Introduction
This project addresses the urgent matter of
“wasting syndrome” among farmed mule deer.
This syndrome is not to be confused with Chronic Wasting Disease, the
neurological disorder with some similarity to Mad Cow Disease. Wasting syndrome is characterised by chronic
scouring and loss of condition that often results in death. This has hindered growth of a mule deer
industry in
Several explanations, none wholly
satisfactory, have been offered. Deer
producers hold the view that wasting is a greater problem on grass/legume than
bush pasture. There are several possible
explanations:
Mule deer are less stressed on bush pastures
providing some security cover
Browsing offers fewer opportunities for parasite
transmission than does grazing
Grass/legume pastures are less nutritious for mule
deer because:
deer are not morpho-physiologically adapted to high
grass diets
tannins in browse plants may
enhance by-pass protein or otherwise influence digestion
Some work has already been conducted on
pathogenic causes and on morpho-physiological adaptation to diet. Therefore, our study addressed issues related
to the chemical composition of diets, specifically minerals and tannins. Because so many diet combinations and
permutations are possible, our study was guided by two hypotheses:
Mineral
“supportive therapy” hypothesis
A
local deer farmer developed a mineral-fortified supplement that seems to reduce
the expression of chronic wasting in mule deer.
Our hypothesis is that deer do not generally have unusually high mineral
requirements but may benefit from heavy fortification when electrolytes and
minerals are flushed from the animal’s body with chronic scouring. To
address this “supportive therapy” hypothesis, we compared: 1) a
promising new mineral fortified supplement (labelled
Granola because of its physical appearance) gaining widespread use
within the mule deer industry in western Canada
with 2) a
standard complete feed for deer (labelled Ministik because
it is the general herbivore diet at the research station) derived from that used
successfully for several decades at the University of British Columbia. The “supportive
therapy hypothesis” would be supported if beneficial effects of mineral-fortified
diet were more marked among mule deer expressing the syndrome than those that
were not. However, because we did not
experience chronic wasting in the few mule deer that we raised, this experiment
focused on determining whether specialized supplements were needed to keep mule
deer healthy enough to conduct meaningful studies related to the tannin
hypothesis (vide infra).
Tannin
hypothesis
The
tannin hypothesis is based on the observation that diets of farmed deer lack
condensed tannins, which are found at levels exceeding 20% in browse selected
by wild deer. It also is known that
odocoileine deer and many other browsers have proline rich saliva that
complexes tannin thereby reducing its negative effects on rumen
micro-organisms. Our second experiment explored whether dietary tannins
improved the performance of both white-tailed and mule deer and whether chronic
wasting was prevented by tanniferous diets.
We sought to provide nutritional information
needed to formulate suitable diets for farmed deer that would allow the
establishment of a viable mule deer industry and improve the productivity of
farmed white-tailed deer. This report
describes these two experiments as well as a method for assessing biological activity
of condensed tannins and additional information arising from our experience
raising white-tailed and mule deer fawns.
Establishing the Research Herd
We raised fawns in 1998, 1999, and 2000. Unfortunately, few mule deer were available
so a successful special appeal was made in 2000 that will allow us to work with
balanced numbers in the future. Results
if this study are sufficiently encouraging to warrant further investigation.
The 1998 Cohort
Fawns were obtained from across
The 1999 Cohort
In the spring of 1999, additional orphaned
and abandoned white-tail and mule deer fawns were collected by Alberta
Environmental Protection. Mild cases of
diarrhoea appeared from the onset of the project. During the first two weeks mild diarrhoea was
observed in two of the fawns and was attributed to stress and diet
changes. By the third week, fourteen
fawns had been collected and the first serious case of diarrhoea was
encountered. The infected fawn was a male
mule deer. The symptoms progressed for
eight days before the fawn died from emaciation. Treatment was administered throughout the
course of the infection with kaopectate, antibiotics, and electrolyte.
In the weeks following the death of the male
mule deer, several other fawns developed severe diarrhoea and despite all
efforts died. Fecal samples were
collected from three white-tail fawns suffering from the unexplained
outbreak. The lab results that returned
were inconclusive and provided no explanation or cause. A quarantine was placed on existing fawns
until the origin of the infection was determined. Newly arrived fawns were placed in a separate
enclosure and separate bottles, boot dip and personnel restriction was
implemented in order to prevent further infection. Reports from the provincial animal pathology
lab had been negative for Cryptosporidium and other enterotoxic bacteria
until fecal samples from three scouring mule deer fawns and one live but very
sick fawn were analysed. Both the fecal
samples and intestinal samples from the live fawn were revealed Cryptosporidium
using the Kinyoun acid-fast staining method and Sheather’s sugar flotation
method. In the course of the summer,
four fawns were confirmed to be infected with Cryptosporidium in
addition to another twelve undiagnosed fawns that suffered from identical
symptoms and died.
Two tentative conclusions were drawn. The first was that young fawns and those that
were weak upon arriving at the Ranch were most at risk of infection. Secondly, there was a marked difference in
the tolerance of infection between white-tail and mule deer fawns. Mule deer fawns exhibited symptoms for longer
than one week before succumbing to the infection whereas white-tail fawns
generally lived only three to four days after the symptoms were first observed.
Cryptosporidiosis
Diarrhoea in young animals
is most commonly attributed to E. coli, rotavirus, or coronovirus
(Tzipori et al., 1981). However with the
intensification of livestock operations the protozoan parasite Cryptosporidium
is becoming a more common causal agent (O’Donaughue, 1995). C. parvum lacks host specificity
(Nielson and Ward, 1999), and has been reported to infect mammalian, avian, and
reptile species (Taylor and Webster, 1998)).
A common strain, C. parvum has been reported in deer species
(Tzipori et al., 1981), in which neonates and immuno-compromised individuals
are most are most at susceptible (Garber et al, 1994). Symptoms may range from intermittent to
profuse diarrhea, abdominal pains, and dehydration, resulting in stunted growth
and possible death in the more serious cases (Taylor and Webster, 1998). Those animals that survive may develop an
immunity to further clinical episodes.
There is no cure for Cryptosporidium
so prevention is essential for control of the parasite. Fecal-oral contamination is the primary route
of infection (Nielson and Ward, 1999).
Therefore it is valuable to know how long the parasite’s oocysts
continue to be shed after initial infection.
As mentioned fawns and immune depressed individuals are most at risk to
suffer acute infections of Cryptosporidium. However, animals with strong immune systems
may contract and carry the parasite.
Seasonal peaks in shedding during spring and autumn periods have been
recorded (Taylor and Webster, 1998).
These peaks correlate with high stress periods such as fawning and the
rut.
Fecal samples were collected from the
hand-reared fawns in late August after no further clinical signs of Cryptosporidium
were observed. A control group of fecal
samples were collected from white-tail and mule deer at the
All fecal samples were negative for Cryptosporidium
oocysts. This suggests that oocysts are
not shed, or are shed at levels undetectable to the staining methods used. In experiment 1,the fawns that were used in
this experiment were exposed to a Cryptosporidium outbreak one month
prior to fecal sample collection.
Previous research has shown that oocysts of Cryptosporidium are
excreted in the feces of an infected animal for 2 days to several months,
depending on the species and Cryptosporidium strain in question
(O’Donoghue 1995). A modified Kinyoun acid-fast staining technique was used in
correlation with unstained fecal smear examination to determine if oocysts were
present. This method is accepted, however
more sensitive tests are now available.
Such tests include the monoclonal antibody test, PCR test.
Mineral Supportive Therapy Hypothesis
A local mule deer farmer developed a
promising mineral-fortified diet that seems to minimize the expression of scouring
and chronic wasting. Studies on deer and
other wild ruminants does not suggest unusual mineral requirements. However, it is possible that stress and
scouring may increase this requirement [As evidence, the last revision of NRC
Nutrient Requirements for Beef Cattle reports mineral requirements for stressed
and non-stressed animals]. Minerals, in
such cases, could be seen as supportive therapy, replacing minerals lost during
scouring.
Support for this hypothesis would be provided
by finding that mineral fortification improves performance of scouring deer but
not healthy deer. The original plan
could not be executed because chronic wasting did not appear in our research
herd. Therefore, the experiment
addressed the simpler proposition that mule deer need mineral-fortified diets
to survive, as suggested by industry experience at the time our study was
initiated.
Experimental
We compared the performance of mule and
white-tailed deer on the mineral-fortified supplement (Granola) to a
standard herbivore pellet (Ministik) designed as a complete feed. Proprietary considerations prevented analysis
of the ingredients and composition of the Granola diet. The Ministik deer
herbivore diet originally developed at the
In April 1999, deer from the 1998 fawn cohort
were transferred to the Ministik Research Station to begin feeding trials.
Several feeding trials were conducted between May and November, 1999. At the start of the trial there were 24 deer
in total, 6 mule deer and 18 white tail deer.
During the trial, one mule and one white-tail deer died, from causes
unrelated to diets fed or trial procedures.
Deer that did not complete both feeding trials were excluded from
analysis.
Table 1. Ingredients and analysis of the
Ministik herbivore pelleted diet
|
Ingredient |
(kg/tonne) |
|
Dehydrated alfalfa |
255 |
|
Barley |
310 |
|
Wheat shorts (middlings) |
140 |
|
Soybean meal (48% CP) |
85 |
|
Beet pulp |
160 |
|
Molasses |
40 |
|
Trace mineralized salt |
4 |
|
Vitamins A,D,E mix |
2 |
|
Binder/anti-mould |
1 |
|
|
|
|
Analysis
(%) |
|
|
Protein |
16.6 |
|
Neutral detergent fibre |
27.2 |
|
Acid detergent fiber |
16.8 |
|
ME (kcal/g) |
2.8 |
Groups and
treatments was as follows:
From the time they arrived at the Ministik Research
Station from Kinsella on 19 Mar 1999 until the experiment began, mule and
white-tailed deer of the 1998 cohort were fed in a single group on the Ministik
pelleted diet with forage ad libitum.
On 29 March, they were randomly allocated to two 140
ft x 50 ft closely grazed pens containing a few trees, dividing equal numbers
of mule and white-tail deer. Animals receiving the mineral-fortified granola
diet were fed 0.5 kg/head/day concentrate and ad libitum grass-legume
hay. Deer in the Ministik diet group
were fed ad libitum pellets and the same grass-legume hay.
On May 10, each group was further subdivided into
duplicate pens. This reduced social
competition around feeders as well as provided limited replication to test pen
effects.
On June 11, deer were recombined into their original
two groups and given access to bush pasture.
This permitted comparison of supplementation of pasture as well as
grass/alfalfa hay.
On July 21, deer on the granola diet were gradually
returned to ad libitum Ministik pellet diet until the beginning of the
tannin experiment in January 2000. This allowed us to study recovery as
additional evidence of the comparative value of the two feeding programs.
Deer were weighed every two weeks during the
trial using a Western Electronic Balance Platform scale. Fecal samples were also obtained monthly to
assess parasite loads.
Results and Discussion
In this preliminary feeding experiment, no
deer had to be removed for reasons related to diet (i.e. emaciation, hoof
problems/founder). Both deer species
were at similar weights at the beginning and end of the period (Fig. 1). Weight gains of white-tailed deer were the
same irrespective of diet and they seem to have reached their target weight of
60 kg by 1 August. However, mule deer
performed much better on the Ministik complete diet, falling behind both in
individual pens and when the same diets supplemented pasture. When placed on the Ministik diet, mule deer
narrowed the weight difference with catch-up growth. With each nutritional transition, there was a
tendency of mule deer to pause before adapting to the new diet.
Each diet in
pen reps
Figure 1. Weights of white-tailed and mule deer during trials to
compare Granola and Ministik diets.
Mineral-fortified Granola was fed according
to the developer’s instructions at 0.5 kg/hd/d along with forage or pasture and
generally was completely consumed so intakes remained constant for the duration
of the experiment. The Ministik pelleted
complete feed is intended to be offered ad libitum and is suitable as a
sole ration but can be fed along with forage or pasture. Daily intake per head rose to a peak in June
and gently declined as animals met their growth targets. When deer receiving Granola were placed on
the Ministik diet, intake increased and exceeded that of deer that had been
kept on the Ministik diet throughout (Fig. 2). This presumable accounted for their
compensatory growth.
Figure 2. Intake of Granola or Ministik diets expressed per head
per day.
These results indicate that the Ministik base
herbivore ration is a suitable complete feed for deer and mineral fortified
diets are not needed to maintain mule deer that have not developed chronic
wasting syndrome. Although chronic
wasting often appears several years after farm establishment, this usually is
attributed to the removal of pasture browse.
Our deer were held in pens without access to natural browse and no
evidence of chronic wasting was found.
Because deer do not use grass-legume hay
efficiently, it is not surprising that concentrates fed on a limited per head
basis did not support the same level of gain.
But the diet was designed as a supplement so the comparison should be
viewed as one of feeding systems rather than as diets. Complete feeds that do not need to be metered
out daily are much more convenient for producers and certainly for researchers
interested in controlled experimentation.
Limited rations create variation in performance among individuals within
groups because of social dominance and facilitation.
Beyond determining that complete feeds could
be used for mule deer research and offered many important advantages, the
results of the feeding trial have further industry implications. Many deer farmers practice frequent diet
change when they notice poor mule deer performance. Our study suggests that mule deer do not
readily adapt to diet change unlike the more adaptable white-tails.
Tannin Hypothesis
Morpho-physiological adaptation of wild
ruminants has received a great deal of attention largely because of the seminal
work of Hofmann (1989). Cattle and sheep
are grazers adapted to digest fibrous feeds rich in plant cell wall. Because of
this adaptation, they have a significantly slower fermentation and passage rate
than deer, a larger rumen and broad incisor bar disallowing specific foliage
selection. In contrast, odocoileine deer (including white-tail and mule deer)
are browsers, digesting plant cell soluble contents and poorly digesting fibre.
Their narrow mouth and long, mobile tongue allow for specific selection of
forbs and foliage. Deer also have a relatively small rumen with a high
fermentation rate, rapid absorption and turnover requiring frequent cycles of
eating and ruminating.
Browse and other dicotyledonous plants
contain rapidly-fermented carbohydrate accompanied by indigestible lignin. Deer
thrive on browse by using the rapidly-fermented portion and propelling lignin
rapidly through the gut. However, this
adaptive nutritional strategy means that deer do not use grass pastures or hay
efficiently. Work remains to be done on
fibre characteristics but attention is now turning to chemical composition of
plants rather than physical characteristics that determine rates of
fermentation and passage. Plant secondary compounds, particularly tannins and
other polyphenols have attracted attention.
Plant browsed by deer often contain more than
20% condensed tannins. In domestic
ruminants, tannins decrease digestibility and voluntary intake (Panda et al., 1983; Van Hoven, 1984), reduce
liveweight gain (Panda et al.,1983;
Barry, 1985; Mehansho et al., 1987),
wool growth (Barry, 1985) and mineral absorption (Disler et al., 1975; Roy and Mukherji, 1979), and may damage the intestine
(Sandusky et al., 1977) or liver
(Jones and Hunt, 1983). Therefore, studies have explored the adaptive tolerance
of browsers to these negative effects.
One such adaptation is saliva composition. Deer possess the proline-rich salivary
proteins, which bind condensed tannins (Hagerman and Robbins 1993). This tannin-proline complex minimizes tannin
absorption and results in reduced toxicity at high tannin consumption (Barry
and McNabb 1999). Grazers do not produce
proline-rich-proteins (PRP) and therefore are more susceptible to toxic effects
of tannins (Austin et al. 1989, Robbins et al. 1987). As expected, CT tolerance decreases among
ruminant species in the order: deer>goat>sheep>cattle.
Although most attention has been directed to
understanding tolerance to tannins, recent work points to a number of
beneficial effects of tannins at moderate concentrations (e.g. <5% dry matter) (Hoskin et al. 1999; Kaitho et al.
1998; Barry and McNabb 1999; Aerts et al. 1999). Tannins may act as a
natural detergent to reduce bloat (Waghorn and Jones, 1989; Tanner et al., 1995) and may be a suitable
substitute for synthetic anthelmintics (Schrägle
and Müller, 1990; Robertson et
al., 1995; Niezen et al., 1995).
Tannin-protein complexes in the rumen may reduce the fermentation of forage
protein to ammonia, increasing the quantity of protein digested in the small
intestine, thereby augmenting biological value (Barry and Manley, 1984; Waghorn
et al., 1987, 1994; McNabb et al., 1996). For example, improved
production efficiency has been demonstrated with animals on diets having
moderate concentrations of tannins (i.e.
2%-4%), through increased wool growth, liveweight gain, milk yield, and
ovulation rate, while maintaining voluntary intake (Terrill et al., 1992; Wang et al., 1996a, 1996b).
Deer might be expected to respond positively to even
higher levels. This certainly is the
case with
This study was conducted to search for
evidence that deer respond positively to relatively high levels of dietary
tannins (10% dry matter). We worked with
Dr Peter Sporns, a food chemist at the
Experimental
Purification of bark tannins
and feed formulation
Feeding trial
The second feeding experiment involved a
winter trial (January 6 –
Thirteen deer completed both winter and
spring feeding trials and were included in the analysis. Deer were randomly
allocated to four groups to achieve an equal distribution of white-tail and
mule deer, and males and females. These groups were divided into duplicate
subgroups to test pen effects. Control group 1 (C1), control group 2 (C2), and
tannin group 1 (T1) each contained two white-tail deer and one mule deer. Tannin
group 2 (T2) consisted of two white-tail deer and two mule deer. Pens of 40 ft
x 100 ft dimensions were used and contained minimal browse (a few trees and an
insignificant amount of fescue grass).
Weekly pellet weigh-backs were collected,
weighed and subtracted from the overall weekly intake. Pre-trial weights were
obtained on January 4 for all deer. Deer
were weighed on January 18, Feb 1 and every week thereafter using a Western
Electronic Balance Platform scale. Fecal
and urine samples (‘yellow snow’) were collected pre-trial, before the increase
in tannin concentration, and at the end of the trial. These samples were analysed for urea
nitrogen, creatinine, tannin, potassium, and cortisol.
Results and Discussion
Performance on
tannin-enriched diets
Diets containing 5% and 10% condensed tannin
were preferred by our penned deer.
Overall, intake of the tannin diet (TG) was greater than that of the
control group (CG) intake (Fig. 3) even when intake was corrected for body
weight. Intake steadily increased after
the tannin level increased to 10%.
Seasonal factors and weather also probably influenced this increase.
Figure 3. Average intake/head/day of Control (open symbols) and
Tannin (closed symbols) diets.
Mule deer fed supplemental bark tannin at 5
and 10% performed better than mule deer fed a control diet without tannin (Fig.
4). Similarly, performance of white-tail deer fed the tannin diet marginally
improved (but not significantly) in comparison to white-tail deer not receiving
tannin. However, numbers of mule deer
available were low and the result therefore could be spurious. The starting weight of mule deer on the
tannin diet was 6 kg higher than that of mule deer on the control diet.
Therefore, the difference in performance may simply reflect the vigour of
individual animals.
Figure 4. Weight dynamics of white-tailed and mule deer on control
and tannin diets.
Diet digestibilities
Acid-insoluble ash content of feed and feces
collected during periods when supplemental hay and grazing were denied can be
used to estimate digestibility. Using
composite samples, digestibility of the control diet was calculated to be 70%
in agreement with other trials at the
Fecal and urine chemistry
Urinary urea nitrogen, potassium and cortisol
were expressed as ratios to creatinine to correct for urine dilution. Creatinine is constantly produced by lean
tissue and estimates total lean body mass. Urea indicates protein breakdown
from either feed or body tissues. Potassium usually indicates muscle breakdown
and cortisol indicates stress. These
indices are usually interpreted together.
High levels of all three indicators indicate nutritional stress. High urea but low potassium and cortisol
indicate a good nutritional environment whereas low values of all three
indicators indicate an adequate nutritional environment.
Differences between diets and species were
too small to be of biological significance (Fig. 5). Generally, both diets
adequately met the needs of both deer species.
The higher cortisol levels of both white-tailed and mule deer at the end
of the 10%tannin feeding period may warrant further investigation but the
magnitude is so small relative to the wide seasonal variation in wild deer that
it is unlikely to be meaningful (Saltz and White 1991, Saltz et al. 1992,
Servello and Schneider 2000).
Conclusions and Recommendations
The original objectives of this study were
only partially met because of the limited availability of mule deer fawns and
absence of chronic wasting syndrome in the research herd. The following is what we can conclude from
the original questions:
What causes wasting syndrome
in mule deer?
Because chronic wasting syndrome was not
expressed in our research herd, we were unable to conduct definitive studies on
its possible nutritional basis. However,
we gained evidence that chronic wasting is not a necessary consequence of
maintaining mule deer without natural browse.
We also determined that mineral-fortified supplements are not necessary
to prevent expression of the syndrome.
Of course, this statement is made only on the basis of 2 years of study
and chronic wasting often appears several years after establishment of
commercial operations.
Nutritional differences of mule and white-tailed deer
Although the number of mule deer is small, we
saw evidence that mule deer may be more sensitive to diet change than
white-tailed deer. The lag in
performance of mule deer occurs even with subtle diet change and even when the
change has been to a better diet. The
general sensitivity of deer to diet change may warrant the following practical
recommendation:
If a diet is of sufficient quality (e.g., 16%
protein), there is no need to reformulate for different seasons, ages, genders
or physiological status and continuous experimentation may be costly. Deer meet varying nutritional needs by
adjusting voluntary consumption and the proportion of daily intake obtained
from pasture or hay. Unless substantial cost savings can be gained, the best
policy may be to stay with the same compounded feed offered throughout the
year.
Complete feeds offered ad libium
throughout the year are better than supplements that must be offered in
measured daily amounts. More important
that the labor saving is the opportunity it offers animals to establish
distinctive feeding/ruminating cycles and minimize social interactions that
lead to variations in access and individual intakes.
Tannins as feed additives
Aerts et al. (1999) suggested that dietary
condensed tannins may improve animal performance, reduce nitrogen excretion,
prevent bloat and reduce chemical control of parasites. Our study suggested that performance of mule
deer may be improved by tannin but we found no evidence of reduced nitrogen
excretion. Because parasite loads were low, we were not able to evaluate
anti-parasitic effects.
Although our results were not definitive,
they corroborate work on
Future research
The results of this study are suggestive but
not definitive because too few mule deer were available for the study. With the 2000 cohort, we will be able to
create balanced groups needed to obtain definitive comparisons.
In future work, we intend to study the
inclusion of coarse-ground bark rather than tannin extracts because of the high
extraction cost and the expectation that bark fibre is a better alternative to
feed sources such as beet pulp which is inconveniently hygroscopic. We also expect bark tannins to prevent mould
in self-feeders.
As our research herd and parasite populations
build, we will be able to explore the anti-parasitic effects of tannins.
Acknowledgements
This project was mentored and supported by
Drs Hansen (Veterinary Laboratory) and Okine (nutritionist with AFRD). Barry
Irving managed the project from collecting and rearing fawns to analysing data
and informing the industry. Paul Hanson
managed facilities and animals and collected data on weights and feed intakes.
Judy Irving spent many long hours collecting fecal samples. Kami Swanson and Francine Morley raised fawns
and participated fully in the experiments. They took aspects of this study to
fulfil the requirements for a directed studies course (ANSC 400). Noble Donkor reduced tannin extracts to a
fine brown powder. Dr Jay Gedir developed and conducted the tannin assays.
Assistance was provided by Alberta
Environmental Protection who granted the University ownership of orphaned deer
fawns. Financial support for this
research was provided by Alberta Agriculture, Alberta White-tailed and Mule
Deer Association, and Buchanan Lumber.
REFERENCES
Aerts
R.J., Barry T.N. and McNabb W.C. 1999.
Polyphenols and agriculture: beneficial effects of proanthocyanidins in
forages. Agriculture, Ecosystems and
Environment 75: 1-12.
Barry, T.N. and Manley,
T.R. 1984. The role of condensed tannins in the
nutritional value of Lotus pedunculatus for sheep. 2.
Quantitative digestion of carbohydrates and proteins. Br. J. Nutr.
51: 493-504.
Barry, T.N. 1985.
The role of condensed tannins in the nutritional value of Lotus pedunculatus for sheep. 3. Rates
of body and wool growth. Br. J. Nutr.
54: 211-217.
Barry
T.N. and McNabb W.C. 1999.
The implications of condensed tannins on the nutritive value of temperate
forages fed to ruminants. Br. J. Nutr. 81:248-254.
Disler, P.B., Lynch, S.R.,
Charlton, R.W., Torrence, J.D., Bothwell, T.H.,
Garber, L.P., M.D. Salman, H. S. Hrd, T. Keefe, and J. L.
Schlater. 1994. Potential risk factors for Cryptosporidium
infection in dairy calves. JAVMA
205:86-91.
Hagerman, A.E. and C.T.
Robbins. 1993. Specificity of
tannin-binding salivary proteins relative to diet selection by mammals.
Hofmann.
R.R. 1989. Evolutionary
steps of ecophysiological adaptation and diversification of ruminants: a
comparative view of their digestive system.
Oecologia 78: 443-457.
Hoskin,
S.O.,
Jones, T.C. and Hunt, R.D.
1983. Veterinary Pathology (5th
Edition). Pp. 161-168. Lea and Febiger,
Kaitho
R.J., Umunna, N.N., Nsahlai I.V., Tanmminga S. and Van Bruchem J. 1998. Utilization of browse supplements with varying
tannin levels by Ethiopian Menz sheep.
Agroforestry Systems 39:
145-159.
McNabb, W.C., Waghorn, G.C.,
Peters, J.S. and Barry, T.N. 1996. The effect of condensed tannins in Lotus
pedunculatus on the solubilization and degradation of
ribulose-1,5bisphosphate carboxylase (EC 4.1.1.39; Rubisco) protein in the
rumen and the sites of Rubisco digestion.
Br. J. Nutr. 76: 535-549.
Mehansho, H., Ann, D.K.,
Nielson, C.K. and Ward, L.
A. 1999.
Enhanced detection of Cryptosporidium parvum in the acid-fast
stain. J. Vet. Diagn. Invest.
11:567-569.
Niezen, J.H., Waghorn, T.S.,
O’Donoghue, P. J. 1995. Cryptosporidium and cryptosporidiosis
in man and animals. Int. J. Parasitol.
25:139-195.
Panda, S.K., Panda, N.C. and Sahue, B.K. 1983. Effect of
tree leaf tannin on dry matter intake by goats. Indian Vet. J. 60: 660-664.
Perryman, L. E. and.
Bjorneby, J. M. 1991. Immunotherapy of Cryptospoidiosis in
immuno-deficient animal models. J.
Protozool. 38:98S-102S
Robbins,
C.T., Mole S., Hagerman, A.E. and Hanley, T.A. 1987. Role of tannins in defending plants
against ruminants: reduction in dry
matter digestion? Ecol. 68:
1606-1615.
Robbins, C.T., Hanley, T.A.,
Hagerman, A.E., Hjeljord, O., Baker, D.L., Schwartz, C.C. and Mautz, W.W. 1987.
Role of tannins in defending plants against ruminants: reduction in protein availability. Ecol. 68:
98-107.
Robertson, H.A., Niezen,
J.H., Waghorn, G.C.,
Saltz, D. and White,
G.C. 1991. Urinary cortisol and urea nitrogen responses
to winter stress in mule deer. J. Wildl.
Manage. 55 (1): 1-16.
Saltz, D. and G.C. White et
al. 1992. Urinary cortisol, urea nitrogen excretion,
and winter survival in mule deer fawns.
J. Wildl. Manage. 56 (4): 640-644.
Schragle, R. and Muller,
W. 1990.
The influence of selected tannin-containing plant species on the
tenacity of pathogenic bacteria in an in vitro rumen system. J. Vet. Med. B 37: 181-186.
Servello, F.A. and
Schneider, J.W. 2000. Evaluation of urinary
indices of nutritional status for white-tailed deer: Tests with winter browse
diets. J. Wildl. Manage. 64: 137-145.
Tanner, G.J., Moate, P.J.,
Taylor, M. A. and Webster,
K. A. 1998. Recent advances in the diagnosis in livestock
of Cryptosporidium, Toxoplasma, Giardia and other protozoa of veterinary
importance. Res. Vet. Sci. 65:183-193.
Terrill, T.H.,
Tzipori, S., Angus, K. W.,
Van Hoven, W. 1984. Tannins
and digestibility in greater kudu.
Waghorn, G.C., Ulyatt, M.J., John, A. and
Fisher, M.T. 1987. The effect of condensed tannins on the site
of digestion of amino acids and other nutrients in sheep fed on Lotus
cornicalatus L. Br. J. Nutr.
57: 115-126.
Waghorn, G.C.,
Wang, Y.,
Wang, Y.,
Appendix 1. Quantification
of bioactive tannins
The absence of an appropriate analytical
technique for quantifying tannins has hindered investigations into
tannin-protein associations in ruminants. Past methods, although precise, often
involve elaborate sample preparation and tedious analytical procedures (e.g. spectrophotometric assay of
dye-labeled protein (Asquith and Butler 1985), precipitation of
ribulose-1,5-bisphosphate carboxylase/oxygenase by tannic acid (Martin and
Martin 1983), precipitation of iodine-125 labelled bovine serum albumin (BSA)
(Hagerman and Butler, 1980)). These extraction and quantification techniques
have only been used to measure tannins in plant material, and attempts to
extract tannins from feces have demonstrated little success. Our objective was
to modify Hagerman’s (1987) radial diffusion assay method for quantifying
tannins in plant extracts, and develop and improved protocol for extracting and
quantifying tannins in cervid feed and feces.
Methods
Tannin
Extraction. Samples of feed or feces were freeze-dried for 48 h at –60°C and ground through a 20-mesh screen in a
Wiley mill. Lyophilized samples (~500-600 mg feed/feces, ~200 mg tannin
standard) were dissolved in 5 ml ‘extraction solvent’ (ES) (i.e. either 70% (v/v) acetone or 50%
(v/v) methanol), vortexed several times over 20 min., centrifuged (5 min. @
3,000 r.p.m.), and supernatant transferred to a 15 ml plastic centrifuge tube.
In total, three extractions were performed. The supernatant (~15 ml) was
evaporated dry using a Multivap Nitrogen Evaporator. Pellets were resuspended
in 3 ml ‘precipitation solvent’ (PS) (i.e.
either 70% acetone or 50% methanol), Vortexed several times over 30 min., and
centrifuged (5 min. @ 3,000 r.p.m.).
Following the third extraction, after
removing the supernatant from the tube containing tannin standard, the residue
was air-dried and weighed. Also, following addition of tannin standard to gel
plate wells, solution was centrifuged (5 min. @ 3,000 r.p.m.), supernatant
removed, and residue air-dried and weighed. Residue weights were cumulated and subtracted
from tannin standard weighed out to correct for impurities.
Gel Preparation. Gel plates were
prepared according to Hagerman (1987). A solution of acetic acid and ascorbic
acid was mixed, forming a solution of pH 2.8. This was adjusted to pH 5.0 by
adding NaOH, as this has been reported as the optimum pH for BSA precipitation
(Hagerman and Klucher, 1986). Solution was heated to boiling while stirring, as
1% (w/v) agarose was added. Solution was cooled to 45°C and 0.1% (w/v) BSA
added, while stirring. Aliquots of 9.5 ml were added to 8.5 cm diameter Petri
dishes on a level surface (to obtain a gel of uniform thickness) and cooled.
Gel plates were stored at 4°C to prevent bacterial
growth.
Radial Diffusion
Assay. Wells of a uniform 5.0 mm diameter were punched in the gel plates,
spaced 1.5 cm apart. Using a micro-pipette, prepared solutions were added to
wells (in duplicate) in 10 l
aliquots, being the well capacity. As the solution is absorbed by the gel,
several successive 10 l
aliquots were added, depending on the dilution of the solution. Wells were
rinsed with successive 10 l
aliquots of PS (i.e. either 70%
acetone or 50% methanol) to ensure all tannin is available to complex with BSA.
Petri dishes were covered and sealed with Parafilm, and incubated at 30°C for
24 h to 120 h.
Ring Measurement. Following
incubation, perpendicular ring diameters were measured to account for
non-uniform ring development (rare) using micro-calipers (actual size) and a
standard mm ruler (enlarged copies). Tannins were quantified by squaring the
mean of the perpendicular diameters (hereafter referred to as “area”) and
substituting this value into a calibration equation. This is the best-fit
equation when ring area (mm2) is regressed against known quantities
of tannins from feed and tannin standard (mg).
Results
The incubation period required for
precipitation rings to achieve equilibrium (i.e.
maximum size) was 24 hours for the range of tannin weights used in this study.
This time will vary according to quantity of tannin, with longer incubation
time required for higher tannin concentrations in wells.
Precipitation ring area (D2) (mm2)
demonstrated a linear relationship with tannin content in wells (mg) (Fig. 1).
The detection threshold for accurate measurement of precipitation rings is
approximately 0.5 mg. It is
recommended that tannin applied to wells does not exceed 4.0 mg, as ring overlap precludes accurate
measurement. This technique for extracting condensed tannins from cervid feces
worked well, as fecal samples from animals on tannin diet were within the
expected relative magnitudes (Table 1).
Removal of non-tannin residue from bark
extract after the tannin extraction procedure revealed a purity of 75±5%. This
impurity was accounted for when determining standard calibrations curves.
Methanol (50%) was a more effective
precipitation solvent than acetone (70%). Methanol demonstrated greater
specific activity (p<0.05). When equal quantities of tannin were
dispensed in wells, methanol formed rings with 19±7% greater area than acetone.
Methanol also exhibited superior predictive capabilities to acetone for
estimating tannin content. All
measurements of ring area included the area of the well. At 200% enlargement,
the diameter of the well is 10 mm, and therefore the well area (D2)
is calculated as 100 mm2. This suggests that that the best-fit
equation for perfect predictability when ring area (mm2) is
regressed against tannin weight (mg), would have a y-intercept of 100. The best-fit regression equation when 50%
methanol was incorporated as the PS, had a y-intercept
of 97.3 (area = 97.3 + 43.0(tannin weight), R2 = 0.98, p<0.001),
which is very similar to that expected. Furthermore, precipitation rings with methanol
PS had a distinct, clearly visible outer ring, facilitating easy, rapid
measurement, while rings with acetone PS faded towards the outer reaches.
Measuring ring diameter with a standard mm
ruler on a photocopy enlarged 200% provided the most precise estimate of ring
area (R2 = 0.98). Accuracy of microcaliper measurements did not
overcome difficulty in visualizing the outer boundary of precipitation rings
measured directly on the plate nor on an actual size photocopy (R2 = 0.89).
Discussion
Radial diffusion assay reflects the ability
of tannins to precipitate proteins. It is derived from the principle of radial
immunodiffusion (e.g. Vaerman 1981),
in that tannin-protein interactions are very similar to antigen-antibody
reactions. For tannin quantification, more ecologically significant gel
proteins could be substituted for BSA, however, Hagerman (1987) selected BSA
for its homogeneity, solubility, and relatively low cost. The simplicity of the
technique and lack of complex reagents involved enhances reproducibility.
An advantage of the radial diffusion assay
method is its specificity for either hydrolyzable or condensed tannins. Other
tannin quantification techniques that depend on functional groups have been
unreliable, as these functional groups are often not unique to tannins (e.g. Folin-Denis assay (Burns 1963),
Prussian blue (Price and Butler 1977)). Our samples which did not contain
tannins, did not form precipitation rings. This indicates that the method is
highly specific to condensed tannins and experiences no apparent interference
from other phenolic compounds.
The 24 h incubation required for
precipitation rings to reach equilibrium using this assay method is shorter
than Hagerman (1987) recorded with a similar technique. Hagerman found a 0.5 mg
aliquot of tannic acid (a hydrolyzable tannin) took 48 h to reach equilibrium,
while 1.0 mg required 96 h. Within 24 hours, precipitation rings for the entire
range of tannin weights analyzed in this study (0.5-4.0 mg) achieved maximum
diameter. This is convenient in that many analytical runs may be completed
within one week. The 70% acetone PS exhibited a faster diffusion rate through
the BSA gel than 50% methanol, however, this is likely due to the greater
precipitation activity of methanol.
The tannin detection threshold of our
technique (0.5 mg) exhibited lower sensitivity than that described in Hagerman
(1987) (0.025 mg). This should not present a problem, as difficulties of tannin
detection in dilute samples can be obviated by continued addition of aliquots
to wells until suitable measurements may be obtained. The same applies for
difficulties encountered with ring overlap occurring with high concentration
solutions (i.e. add lesser volumes to
wells). Adjustments in volumes of solution dispensed in wells will provide
optimal ring sizes for accurate measurement.
Acetone is considered the most efficient
solvent for extracting tannins (Fletcher et
al., 1977). Although, Hagerman (1987) found no difference between specific
activity of 50% acetone and 50% methanol, acetone as a PS has been known to
inhibit protein precipitation (Hagerman and Robbins 1987). The combination
of 70% acetone for extracting and 50% methanol for precipitating tannins,
provided optimal tannin extraction and BSA precipitation. Furthermore, the
clearly defined outer ring produced from methanol (i.e. versus faded outer edges with acetone-precipitated rings)
facilitates rapid, accurate measurement.
Selection of the standard is paramount to the
success of tannin assays. Nelson et al. (1997) noted that errors can arise from
an incorrect choice of external standard, particularly when it differs from the
tannin being analyzed. Structural variation among condensed tannins can
introduce error when using assays for quantification. In this study, we had the
advantage of the standard being identical to the analyzed tannin, which likely
contributed to the superior predictive capabilities demonstrated.
Although the bark extract employed as
standard in this study contained contaminants, removal of impurities during
sample tannin extraction was very simple. In order to reliably predict tannin
quantities extracted from fecal samples, for every batch it is necessary to
have duplicate standard reference wells on each plate, as well as producing
precipitation rings from a suitably wide range of weights of tannin standard.
From this, the simple regression calibration equation is developed, by
examining precipitation ring area (mm2) in relation to known
quantities of standard in each well. Several methods reviewed in Maxson and
Rooney (1972) had significant variation in standard curves between days and/or
laboratories. This highlights the importance of preparing individual standards
for each analytical run. This ensures that samples and standards are prepared
and run under identical conditions. Although variation in calibration curves
among our batches was negligible, each batch should be restandardized with
purified tannin.
The unique, yet very simple technique of ring
measurement developed in this study, provides an accurate and rapid method for
determining ring area. Assay methods used in the past have depended on
colorimetric measurement of absorbance using spectrophotometric laboratory
equipment (e.g. two-dimensional
thin-layer chromatography (Peng and Jay-Allemand 1991), dye-labeled protein
assay (Asquith and Butler 1985), vanillin-HCl assay (Price et al. 1978), Prussian blue (Price and Butler 1977), 1-butanol-HCl
(Porter et al. 1986)). This ‘gel
plate photocopy’ technique is convenient in that laboratory instruments are
unnecessary. In fact, gel plates can be photocopied immediately following
incubation, and measurement carried out anywhere and at anytime thereafter. As
well, photocopies can be re-accessed later, if further analysis is desired.
Perhaps, to reduce possible measurer bias, photocopies could be digitized and
computer-analyzed.
Hagerman (1987) found the radial diffusion
assay method for tannin quantification to be simple, sensitive, and specific,
facilitating analysis of large numbers of samples. Complex reagents and
elaborate instruments were not necessary, and there was no apparent
interference from non-phenolic compounds. Our results suggest several
modifications to Hagerman’s (1987) technique, providing a very convenient
method with augmented precipitation specific activity and enhanced accuracy and
precision of tannin quantification. Moreover, our tannin extraction method was
thorough, overcoming past difficulties in obtaining suitable extractions from feces.
Further research into radial diffusion assay
should attempt to account for the presence of bound tannins in feces. Our
technique only has the ability to detect unbound tannins. Feces may contain
tannins that have formed associations with proteins while passing through the
gastrointestinal system. Phenol is a very strong solvent, which dissociates
tannin-protein complexes (Hagerman and Butler, 1980). Perhaps simultaneous
comparison of precipitation rings from tannin solutions that have been exposed to
phenol fractionation, to those that have not, will reveal ratios of unbound to
bound proteins present. Furthermore, effects of antioxidants on specific
activity of protein precipitation should be investigated. Pend and Jay-Allemand
(1991) demonstrated an increase of 30% and 75% when ascorbic acid and sodium
metabisulfite, respectively, were added to the extraction solvent. The addition
of diethyldithiocarbamic acid to the agarose gel resulted in an increase in
specific activity of 20%.
References
Asquith, T.N. and
Burns, R.E. 1963. Methods of tannin analysis
for forage crop evaluation.
Fletcher, A.C., Porter, L.J., Haslam, E. and Gupta, R.K. 1977. Plant
proanthocyanidins. Part 3. Conformational and configurational studies of
natural proanthocyanidins. J. Chem. Soc.
Perk. I 1977: 1628-1637.
Hagerman, A.E. 1987. Radial
diffusion method for determining tannin in plant extracts. J. Chem. Ecol. 13: 437-449.
Hagerman, A.E. and
Hagerman, A.E. and Klucher, K.M. 1986. Tannin-protein interactions.
Pp. 67-76. In Plant Flavanoids in Biology
and Medicine: Biochemical, Pharmacological, and Structure-Activity
Relationships. V. Cody, E. Middleton, and J.B. Harbone (Eds.). Alan R.
Liss,
Hagerman, A.E. and Robbins, C.T. 1987. Implications of soluble
tannin-protein complexes for tannin analysis and plant defense mechanisms. J. Chem. Ecol. 13: 1243-1259.
Martin, J.S. and Martin, M.M.
1983. Tannin assays in ecological studies: Precipitation of
Ribulose-1,5-bisphosphate carboxylase/oxygenase by tannic acid, quebracho, and
oak foliage extracts. J. Chem. Ecol.
9: 285-294.
Maxson, E.D. and Rooney, L.W. 1972. Evaluation of methods for tannin
analysis in Sorghum grain. Cereal Chem.
49: 719-729.
Nelson, K.E., Pell, A.N., Doane, P.H., Giner-Chavez, B.I. and
Schofield, P. 1997. Chemical and biological assays to evaluate bacterial
inhibition by tannins. J. Chem. Ecol.
23: 1175-1194.
Peng, S. and Jay-Allemand, C. 1991. Use of
antioxidants in extraction of tannins from walnut plants. J. Chem. Ecol. 17: 887-896.
Porter, L.J., Hrstich, L.N. and Chan, B.G.
1986. The conversion of procyanidins and prodelphinidins to cyanidin and
delphinidin. Phytochem. 25: 223-230.
Price, M.L. and
Price, M.L., Van Scoyoc, S. and
Vaerman, J.P. 1981. Single radial immunodiffusion. Methods Enzymol. 73: 291-305.
Fig 1.
Comparison of precipitation solvents for quantifying tannins. Ring area
(mm2) (measured from 200% enlargement) is regressed against
known tannin weight (mg). Solid line represents 50% aqueous methanol (ê/â) (y
= 97.3 + 43.0x, R2=0.98, p<0.001). Dashed line represents
70% acetone (ë/ã) (y
= 56.0 + 39.4x, R2=0.98, p<0.001). Closed symbols
represent tannin standard bark extract from Populus tremuloides, while open symbols represent tannin
extracted from cervid feed. Each point is the mean of duplicate ring
areas.
Figure 1. Comparison of precipitation solvents for quantifying
tannins. Ring area (mm2) (measured from 200% enlargement) is
regressed against known tannin weight (mg). Solid line represents 50%
aqueous methanol (/) (y
= 97.3 + 43.0x, R2=0.98, p<0.001). Dashed line represents
70% acetone (/) (y
= 56.0 + 39.4x, R2=0.98, p<0.001). Closed symbols
represent tannin standard bark extract, while open symbols represent tannin
extracted from cervid feed. Each point is the mean of duplicate ring areas.
