in some areas as low as 0.08 ppm. Only half a dozen famers have had as much as 0.6 ppm in the last two years [6]. A New Zealand soil survey rating classes < 0.3 ppm v.low, 0.3-0.5 ppm low, 0.5-0.9 ppm average, 0.9-1.5 ppm high [7]. Goodwin-Jones restores trace elements according to the need of the individual farm and monitors the results in terms of the response. For an optimum response in cattle and sheep he has found levels should be 0.8-1.2 ppm. Responses specific to selenium include the elimination as a herd or flock problem of failure to conceive (farmers report an increase in cow conception rates from 45 per cent which is the national average, prior to treatment, to up to 90 per cent post treatment), malpresentation especially in lambs, dystocia, large calves at birth ie. gone beyond their time, thick amniotic sacs, retained placentas, contratile ankles in calves and lambs, unwillingness or inability of calves or lambs to suck, some forms of lameness, recurring mastitis and the classic symptom of selenium deficiency in cattle the raised or pumphandled tail. There is no one else to my knowledge, other than Goodwin-Jones who is steadily analysing trace element levels in soils on UK livestock farms, including re-analysis when requested every 10 years or so, three to five years in some cases, and monitoring the responses with the help of farmer feed-back.

  When post-war the Soil Surveys of England and Wales and of Scotland mapped the soils of Britain for their textural qualities and major element availability, almost no attention was paid to trace elements [8] and was begun in the late seventies for Scotland. But selenium and iodine were then thought unimportant and were not included. Only in the 1980s was the seriousness of selenium's omission recognised. Selenium analyses derived from stored soil samples collected since 1948 were added to the inventory by the new body for England and Wales, the Soil Survey and Land Resource Centre. The Macaulay Land Use Institute had no funding for a similar selenium survey in Scotland, hence it was not done. Iodine has been omitted altogether.

  I have not yet seen the data for selenium but the levels at sampling will be totally different to present day levels due to its accelerated removal (a) in crop, meat, milk etc by arable and dairy farm intensification since the mid seventies (b) by acid rain, since sulphate displaces selenium from binding sites (c) by ammonium sulphate fertilisers for the same reason. Goodwin-Jones finds selenium levels need to be topped up every 5 years or so in order to maintain the reponse. Since 1985 the Finns have solved the problem of its deficiency and also unavailability in their soils with a blanket application every year equivalent to 3gm Se/ha by making it mandatory for every bag of NPK fertiliser to contain 6mg Se/kg [9].

  When trace elements are spread on the pastures the animals get them in their food and in the most available form. Mineral licks, injections and boluses are less desirable, less effective and more expensive. Minerals may not be taken by some. High phosphate mineral licks will tie up selenium—it was high phosphate fertilisation that precipitated clinical deficiency in New Zealand. The Se levels incorporated into cake may be sufficient to avoid clinical deficiency on dairy farms, but they are not enough to correct the symptoms of subclinical deficiency described above. Boluses were devised because the word was put out that "trace elements would leach and were a waste of money." Goodwin-Jones has demonstrated the fallacy of this argument in the improved health of the animals and improved economic viability of the treated farms. Once zinc, copper, cobalt and iodine levels have been restored they can last for years. Only selenium losses have to be watched.

2. SELENIUM LEVELS IN UK FOOD

  The low levels of selenium in UK soils are reflected in its low levels in our food. Our intake used to be about 60mcg/day until accession to the EEC in the 1970s when we were obliged to cut our imports of high selenium Canadian wheat containing 200-500mcg/kg from 1m tonnes to 200,000 tonnes/yr, (letter from MAFF`s Nutrition Unit to writer Jan. 22.97) and depend for our flour and bread on UK (ie EU) wheat containing 20-50mcg/kg [10]. As a result, and combined with falling soil levels our Se intake has declined steadily. Oct.1997 MAFF published their 1995 Total Diet Survey showing the population's intake is now as low as 29-39mcg/d [11]. The accepted Reference Nutrient Intake (RNI) given by the Committee on Medical Aspects of Food Policy (COMA) 1991 is 60mcg/d for women and 75mcg/d for men [12]. The cereal figures 20-50mcg/kg reflect the low selenium levels on our arable farms. Moreover the UK intake from home-grown food depends almost entirely on cereals. The Finns estimate that since 1985 40 per cent of their population's intake comes from meat, over 25 per cent from dairy products and eggs and 20 per cent from cereals [9]. The selenium content in our milk in 1995 was reported 10mcg/kg dry matter (DM), equivalent to 1-1.3mcg/l. thus this amount is supplied to the calf. The Finn's figure for milk reported by their Working Group on Selenium 1994 was 120-160 mcg/kg or 12-15mcg/l. Nevertheless it must be added that intakes do not take account of the complexities of selenium bioavailability from different foods. Plant SE-bioavailability compared to that of animal origin is relatively high [13 rev], so that for the human population intakes from cereals are a major source.

3. SELENIUM BLOOD LEVELS

  Limited surveys of Se blood levels in the human population cite 47-60mcg/l [9,10], considerably below the 100 mcg/l considered adequate [14,15]. The Finns' levels rose from 60-70mcg/l prior to 1985 to their present levels of 100mcg/l. Selenium levels in cattle and sheep are usually measured in terms of the activity of Se-GSHPx, an easier parameter to measure and it is thought, providing good correlation with selenium levels. VI centres regularly determine GSHPx levels in cattle and sheep and MAFF asserts that their results do not support the statement "selenium levels are desperately low in UK soils and in the intakes of farm animals." Selenium levels derived from whole blood are reported as GSHPx activity units (U) per ml of packed erythrocytes. Reference values for an adequate selenium intake are now (Shrewsbury VI centre pers. comm):

cattle > 30 U/ml (or 37.5U/g HB) sheep > 50 U/ml (or 62.5U/g HB)

Hb = haemoglobin

  Goodwin-Jones argues that these reference values are inadequate, because cattle and sheep with activities considered adequate have problems that respond when their pastures are treated with selenium or when they are given a selenium injection. He is supported for example by Mclure et al. 1986 [16]. Dairy cows paired as treated and untreated in three New South Wales herds with a history of infertility were orally dosed with long acting selenium pellets, resulting in a mean first service pregnancy rate of 58 per cent compared to 30 per cent in untreated controls. The mean GSHPx activity in the treated cows was 70 U/g Hb, equivalent to 86.9mcg Se/l blood, but these levels are probably still too low.

  Farmers whose soils have been restored with trace elements by Goodwin-Jones report GSHPx activities in cattle of 100-150 U/ml (125-187.5 U/g Hb). In a trial of 40 ewes organised by a farmer in North Wales, the whole blood GSHPx activities in a group whose pasture was treated with selenium-only rose from 17 U/ml (21.1 U/mg Hb) to 107 U/ml (133.8 U/mg Hb); and in a group on pasture treated with the indicated levels of Se, Cu and iodine GSHPx activities rose from 16.6 U/ml (20.7 U/mg Hb) to 129.2 U/ml (161.58 U/mg Hb), suggesting a positive interaction between trace elements. On a South Wales farm with severe deficiency problems, the pastures were treated with Se, Cu, Zn and Co. GSHPx activities in lambs rose in two months to 150-190 U/ml (187-237 U/mg Hb) with marked improvement in health and performance [6].

4.  SELENIUM BLOOD LEVELS AND GSHPX ACTIVITIES: AN UNDEPENDABLE CORRELATION

  Confusion may arise because in humans GSHPx activities only correlate well with selenium status where the selenium blood levels and daily intake are on the low side, as in Northern Europe, New Zealand and parts of the USA ie up to about 79mcg/l, corresponding to an intake of 50-80mcg/d. After that GSHPx activities plateau in the region of 40 U/mg Hb [17, 18, 19], while selenium levels continue to rise. Hence GSHPx is no longer considered a valid index of selenium status in humans and the current method of basing adequate intake on the need to maintain functional GSHPx saturation is not acceptable. In ruminants however, whole blood GSHPx activities correlate well with blood or plasma Se levels to much higher GSHPx values. In a recent study comparing plasma Se and erythrocyte GSHPx activities between cattle and camels fed 2mg Se/day, activities in cattle continued to rise until about 110 U/mg Hb when supplementation was stopped, but camel activities kept on rising at least for another three months thereafter [20]. An examination of an early study [21] to determine the selenium requirement on the basis of tissue GSHPx activities in ewes and lambs on a corn and silage diet supplemented with levels from 0.05ppm to 0.52ppm Se, shows that there was no plateau-ing of erythrocyte GSHPx. After four months, activities were 70 and 75 U/gm Hb in ewes and lambs respectively and still rising.

  It should be noted however that 0.12ppm on a dry matter basis, corresponding to an intake in sheep of 0.25mg/d was then considered optimum and 0.52ppm (1 mg/d) believed far too high. This beliefe led the authors to overlook the fact that there was a drastic reduction between birth and weaning in the GSHPx activities of the lambs whose dams were supplemented with 0.12ppm, demonstrating selenium deficiency in the milk: just before weaning at 10 weeks GSHPx activities had fallen to 34 U/g Hb. They scarcely rose at 16 weeks. Selenium levels fell in parallel, but rose again to near their levels at birth. The milk at the 0.52 ppm intake was twice that at 0.12ppm and GSHPx activities were 75 U/g Hb at weaning and 84 U at 16 weeks, with selenium levels rising in parallel.

  This study demonstrates (i) that although at weaning GSHPx was undersaturated, there was still a diversion of selenium post-weaning to other tissues; (ii) that levels once considered adequate, later prove to be deficient, and this is a general finding with other minerals and vitamins; (iii) that as evidence now indicates, although "classical" GSHPx has an anti-oxidant role, its major function—other than in the reticulocytes—is to serve as a buffer, directing Se flux towards more critical tissues and enzymes [22] such as (a) the immune system [23, 18rev pp 180-202]; (b) the thyroid [24]; (c) phospholipid hydroperoxide glutathione peroxidase (PLGSHPx). It is this enzyme, situated at the interface of cytosol and cell membrane, which is now recognised as the major Se antioxidant, overlapping in that function with vitamin E[22]; (d) plasma selenoprotein P, the first to be repleted when deficiency is corrected [25]; (e) possibly other seleno-proteins not yet identified; (f) one suspects, the brain. Thus selenium is preferentially diverted to those organs and tissues most vital to the organism and when supply is limiting, less essential tissues will go short [26]. GSHPx activity plateaux differ between species no doubt because species differ in their demand and priorities for selenium distribution. Clinically in humans it is found that supplementation of 100mcg Se/d may help asthma. But at least 300 mcg/d in low selenium areas such as Northern Europe could be required to help protect against cancer [27]. One must assume there is low priority to protect an ailing organism.

  The only conclusive test of an adequate selenium intake is when a human or animal responds and some malfunction is corrected, whether this be impaired immuno-resistance, impaired metabolism, infertility, cardiovascular disease, susceptibility to cancer and so on. Only rarely, in severe deficiency is there a specific selenium deficiency disease—white muscle disease in young animals, Keshan and Kashin Beck in humans. But a subclinical deficiency will aggravate a range of weaknesses. Hundreds of farmers have found that problems in their cattle and sheep melted away after correction of soil deficiencies at levels greater than those currently thought to be adequate.

4.  SELENIUM DEFICIENCY AND THE IMMUNE SYSTEM

  Selenium is present in the majority of immune cell types. Phagocytic cells depend on GSHPx or PLGSHPx activity for controlling the peroxides released in the respiratory burst. But also by some unknown mechanism selenium plays a part in lymphcyte proliferation [18 rev pp189-194]. This is shown by the increased lymphocyte response to phytoalexins and other mitogens when selenium is supplemented in vitro. In clincial trails, Se supplementation of 100-200mcg/d to patients with marginal deficiency resulted in stimulation of the immune response in all subjects studied, with enhanced lymphocyte proliferation in response to mitogens or antigens [28]. In the early stages of HIV/AIDs, a low plasma level of Se is associated with dysfunction of T-lymphocytes [29] including reduced natural killer (NK) cell activity [30]. A recent report in New Scientist suggests "treatments for HIV may need to strengthen the immune system, rather than target the virus" (9 Jan 99 p 14). But so far the identity of the seleno-protein(s) that stimulate T-cells is a mystery. I have hypothesised (unpublished research) that the macrophage depends on a regulatory seleno-protein, and that in selenium deficiency it protects itself from oxidative self-destruction by down-regulating the T-cells that activate it. There is also evidence that the thyroid stimulates the immune system (see below).

  Conversion of the prohormone thyroxine (T4) to the active 3,3, 5-triiodothyronine (T3) is catalysed by the seleno-enzyme type 1 iodothyronine 5-deiodinase (IDI) and its underactivity would thus affect an immune response. Whatever the mechanism(s) it is clear selenium is of critical importance to the immune system in some way other than as an anti-oxidant; and that a higher intake than that provided by UK pastures is needed to confer resistance.

5.  COPPER DEFICIENCY AND THE IMMUNE SYSTEM

  Copper deficiency in UK soils is fairly widespread. Goodwin-Jones finds most soils require restoration to 10-12 ppm available Cu, corresponding to an optimum response in livestock copper deficiency problems.

  Copper is crucial to antioxidant function. The antioxidants superoxide dismutase (Cu,Zn-SOD) and ceruloplasmin are copper-dependent enzymes, catalysing the reduction of superoxide (O2-) to H2O2 in the cells and plasma respectively. Copper deficiency lowers Cu,Zn-SOD activity in the macrophage and severely impairs control of the respiratory burst. This is because in the activated macrophage nitric oxide (NO) and O2- are produced simultaneously and unless O2- is removed by CU,Zn-SOD it reacts with NO to form excess peroxynitrite (ONOO-) [31]. ONOO- oxidises sulphydryl groups and initiates lipid peroxodation. It is a powerful bacteriacide, but a runaway reaction will damage the surrounding lung tissue, not least because the gaseous diffusion of O2- and NO allows its activity to be initiated both in the macrophage itself and also in neighbouring cells.

  Although the immune system is altered by copper deficiency, the immunological and biochemical mechanisms are not known [32 rev]. In the macrophages of Cu-deficient mice it was found that a 49 pre cent fall in SOD activity was accompanied by a 28 per cent fall in that of copper-dependent cytochrome c oxidase compared to Cu-replete mice. A lowered activity of cytochrome oxidase would reduce the available energy to the macrophage-dependent immune response.

  Part of the answer must also lie in the effect of lowered levels of Cu-dependent neuroendocrine enzymes and peptide modulators on immune function. (i) Cu-dependent dopamine-B-hydroxylase catalyses noradrenaline (NA) synthesis. Noradrenaline metabolism is known to be altered in copper deficiency and to interfere with the regulation of cytosolic calcium [33] and thus with the activities of all cells, including immune cells. (ii) Peptidyl glycine amidating mono-oxygenase is a copper-dependent enzyme required in vasoactive intestinal peptide (VIP) synthesis. VIP couples with its lymphocyte receptor and alters the immune reponse [34]. (iii) a copper-dependent neuromodulator enhances the release of luteinising hormone releasing hormone (LHRH) [35] and thus of luteinising hormone (LH). LH affects the thymus [36].

  When copper becomes limited (or in excess) there are changes in the activity of a number of other enzymes [25]. One of these is GSHPx [18] where copper deficiency down modulates GSHPx activity and GSHPx mRNA. It would be logical to deduce that this is the body's response to copper deficiency since if cu,Zn-SOD was underactive, less H2O2 would be made; less GSHPx would be required to reduce it to water and more precious GSH conserved. In copper deficiency, lowered Cu.Zn-SOD activity occurs primarily in the liver where it is accompanied by a down-modulation of Se-GSHPx, with faecal excretion of some selenium but also a considerable diversion to provide extra selenium to brain, lungs, heart and thymus [37]—perhaps in order to compensate for copper deficiency in those tissues. Copper-selenium interactions need investigation, not only where one or other or both are deficient, but also where one or other or both are present in excess. Goodwin-Jones has discovered from responses in the field that raised selenium levels protect against copper toxicity in sheep [6].

  In addition to its direct effect on phagocyte function, copper deficiency affects cell-mediated immunity. It alters the pattern of lymphokine release and thereby lowers the ability of T-cells to activate macrophages [32 rev].

  MAFF cautions that copper treatment of soils would require care because ionic copper is an oxygen stressor. Current literature does not support the idea that Cu ions initiate auto-oxidation reactions as do Fe2+ and Fe3+ ions. Nor indeed as MAFF suggests, do CO ions (Letter to the writer Nov 16. 98). These misconceptions arise from chemically induced experimental free radical generation. Such experiments cannot be extrapolated to biological systems. Ionic copper has a high affinity for ligands capable of bonding to it and in consequence all measurable copper in biological systems exists as complexes or chelates tightly bonded [38]. The evidence demonstrates unequivocally that adequate levels of copper to activate superoxide dismutase and a range of other copper-dependent enzymes are essential to leukcyte and lymphocyte functions, and thus we may conclude to resistance against M.bovis. Nor must it be forgotten that the Cu-depedent lysyl oxidase mediates crosslinking of collagen and elastin and will contribute to repair of TB lesions.

6. ZINC DEFICIENCY AND IMMUNE FUNCTION

  Zinc deficiency in UK soils is also fairly widespread. Zinc may be less available on limestone or chalk soils of raised pH. Goodwin-Jones finds levels of 12-15 ppm should be aimed for to prevent symptoms of subclinical deficiency. Once again interactions with other trace elements are important. Interactions impairing Cu. Se, and Co are described below and those with iodine in the thyroid section.

  Deficiency reduces the numbers of lymphocytes and of macrophages [39 rev]. The remaining lymphocytes however are fully functional: proliferation and interleukin 2 (IL-2) production by mitogenic or allogeneic cell stimulated splenic T-cells from deficient mice were similar to those from adequately fed mice, regardless of whether the zinc level of the culture system was adequate or limiting [40]. IL-2 affects activated macrophages, contributing to the regulatory signals involved in inflammation and in wound repair [41]. Thus in zinc deficiency minor challenges can be dealt with but insufficient lymphocytes are available for mobilisation against major ones.

  Zinc deficient macophages on the other hand, cannot mount an effective respiratory burst, and thus cannot kill intracellular pathogens, which are likely to kill the host instead [42]. Reasons for this include: (i) a number of enzymes in the cell are zinc-dependent, notably phospholipase A2 (PLA2) and phospholipase C (PLC) which activate latent NADPH oxidase in the phagolysosome membrane and accept electrons from NADPH, transferring them to O2 with release of O2- into the phagolysosome. O2- initiates the free radical cascade (ii) The zinc in Cu, ZN-superoxide dismutase unlike copper plays no part in the dismutation (production of H2O2 from O2-). Zinc ions have a structural role. They keep enzymes inactive by binding their sulphydryl groups, protecting against disulphide formation and enzyme inactivation [43]. A large number of enzymes and proteins including those involved in replication and transcription are protected in this way [44]. A decreased activity of SOD in the macrophage will mean as was the case in copper deficiency, that O2- reacts with NO to initiate a peroxynitrite free radical cascade, damaging to lung tissue.

  Zinc deficiency induces an underactivity of catalase, the antioxidant that directly reduces peroxides to water or alcohol, overlapping with GSHPx, and also of glutathione reductase the enzyme that restores levels of GSH, the substrate for GSHPx [45]. It thereby interferes with selenium function. A depression in protein synthesis by zinc deficiency suggests that the reduced activities of SOD, catalase and glutathione reductase is a result of inhibition of their biosynthesis, again due to inadequate binding of sulphydryl groups [46]. However the regeneration of enzyme activity and of macrophage function in response to zinc is very fast. Preincubation of macrophages from deficient mice for one hour with zinc chloride at five times the physiological level completely restored their ability to kill an intracellular parasite [42].

  Goodwin-Jones had a number of reports from farmers where sheep with all the signs of vitamin B12 deficiency did not respond to cobalt. The cobalt deficiency symptoms disappeared when they were given as little as 3mg zinc/day. This is because rumen bacteria require zinc to synthesise B12 [47].

  Zinc is a structural component of cell membranes. It stabilises membrane structure by binding to the sulphur and nitrogen groups in proteins and to phospholipids via their phosphate groups. It also takes part in the assembly of microtubules on which membrane structure depends [48]. Loss of zinc from membranes is the first consequence of dietary zinc deficiency, or of zinc excretion from the body due to stress. It induces changes in molecular conformation and lipid packing, reducing membrane fluidity and interfering with intercellular communication [49]. It would also I suggest interfere with transmembrane methylation [50] thereby affecting gene expression. Zinc overlaps in function with vitamin E. Both are antioxidants and vitamin E too regulates membrane fluidity [51 rev].

7.  COBALT DEFICIENCY AND THE IMMUNE SYSTEM

  Cobalt deficiency is particularly acute on soils where it is deficient in the parent rock, such as granite, and on limestone soils where the raised pH may make both cobalt and zinc less available. Optimum levels of available cobalt are 1.8-3.0 ppm.

  Adequate cobalt ie vitamin B12 levels have powerful effects on the immune system, particularly in the ruminant. (i) B12 can help suppress internal parasites and a number of infections. Goodwin-Jones finds a response to cobalt in lower worm counts, and prevention or clearance of diseases such as orf and infectious keratitis (New Forest). (ii) As the cofactor of methyl malonyl CoA mutase in the propionate pahway for the synthesis of liver glucose, it is crucial to ruminant energy metabolism, and thus indirectly of immune function. (iii) As cofactor of 5-methyl tetrahydrofolate homocysteine transferase, it catalyses the reconversion of homocysteine to methionine with recycling of S-adenosyl methionine (SAM), the methyl donor on which all cellular methylation reactions depend eg for transmitter synthesis and degradation; for degradation of histamine, cell metabolites and xenobiotics; for fatty acid transport and membrane synthesis; and for regulation of gene expression, including the rapid requirements for gene expression in the immune system and its cytokine network.

8.  IODINE DEFICIENCY AND THE IMMUNE SYSTEM

  Iodine levels can be very good or very poor, although in general they should not be too poor in a country where nowhere is far from the sea. They may vary inexplicably between fields. Optimum levels for total iodine are about 6ppm.

  Iodine may directly affect phagocyte function : iodene released from a wound dressing "iodosorb" was found to modulate secretion of cytokines by human macrophages [52]. This is an area that should be researched. Iodine's main function is in the iodination of tyrosine for the production of thyroid hormones—thyroxine (T4) and (T3). subclinical iodine deficiency produces under-active thyroid hormones, lowering energy metabolism and impairing regulation of gene expression, thereby indirectly affecting differentiation and proliferation in the immune system. However not only iodine but a number of other elements are required in thyroid biochemistry: (a) in thyroid hormone production and (b) in thyroid hormone activity.

(a) Iodine, selenium zinc, cobalt, and probably copper are all required in thyroid hormone production. Selenium contributes as cofactor of GSHPx to the removal of excess H2 O2 produced in the oxidation of iodide during the iodination of tyrosine for hormone synthesis [18 rev pp140-146]. Zinc deficiency interferes with thyroid function [53 rev] and induces T3 deficiency. Trials with mild experimental human zinc deficiency found serum TSH, total T4 and free T3 declined by about 20 per cent in one month, followed by a significant reduction in metabolic rate [54]. Cobalt deficiency appears to exacerbrate goitre where there is iodine deficiency [55]. Copper could be important for the production of thyroid releasing hormone (TRH). In a study of Cu-deficient rats the levels of immuno-active plasma thyrotropin (TSH) and of total thyroxine were depressed, suggesting a lack of bioactive TRH [56]. There is evidence that a copper-dependent enzyme is required for the alpha amidation of the COOH terminal of the TRH peptide [57].

(b)  The seleno enzyme type I deiodinase (IDI) is required for thyroid hormone activity. It catalyses the conversion of T4 to T3. A depressed activity of IDI in selenium deficiency will affect T3 even if iodine is adequate, and amplify some of the consequences of iodine deficiency despite a number of homoeostatic mechanisms to conserve T3. This is particularly the case in the ruminant. Humans can produce T3 from T4 via IDI in the thyroid, assisted by a preferential diversion of selenium. But the ruminant must rely entirely on its production in the peripheries. Hence it may be more susceptible than either rodent or human to thyroid related effects in selenium deficiency [18]. Vitamin B12 and zinc are essential to the T3 activation of nuclear processes: B12 for the synthesis of the methylating agent S-adenosyl methionine and consequently the regulation of gene expression; and zinc for the protection of nuclear enzyme sulphydryl groups.

  Thyroid underfunction is known to be conducive to milk fever and hypomagnesaemia. Farmers report that once their fields have been treated with copper and selenium they no longer see milk fever or staggers (hypomagnesaemia) and magnesium supplementation is not necessary [6].

9.  THYROID HORMONE ACTIVATION OF MACROPHAGE PHAGOCYTOSIS AND SUPPRESSION OF INTRACELLULAR BACILLI

  The role of thyroid hormones in macrophage activation was researched by Lurie some 50 years ago when there was an extensive literature on resistance to TB [2]. It had been discovered that hyperthyroidism increased the activity of phagocytes, and "the preponderance of available experimental evidence favoured the view that, within certain limits, hyperthyroidism tends to increase resistance against tuberculosis, while hypothyroidism exercises an opposite effect."

  Lurie used the rabbit as a model to study human resistance to M.tuberculosis. He exposed rabbits to tuberculin bacilli by inhalation and administered intramuscular injections of T3 or T4. T3 was the most effective, as was to be expected. It was found that:—

(i)  During four weeks exposure when 40mcg T3 was administered every other day to rabbits of intermediate susceptibility, the inception of TB was almost completely suppressed. There were few intracelular bacilli and tubercles were absent or very rare. Untreated controls had tubercles with extensive, far advanced caseous centres containing moderate numbers of bacilli and widespread tuberculous granulation tissue with maturing epitheloid cells harbouring some bacilli. 100mcg T4 every other day had a similar effect.

(ii)  In a second study designed to follow the progress of the disease during more severe exposure, T3 restrained the intracellular accumulation of bacilli in the lungs and increased in varying degrees their spread to the draining lymph nodes. About 20 times more inhaled bacilli were required to produce one tubercle in the treated rabbits as in the controls. T4 had a similar effect.

(iii)  The transport of bacilli from the lungs to draining lymph nodes was greater in the natively resistant animals than in the natively susceptible.

(iv)  The development of an existing TB was suppressed in some degree. T3 was administered three weeks after initiation of the disease to rabbits with matching levels of infection to the controls. The disease regressed during the 42 - 54 days of treatment, for when the rabbits were killed there was a two to three fold reduction in the number of residual tubercles per unit of inhaled bacilli in the treated compared to the controls.

(v)  Very susceptible strains were not significantly protected by hormone administration, compared to controls.

(vi)  Thyroidectomy or a daily subcutaneous injection of the thyroid antagonist propyl thiouracil increased the severity of bacillary infection and reduced the transport of bacilli from the lungs to draining tracheobronchial lymph nodes compared to the controls. Two of the 12 thyroidectomised rabbits had bacilli accumulating in the spleen, indicating that hypothyroidism promoted multiplication of bacilli.

(vii) The basic metabolic rate (BMR) was closely monitored, and increased by 35 to 40 per cent above baseline during T3 treatment. However it was shown that it was not the increase in basic metabolic rate itself that conferred resistance because (a) when 2,4- dinitrophenol was administered—a substance that shares with thyroid hormones the capacity to increase oxygen consumption—the BMR rose in exposed rabbits by 28 per cent compared to 11 per cent in the controls and to 41 per cent in the T3-treated, but the lesions of those given 2,4-dinitophenol contained greater numbers of bacilli than the controls. (b) the BMR of highly susceptible rabbits rose markedly in response to T3 but there was no corresponding increase in resistance.

(viii) There was a reduction in the inflammatory response during hormone treatment, but mere hyperthyroidism without a pronounced concomitant restriction of the growth of the bacilli did not reduce the allergic inflammation. This was demonstrated when heat-killed bacilli were repeateedly injected and there was no significant reduction in the inflammatory reaction. Lurie concluded that "the reduced tuberculin sensitivity suppressed by hyperthyroidism was due to the inhibition of bacillary growth and hence to reduced antigenic stimulation."

(ix) However, thyroidectomy and propyl thiouracil both suppressed the inflammatory reaction, even while exacerbating the disease. Hence he proposed that hypothyroid suppression of allergic inflammation must be by a different mechanism.

(x) Hyperthyroidism is associated with hyperplasia of the lympathic tissue of the spleen and hypothyroidism with its atrophy.

  In searching for a mechanism whereby thyroid hormones activate macrophages he noted that they provide the energy for phagocytosis and mobilisation of leukocytes by enhancing glycolysis, cellular utilisation of glucose and stimulation of insulin secretion. Hence there effects—immuno-resistance, growth stimulation, tissue repair, hypertrophy of lymphatic tissue—were the direct opposite of those of glucocorticoids—immuno-suppression, inhibition of growth and tissue repair, atrophy of lymphatic tissue. But he was forced to conclude that the mechanisms responsible for the suppression of intracellular bacilli in the hyperthyroid state and virulence in the hyperthyroid state could not be identified in the present state of knowledge (1964).

  I suggest that advances in molecular biology and genetics now allow us to make a speculative interpretation of (a) the stimulatory effects of thyroid hormone on the immune system; (b) the suppresion of inflammation and resistance in the hyperthyroid state; (c) Lurie's insight that it was thyroid hormone's effect on glucose metabolism which was connected to the enhancement of immuno-resistance.

  Support for these ideas comes from research by Oppenhemer and Schwartz [58 rev].

1.   Studies of mRNA activity profiles.

  Hepatic mRNA was extracted from hyperthyroid, euthyroid and hypothyroid rabbits and from hypothyroid rabbits treated by T3 [58]. The impact of the thyroidal states and of T3 on gene expression was analysed by means of a two dimensional electrophoretic technique following generation of protein products in an in vitro translational system. Each spot represented the protein product of a specific mRNA sequence and its intensity was assumed to be proportional to the translational activity of the mRNA. Spots varied in response to altered thyroidal status, some increasing and other decreasing in intensity in the transition from hypo- to hyperthyroidism ie thyroid hormone both attentuates and augments gene expression [59]. The results further suggested "that eight per cent of all genes may be selectively responsive either directly or indirectly to thyroid hormones" [58].

  Hypothesis I: Since thyroid hormones stimulate macrophage phagocytosis [2], I propose that T3 activates the expression of one or more of those enzymes crucial (a) to macrophage activity and release of superoxide and NO eg NADPH oxidase, PLC, PLA2, PKC, NO synthase or of various receptor proteins or immuno-stimulants such as interleukin 12, critical in TB resistance (New Scientist 9 January 1999 p21) and (b) to macrophage free radical control eg SOD, GSHPx, glutathione reductase, catalase.

  In hyperthyroidism, enzymes of high activity are produced in both the activating and inflammatory-control groups. The exceptions will be those highly susceptible rabbits where genetically determined enzyme(s) of lower activity are produced with insufficient production of O2".

  In hypothyroidism enzyme(s) of such low activity are produced that too little superoxide is generated for an effective resistance response, and the limited numbers of inflammatory free radicals can be quenched even if the enzymes controlling it are of low activity (see ix).

2.   Studies of T3-carbohydrate interaction

  Diets high in carbohydrates and containing no fat are known to be potent inducers of lipogenic enzymes and fatty acid synthesis in the liver. The authors discovered that T3 augments by some 40-fold the induction of malate enzyme, fatty acid synthetase, and the pentose phosphate pathway enzymes glucose-6-phosphate dehydrogenase and 6-phosphogluconate dehydrogenase. But what was the mechanism? Studies on malate enzyme showed that administration of T3, carbohydrate feeding or both, synergistically stimulated the synthesis of the specific mRNA for malate enzyme. Other studies showed that glucose or its glycolytic intermediates such as pyruvate, lactate or glycerol ie glucose metabolism was necessary for the induction of malate enzyme mRNA synthesis and interacted synergistically with T3. The finding that T3 appeared to multiply a carbohydrate-generated signal raised the possibility that T3 in general might multiply co-ordinated regulatory signals [58]. Similarly we can conclude that a diminution in T3 induction and augmentation due to trace element deficiency might impair fatty acid synthesis pathways.

  We now return to Lurie's insight that in some way T3 activation of the macrophage was connected to glucose metabolism. And if so we must also take on board that B12 is crucial to macrophage activation in the ruminant. This is because dietary carbohydrate provides glucose for the synthesis of triacyl glycerols. Liver glucose surplus to energy and glycogen requirements is converted to triacyl glycerols and stored in fat cells as a major source of energy or is used in cell membranes as precursors of phosphoglycerides and sphingolipids. But in the ruminant, liver glucose is almost totally derived from rumen propionate and other short chain organic acids, produced from carbohydrate by rumen bacterial fermentation, and dependent for its re-conversion on a pathway that includes methyl malonyl CoA mutase whose co-factor is B12.

  Is there a connection coordinated by T3 between the biosynthesis of fatty acids and the generation of superoxide by phagocytes?

NADPH is required as the reducing agent (electron donor) for the producing of O2- and its initiation of the free radial cascade. It is also required as the reducing agent in the biosynthesis of fatty acids, including membrane fatty acids ie. for the saturation of double bonds, and arises from two major sources, depending on cell type. In the liver, adrenal cortex and mammary gland it is largely formed by the reactions of the pentose phosphate pathway, especially the action of glucose-6-phosphate dehydrogenase

    Glucose-6-phosphate + NADP+ ——— pyruvate + CO2 +NADPH +H+

  In fat or adipose cells NADPH is largely generated by malate enzyme.

    Malate + NADP+ ——— pyruvate + CO2 +NADPH +H+

  Hypothesis II. What is the source of the NADPH generated in the phagolysosome? It would appear to be those lipogenic enzymes in the pentose phosphate pathway, glucose-6-phosphate dehydrogenase and6-phosphogluconate dehydrogenase [39 fig. 1] whose mRNA is induced and augmented by T3 [58]. If so it would mean that T3 not only coordinates the biosynthesis of the phagolysosome fatty acid membrane, but whenever pathogenic signal activates NADPH oxidase, it would also coordinate the production of O2- and free radical destruction of the TB bacilli. This is a key issue and requires research.

  Lurie [2] warns of instances where hyperthyroidism via hormone injections and also BCG vaccinations caused rodent mortalities that did not occur in euthyroid rodents. This it was suggested, could be attributed to an increased susceptibility of hyperthyroid animals to endotoxins. Hence if thyroid hormones were to be administered, it must only be to otherwise healthy cattle.

  However it should not be necessary to make animals hyperthyroid to boost their resistance. I propose that the critical importance of the thyroid in TB resistance be recognised, and that all the necessary trace elements are provided in the cattle diet at continuing level to induct T3 activation of the macrophage. The same could be said for the badger. This can be ensured by raising the nutrient status of the entire food chain.

10.  OUTSTANDING QUESTIONS

1.   Why is bovine TB not rampant in Scotland?

  Selenium and copper levels in cattle are lower in Scotland than in England and Wales if we are to go by the VI centre reports of hyposelenaemia together with white muscle disease, and hypocupraemia regularly published in the Vet Record. Some other factor(s) must also be involved. Evidence suggests one of these may be the weather. King et al. 60 looked at climatic variables, investigating their influence on the survival of M.bovis outside the host on the premise that if (a) there was badger to cattle transmission and (b) it was field based rather than through direct contact, the risks to cattle would depend on the persistence of M.bovis in the environment. Field measurements were not possible because the bacillus is a grade 3 pathogen. Hence strains of M.bovis were isolated from caught wild badgers and bacilli survival measured when exposed to UV radiation, temperature and relative humidity under stimulated environmental conditions. The effect of climatic conditions on the TB incidence in cattle was investigated in a TB pocket in the Bude area with the help of data from the British Meteorological Office and MAFF data on new confirmed breakdowns 1986-1995 in parishes with a 20 mile radius of the meteorological station in Bude. An interactive software program identified the meteorological variables most highly correlated with the inter-annual variation in the number of herd breakdowns.

  Results suggested (a) the survival of M.bovis was favoured by climatic conditions typically experienced in spring: low levels of UV radiation, low temperatures and high relative humidity; (b) the strongest correlation between the number of herd breakdowns and a single environmental factor was temperature. Temperatures as low as 3C reduced the risk of infection, as did longer hours of sunlight. The three meteorological variables correlated to the annual numbers of herd breakdowns were rainfall, sunlight and temperature, but in all cases only when the data referred to the previous year. High rainfall in the previous year was an important risk factor for infection, particularly in the period 25 April to 9 June ie following spring turn-out. The total rainfall in that period was able to explain 80 per cent of the variation in the number of herd breakdowns in the following year.

  The authors concluded that these results supported the hypothesis that transmission was field based, and since none of the three variables showed winter correlations, the cattle were at little or no risk of becoming infected in winter. With respect to M.bovis the authors observed it was more likely that contamination of pastures in spring was new rather than residual, and that over-wintering of the bacilli was not significant. It was the low temperature, low exposure to UV radiation and high relative humidity typical of spring weather that favoured the survival of newly expelled bacilli, and this was also the time of greatest susceptibility of the cattle to infection.

  In my view as a dairy farmer, the data can be interpreted differently. What are we looking at? (a) the survival of bacilli on the assumption that it is transmitted from badger to cattle; (b) the susceptibility of cattle to infection. The fact that the same meteorological conditions seem to favour both does not necessarily mean the two are correlated in the manner indicated. I propose that the correlation between herd breakdown and the meteorological data of the previous year suggests another scenario.

(i) Cattle to cattle transmission in the winter months via exhalation of the confined space of the cubicle house or cattle shed is far more likely than badger to cattle transmission in the field.

(ii) The correlation of total rainfall in the period 25 April to 9 June with 80 per cent of the variation in the number of herd breakdowns in the following year is highly significant. Those months, April to early June can be difficult months for cattle. It is then that weather conditions can be most challenging. There is no mention of "rain plus wind" as one of the meteorological factors, but rain accompanied by wind—typical of those months—chills them more than any other type of weather, precipitating staggers, pneumonia, milk fever and immuno-depression. Muddy fields are conducive to lameness, another stress factor.

(iii) Their selenium intake is likely to be lower from their deficient pastures than from their winter feed. Dairy cows will be getting much smaller rations of selenium-fortified cake and in-wintered sucklers and steers none at all. Magnesium intakes will be lower from the spring grass, as may copper intakes, for copper rises in the grass in response to sunlight and farmers observe staggers occurs when skies are overcast (Goodwin-Jones, pers.comm). Soil temperatures in spring are still low and consequently mycorrhizal fungi that mobilise nutrients for their associated plant species are not yet active.

(iv) When the cattle are turned out their reserves built up from the previous summer may be at a low ebb. Farmers report that after January troubles with lameness, mastitis and other infections often begin. If they are autumn calvers they will be carrying a calf, another stressor.

  I propose that where springtime weather conditions are stressful and nutrient availability low, any animal harbouring M bovis picked up from another during winter housing will be less able to fight off challenge than at any other time. The delayed hypersensitivity tuberculin skin test will be more pronounced and the risk of latent bacilli establishment or lesion development greater. The correlation will of course be with the year before, because if the cattle had had a benign spring—less stress, greater nutrient availability—then they would be more likely to have resisted the challenge.

  But why is bovine TB not rampant in Scotland? Could it be that in the south west of England, also Northern Ireland there are more muggy days, more condensation after a cold snap, higher relative humidity in shed and parlour with water running down the walls, favouring the survival of exhaled M bovis, but in the cubicle house and not as proposed in the field? Is it that just as M tuberculosis infectivity thrives under the combined conditions of malnutrition, damp housing and overcrowding, so does M bovis infectivity thrive under the combined conditions of inadequate trace element nutrition, high humidity and overcrowding?

  But how is TB spread between farms, between districts? Why is it spreading out from the south west to Staffordshire, Derbyshire, Cheshire, Wales? Is it by bought in cattle, or even more apposite—bought in imported cattle? Is it transmitted via wildlife? The badger stands accused because it harbours infection. But starlings, other birds and wildlife are a menace to cattle as a prime source of salmonella infection when coming into contact with feed. (MAFF vet Graham David, Farmers' Weekly, 25 December 1998, p28) where they are carriers of the disease. Can they carry M bovis? Are there cattle carriers of M bovis, just as there are said to be human carriers of M tuberculosis? These are indeed outstanding questions, but the urgent question is whether cattle can be made resistant. If they can, then all resources should be devoted to it and the problem of transmission examined later.

2.   Are organic farms protective against TB?

  There is anecdotal evidence of organic farms being free of TB, while their neighbours are not. This requires an epidemiological survey. Organic farmers do not use nitrogenous fertilisers and selenium levels are thus not depleted by ammonium sulphate leaching with selenium displacement. The management is less intensive. Hence animals are less stressed and immuno-suppressive cortisol levels less likely to be raised. Soil nutrients are mobilised and made available by high microbial activity and turnover, in contrast to chemically fertilised soils that suppress microbial numbers and activity. Organic farming reintroduces deep rooting grasses and herbs that tap nutrients from the subsoil and promote biodiversity with different species and varieties concentrating specific nutrients.

3.   Is the species/strain composition of soil mycobacteria a factor in promoting resistance to TB?

  Environmental mycobacteria are known to play an important role in the initial priming of the immunological response in humans and animals, thus determining their response to subsequent infection with pathogens [61]. Studies have indicated that certain soil mycobacteria induce an immune response to M tuberculosis and M leprae [62] and to M bovis in the badger [63]. A three year study of mycobacterial populations in the soils of a Gloucestershire farm with a recent history of TB infection was undertaken as part of a project examining mycobacterium and animal immunity [64]. Fortuitously it was a dairy farm converting to organic. Some 50 biotypes were isolated including high numbers from streams and drinking troughs, but there was a marked change in population composition in the three years of study. This raises the question as to whether organic farming methods influence population composition. Further studies are required to find out whether the numbers of immuno-enhancing strains are higher in organic than in chemically farmed soils.

4. Badger nutrition

  Badgers are remarkably resilient to TB In a recent study [65] of badger carcases submitted to two VI centres in Devon and Cornwall for the diagnosis of TB, of 118 with no visible lesions 15 were shown on cultural examination to be infected with M bovis. Histopathological inspection of lung and kidney sections identified microscopic lesions. Most of these gave evidence of a host containment reaction often early in the pathogenesis, with inactivation or death of the bacilli. Animals with no visible lesions appeared to represent an early containment or latent phase in the majority of infections, with possible complete resolution or cure in some cases. Where there was reactivation of an arrested infection this could have been triggered by a number of stress factors [65]. It has been suggested one of these is malnutrition, evidenced by a lower body weight of different cohorts of badgers, attributed to over-population and competition for food resources [66]. Vic Simpson of the Truro VI centre refutes there is evidence of malnutrition. The alimentary tracts of Cornish badgers are full of earthworms even on poor soils. Remains of frogs, hedgehogs, fruit, grain are also identified but never in all his experience has he found remains of bird or lamb which hungry badgers are now accused of taking (pers comm). Gallagher et al [65] observe "No simple linear relationship between population density and the prevalence of tuberculosis has been found [67] but populations undergoing nutritional stress would possibly be more likely to succumb to tuberculosis."

  In a report in Farmers' Weekly (28 November1997) a farmer complained that badgers took 5kg of minerals per night from the dry cows' trough. If true, what were they looking for? Is it that their diet of worms and insects cannot supply adequate nutrition? A study by Dublin Veterinary College suggested that mineral deficiency was one of the husbandry factors conducive to TB in cattle [68]. It could be equally conducive to TB in badgers. I suggest that badgers may well get enough worms and insects, just as cattle get ample feed and fodder, but that they too have a trace element dietary deficiency inducing immuno-suppressive nutritional stress.

5. Badger parasites

  Since the 1970s a reservoir of TB has been found in badgers in areas of Gloucestershire, Devon and Cornwall. In the same study [65] of 118 carcases, of the 15 shown to be infected with M bovis 8 had granulomata infested with lungworm parasites and six had adiaspiomycosis (fungal) granulomata. The lungworm species appears to be restricted to Cornwall and parts of Devon, but the fungal parasite has also been found in Gloucestershire. This suggests that badgers succumb more readily to TB if their immune systems are already stressed with a parasite load. Mammal parasites are controlled by adequate levels of B12, a vitamin requiring cobalt for its structure and zinc for its biosynthesis. Large areas of TB-prone Devon and Cornwall are on granite where cobalt is deficient, whereas the Lizard on ultrabasic rock high in cobalt is not a TB area. On the Gloucestershire limestone cobalt and zinc will tend to be unavailable due to raised soil pH. I suggest here may be the explanation for the frequency of badger TB in these areas. Cobalt and zinc deficiency impair the ability of badger gut microorganisms to synthesise B12, and low B12 levels in turn impair badger resistance to parasites, making them more vulnerable to invasion by M bovis Selenium and perhaps other deficiencies will then contribute to some becoming infectious. Badger to badger transmission in the sett I suggest is the likely route for acquiring infection.

6. What husbandry factors contribute to the suppression of cattle immno-resistance?

  Researchers at University College Dublin Veterinary College conducted a matched case-control study of 80 dairy herds with a history of recurrent tuberculosis 1986-1990 and 80 controls that had been free of it since 1982. A 21-page questionnaire was used to identify the factors contributing to recurrent outbreaks [68]. questions were asked on the presence of badgers, but there were no biased assumptions on badger transmission. Farm management factors identified as significantly associated with increased risk were: failure to provide mineral licks, rough grazing areas, purchase of bulls, spreading of slurry. There was some association with cubicle housing. Overall, intensively managed herds were at greater risk.

(i)  Mineral deficiencies: The risk of a chronic TB problem was increased substantially where there were areas of rough grazing and no provision of mineral licks. The relative odds of 1.7 for rough grazing, 2.75 for no access to mineral licks, soared to 30 on farms where both factors were present. The authors deduced that the areas of rough grazing were likely to be on poor, mineral deficient soils and a failure to compensate by feeding minerals would exacerbate the effects of deficiency. In a Russian study reduction in resistance to tuberculosis had been attributed to a deficiency of vitamins A,B,C,D, cobalt, copper, iron, iodine and manganese [69]. (I would put in a caveat here, that high soil manganese makes cobalt unavailable). It was also noted that a higher proportion of case than control herds were fed "straight" concentrates rather than compounds. Again this suggests a deficiency factor. Most compounds are fortified with antioxidant vitamins and selenium.

(ii)  Purchase of bulls: The authors advised there was a need to study the importance of introducing the disease via the purchase of bulls. They reported however that 12 of the bulls (35 per cent) housed during the winter prior to the survey were deemed a reactor. This is an astounding figure, and I suggest it was not the purchase of a bull that introduced disease, it was the vulnerability of the bulls themselves. Bulls have a higher demand for selenium than cows—it is a male mammal requirement—because they lose a fraction of their selenium reserves with each ejaculation. The most abundant protein in the mitochondrial capsule of the spermatozoa is a seleno-protein. Sperm mitochondria are concentrated in the midpiece, a segment of the tail so that in selenium deficiency there is abnormal tail morphology and motility is lost, impairing bull performance. The levels of selenium in the spermatozoa are unusually high, 20-30 ng/mg protein in the bull and rat [70 rev]. If it is a correct hypothesis that selenium deficiency is a major factor in the depression of immuno-resistance to TB, then it can be predicted that bulls are at greater risk than cows. It was those farms that used a bull, home-bred or purchased that were vulnerable. Not only was the bull more at risk, he was a reservoir of infection. The presence of a bull and whether he is housed with the cows should be in any questionnaire surveying factors conducive to TB on UK farms. Meanwhile it would be wise for farmers to be alerted to protect their bulls with extra selenium.

(iii)  Spreading of slurry: Farms that spread slurry on pasture without prior storage were at greater risk than those producing other types of manure ie, farmyard manure or compost, or that stored slurry before spreading it. Cattle were at highest risk if they grazed land where raw slurry had been put out in the previous two months. I propose that this was not due as suggested [68] to cattle inhaling spores of M.bovis during the spreading process, but to the challenge of whatever pathogenic organisms the raw slurry harboured—pathogens that would have been destroyed by other micro-organisms in the farmyard manure heap, and even in the limited digestive processes of the slurry tank. E.coli 0157 although admittedly more persistent than most, was found to survive on grass for 81 days, long after the slurry had washed in (Daclan Bolton, research officer, Ireland's National Food Centre, reported Farmer's Weekly 10 October 1997). It should be noted that some farmers are guilty of putting out their animals even before the slurry has gone in. I suggest that raw slurry is a pathogenic liability in that cattle fighting off a secondary challenge will be less able to muster resistance against any TB bacilli they might be carrying.

(iv)  Cubicle housing: In some set-ups there may be stress due to short uncomfortable cubicles creating hock and other sores; no or inadequate bedding; slippery or uneven yards and walkways; competition at the silage face where the timid lose out. Stress it must never be forgotten, induces secretion of cortisol, an immuno-suppressant.

(v)  Intensification: High productivity puts strain on the cow. Moreover intensively managed farms pay less heed to cows in need of care and protection eg from bullying, the effects of fright and other stresses. Cows with incipient lameness or unwellness may go unnoticed and must then depend on the resources of their own immune systems to pull them through.

11.  SUGGESTED REVISION OF THE PROTOCOL

  I endorse the major criticism of MAFF scientists that the protocol put forward in the first draft is unsuitable for research purposes. No convincing evidence however positive would be gained by treating one farm with a history of TB and monitoring the effects compared to a matched control. Only if a considerable number of farms backed up the hypothesis would the results be significant. Yet to treat more than one or two farms is outwith our resources.

  I propose instead the West of England TB and Badger Study Group's research be focused on a retrospective survey of farms where trace elements have been restored to optimum levels by Goodwin-Jones, to see if there has been any protection against TB compared to neighbouring farms. This information is "fusting unused" and could serve as a pilot trial. The protocol should be revised as below:

(i)  The dates of TB testing before and after treatment and the results ie, number of reactors: nil, positive or negative recorded.

(ii) Soil analyses, husbandry methods and details of cattle/calf health and fertility problems prior to treatment, their breed, the breed of bull and whether in the herd or by AI have all been recorded by Goodwin-Jones. The effects eg on cattle and calf health and on performance are obtainable from the farmer and his vet.

(iii) The soils should be reanalysed for present trace element levels and changes in pasture species and productivity obtained from the farmer.

(iv) Management changes as in fertiliser/liming practice, feeding and mineral usage, present cattle health and performance; numbers and dates of bought in cattle, information on the use of a bull should be reported in a new questionnaire.

(v) Similar information as in (iii) and (iv) should be obtained from one or more of the neighbouring farms where farmers are willing to cooperate.

(vi) Trace element blood levels prior to treatment will usually be unknown. Present levels should be analysed wherever blood is available from routine veterinary tests and compared with those of neighbouring farms.

(vii) Worm and insect counts should be made on the treated and neighbouring farms and worms analysed for trace element content.

(viii) The researchers must depend on MAFF scientists and the local VI centre vets for information on the local badger population, the incidence of TB infectivity and lesions, trace element analyses of fur, blood or livers. In areas where no work is being done on badgers perhaps carcases from road casualties will suffice. Farmers, badger groups and others may be able to give information on badger populations and behaviour.

  It is to be hoped that enough treated farms will be available in areas where TB reactors are reported to give the survey a statistical validity, or at least provide sufficient evidence to demonstrate that trace element treatment should be an urgent precautionary measure while further research was pursued. If from a survey of the treated farms it is found there has been no protection against TB, the hypothesis fails the test. If they have been afforded protection, the next step would be for every farm in TB susceptible areas to be similarly protected—preferably by applying trace elements at levels indicated by the soil analysis; or to avoid delay, by an immediate emergency application of Cu, Zn, Co and I, together with sufficient selenium to provide 3gSe/ha for a year. UK soil selenium levels are far too low for there to be any question of toxicity or mineral imbalance as was argued by MAFF last year when they rejected "the mandatory fortification of fertilisers" practised by the Finns (letter to the writer from MAFF's Nutrition Unit, 22 January 1998). The Finns have found no build-up in soil or waters from run-off after extensive monitoring since 1985. The emergency trace element applications should be followed at leisure by individual farm soil analyses and trace element correction including selenium at levels in line with Goodwin-Jones' experience, bearing in mind that selenium should be topped up at least every five years. The money that would have been spent on the badger culling trials could be diverted into a grant for trace element restoration.

12.  CONCLUSION

  The development of microbial resistance to antibiotics and the instability and sometimes ill effects of certain vaccines make it urgent that preventive treatment and research should once more be directed to the build-up of resistance and discovery of its mechanisms. New tools in cell biology, biochemistry, molecular biology and genetics now enable us to study the recognition factors and plethora of defence reactions mobilised against pathogens. A major target of signal-transduction pathways is the nucleus, where such signals can either activate or suppress genes and where the up or down regulation of mRNA may depend on nutritional factors. Since to the writer's knowledge, the thyroid is not known to play a role in the immune system, the proposed coordination of T3 with enzymes active in resistance mechanisms is an area in need of research.

  Nevertheless it is the understanding of mechanisms that is important and the creation of conditions in the organism that enable them to function correctly. Biologists must beware that reactions involved in pathogen defence may have a different function in other cells or other organs and unpredictable consequences can follow attempts to manipulate individual genes or enzymes. Evidence accumulating in intensive plant research suggests that defence mechanisms in plant and animal might rely less on the "altered expression of a few unique, defence-related genes" than on major shifts in metabolic pathways [71]. In many cases all that is required to induce in an animal a resistance response is an armoury consisting in the necessary nutrients, freedom from stress and freedom from pollution.

  TB as has been said is a disease of malnutrition. It may be that all the distress caused to farmers by the loss of their cattle, the cloud hanging over them of draconian restrictions likely to drive them from their livelihood, the welfare implications for the cattle and the fate of the badgers, has a simple nutritional solution. In my view the evidence presented here is overwhelming: Lurie's research into TB resistance and the mechanisms that can be deduced from this early work; the intensive research since the 1970s mainly in America, into trace element function; and the responses in the field obtained by Goodwin-Jones. The threat of a crisis in our agriculture and rural economy is too great—we can no longer ignore it. Farmers would do well, and MAFF would do well to treat susceptible TB farms with an emergency but sufficient application straightaway. There is nothing to be lost from such a policy, and everything to be gained.

13 January 1999

REFERENCES

  1  Kaplan G et al. Tuberculosis comes of age. Molecular mechanisms and immunologic aspects. Keystone Symposium, Keystone, CO, USA 2-8 April 1998. Reported in Mol Med Today 1998; 4: 330-333.

  2  Lurie M B. Role of the thyroid in native resistance to tuberlucosis in. Lurie M B. Resistance to Tuberculosis. Harvard Univ Press 1964 chap XII pp 265-301.

  3  Gallagher J et al. Tuberculosis in wild badgers (Meles meles) in Gloucestershire: pathology. Vet Record 1976; 98: 9-14.

  4  Report in BBC Wildlife 1998; 16(12): 24.

  5  Staggs A J. Vaccines against chlamydia: approaches and progress. Mol Med Today 1998; 4: 166-173.

  6  Goodwin-Jones R. Unpublished farm reports. Trace Element Services, Pencnwc, Abergorlech Road, Carmarthen SA32 7BA.

  7  Wells N. Selenium in horizons of soil profiles. New Zealand J Sci 1967; 10: 142-153.

  8  McGrath A J, Loveland P J. The Soil Geochemical Map of England and Wales. London, Blackie Acad Prof 1992.

  9  Varo P et al. Nationwide selenium supplementation in Finland—effects on diet, blood and tissue levels and health. In. Selenium in Biology and Human Health ed Burk R F NY, Springer-Verlag 1993; pp 199-218.

  10  Barclay MNI, MacPherson A. Selenium contents of what flour used in the UK. J Food Agric 1986; 37: 1133-1138.

  11  MAFF Joint Food Safety and Standards Group. Food Surveillance Information Sheet. Oct 1997 No 126 pp 1/8-6/8.

  12  Committee on Medical Aspects of Food Policy (COMA), Dept of Health. Dietary Reference Intake for Food Energy and Nutrients for the United Kingdom. HMSO 1991 pp 174-177.

  13  Reilly C. Selenium in health and disease: a review. Aust J Nutr Dietetics 1993; 50: 136-144.

  14  Rayman M P. Dietary selenium: time to act. Br Med J 1997; 314: 387-388.

  15  Shortt C T et al. Selenium status in a group of Scottish adults. Eu J Clin Nutr 1997; 51: 400-404.

  16  McLure T J et al. Improved fertility in dairy cows after treatment with selenium pellets. Aust Vet J 1986; 63: 144-146.

  17  Whanger P D et al. Blood selenium and glutathione peroxidase activity of populations in New Zealand, Oregon and South Dakota. FASEB J 1988; 2: 2296-3002.

  18  Reilly C. Selenium in Food and Health. Blackie Acad Prof 1996.

  19  Diplock A T. Indexes of selenium status in human populations. Am J Clin Nutr 199; 57: 2565-85.

  20  Bengoumi M et al. Comparative effects of selenium supplementation on the plasma of selenium concentration and erythrocyte glutathione peroxidase activity in cattle and camels. Animal Sci 1998; 67: 461-466.

  21  Oh SH et al. Dietary selenium requirement of sheep fed a practical-type diet as assessed by tissue glutathione peroxidase and other criteria. J Animal Sci 1976; 42: 984-992.

  22  Sunde RA. Intracellular glutathione peroxidase—structure, regulation and function in Selenium in Biology and Human Health. Burk RF ed NY, Springer-Verlag 1993: pp 47-77.

  23  Behne D, Hofer-Bosse T. Effects of a low selenium status on the distribution and retention of selenium in the rat. J Nutr 1984; 114: 1289-1296.

  24  Arthur JR, Beckett GJ. Roles in selenium in Type I iodothyronine 5-deiodinase and in thyroid hormone and iodine metabilism in Selenium in Biology and Human Health. Burk RF ed NY, Springer-Verlag 1993: pp 95-115.

  25  Hill KE, Burk RF. Selenoprotein P—an extracellular protein containing multiple selenocysteines in Selenium in Biology and Human Health ibid pp 119-131.

  26  Berman G et al. Tissue specific regulation of selenoenzyme gene expression during selenium deficiency in rats. Biochem J 1995; 311: 425-430.

  27  Clark L et al. Effects of selenium supplementation for cancer prevention in patients with carcinoma of the skin. J Am Med Assoc 1996; 276; 1957-1963.

  28  Peretz A et al. Effect of selenium supplementation on immune parameters in gut failure patients on home parenteral nutrition. Nutrition 1991; 7: 215-221.

  29  Simonoff M et al. Plasma trace element levels in 89 HIV infected patients: correlation with nutritional and immunological status in Trace Elements in Man and Animals—TEMA. Anke M et al eds. Gersdorf, Verlag Media Touristik 1998 pp 768-771.

  30  Mantero-Antienz E et al. Selenium status and immune function in asymptomatic HIV-1 seropositive men. Nutr Res 1991; 11:1237-1250.

  31  Ischiropoulos H et al. Macrophage-derived nitric oxide and superoxide reacts to form peroxynitrite in The Biology of Nitric Oxide vol 2 Enzymology, Biochemistry and Immunology. Moncader S et al eds. London, Portland Press 1992: pp 208-209.

  32  Prohaska JR, Lukasewycz A. Effects of copper deficiency on the immune system in Antioxidant Nutrients and Immune Functions. Bendich A et al eds. NY, Plenum Press 1990: pp 123-143.

  33  Erikson RR et al. The role of NADPH—and reduced glutathione-dependent enzymes in the norepinephrine modulation of the ATP-dependent, hepatic microsomal calcium pump: a new pathway for the noradrenergic regulation of cytosolic calcium in the hepatocyte. J Pharmacol Exp Ther 1987; 242: 472-477.

  34  O'Dorisio MS et al. Vasoactive intestinal peptide and neuropeptide modulation of the immune response. J Immunol 1985; 135: 792s-796s.

  35  Barnea A Copper—a modulator of peptide release in the brain in Biology of Copper Complexes. Sorenson JRJ ed Clifton, NJ: Humana Press 1987: pp 81-93.

  36  Grossman CJ Interactions between gonadal steroids and the immune system. Science 1985; 227: 257-261.

  37  Jenkinson SG et al. Effects of copper deficiency on the activity of the seleno-enzyme glutathione peroxidase and on excretion and tissue retention of 75SeO32-. J Nutr 1982; 112: 197-204.

  38  Sorenson JRJ. A physiological basis for pharmacological activities of copper complexes: an hypothesis in Biology of Copper Complexes. Sorenson JRJ ed Clifton, NJ: Humana Press 1987: pp 3-16.

  39  Cook-Mills JM et al Possible roles for zinc in destruction of Trypanosoma cruzi by toxic oxygen metabolites produced by mononuclear phagocytes in Antioxidant Nutrients and Immune Functions. Bendich A et al eds NY, Plenum Press 1990; pp 111-121.

  40  Fraker PJ et al. Zinc deficiency and immune function. Arch Int Med 1987; 147: 1699.

  41  Wong HL, Wahl SM. Inflammation and repair in Peptide Growth Factors and their Receptors II. Sporn MB and Roberts AB eds NY, Springer-Verlag 1990: pp 509-548.

  42  Fraker PJ et al. Delayed type hyper-sensitivity in zinc deficient adult mice. J Nutr 1982; 111: 409.

  43  Bray TM, Bettger WJ. Free Radical Biol Med 1990; 8: 281-291.

  44  Vallee BL, Falchuk KH. Physiol Rev 1993; 73: 79-118.

  45  Cakmak I, Marschner H. Effect of zinc nutritional status on activities of superoxide radical and peroxide scavenging enzymes in bean leaves in Plant Nutrition—from Genetic Engineering to Field Practice. Barrow NJ ed. Kluwer Acad Pubs 1993: pp 133-137.

  46  Vaughan D et al. Physiol Plant 1982; 54: 1253-1257.

  47  Chvapil M. New aspects of the biological roles of zinc: a stabiliser of macromolecules in biological membranes. Life Sci 1973; 13: 1041-1049.

  48  Fullerton H. Soil Nutrition. Nutr Therapy Today 1996; 6 (2): 8-9.

  49  Bettger WJ, O'Dell BL A critical physiological role of zinc in the structure of biomembranes. Life Sci 1981; 28: 1425-1438.

  50  Hirata F, Axelrod J Phospholipid methylation and biological signal transmission. Science 1980; 209: 1082-1090.

  51  Machlin LJ Vitamin E in Handbook of Vitamins 2nd ed Machlin LJ ed NY, Marcel Dekker 1991: pp 99-144.

  52  Moore et al. Int J Biochem Cell Biol Jan 1997.

  53  Hambridge M et al. Zinc in Trace Elements in Human and Animal Nutrition vol 2. Mertz W. ed 5th ed NY, Academic Press 1996 pp 1-137.

  54  Wada LL et al. Fed Proc, Fed Am Soc Exp Biol 1983; 42: 390.

  55  Blokhima RI. in Trace Element Metabolism in Animals vol 1 Mills CF ed Edinburgh, Livingstone 1970: pp 426-423.

  56  Allen DK et al. J Nutr 1982; 14: 665.

  57  Gibson TR, Glembotski CC. Copper-dependent alpha-amidation in the synthesis of neuroendocrine peptide hormones in Biology of Copper Complexes. Sorenson JRJ ed Clifton, NJ: Humana Press 1987: pp 95-111.

  58  Oppenheimer JH, Schwartz HL. Thyroid hormone action at the nuclear level in Thyroid Hormone Metabolism. Hennemann G ed NY, Marcel Dekker 1986: pp383-415.

  59  Seelig S et al. Thyroid hormone attenuates and augments hepatic gene expression at a pre-translational level. Proc Natl Acad Sci USA 1981; 78; 4733-4737.

  60  King EJ et al. Effect of climate on the survival of Mycobacterium bovis and its transmission to cattle herds in the SW of England in Proc 1st Europ Vertebrate Management Conf Univ York 1-3 Sept 1997.

  61  Abrahams EW. "Original mycobacterial sin". Tubercle 1970; 51: 316-321.

  62  Pallen MJ. The immunological and epidemiological significance of environmental mycobacteria in leprosy and tuberculosis control. Int J Leprosy 1984; 54: 231-245.

  63  Stainsby K et al. Antibodies to mycobacteria in healthy and tuberculous badgers from two English counties. Tubercle 1989; 70: 287-292.

  64  Donoghue HD. A longitudinal study of environmental mycobacterias on a farm in SW England.J Appl Microbiol 1997; 82: 57-67.

  65  Gallagher J. Role of infected, non-diseased badgers in the pathogenesis of tuberculosis in the badger. Vet Rec 1998; 142: 710-714.

  66  Rogers LM et al. J Zoology, London 1997; 242: 597.

  67  Cheeseman CL et al. J Appl Ecol 1981; 18: 795.

  68  Griffin JM et al. The association of cattle husbandry practices, environmental factors and farmer characteristics with the occurrence of chronic bovine tuberculosis in dairy herds in the Republic of Ireland. Prev Vet Med 1993; 17: 145-160.

  69  Udris GA. Trace elements in the preventive treatment of bovine tuberculosis. Vet Moscow 1983; 2: 17-20.

  70  Kleene KC. The mitochondrial capsule selenoprotein—a structural protein in the mitochondrial capsule of mammalian sperm in Selenium in Biology and Human Health. Burk RF ed NY, Springer-Verlag 1994: pp 135-149.

  71  Somssich IE, Hahlbrock K. Pathogen defence in plants—a paradigm of biological complexity. Trends Pl Sci 1998; 3:86-89.

---

---

Prepared 27 April 1999