Aquaculture Nutrition Research

1
Aquaculture Nutrition Research
Largely from DAbramo, et al.
2
Introduction

• Besides the constraints imparted on successful
  aquaculture by site variance, nutritional
  research also shows substantial shortcomings
• this affects the interpretation and comparison of
  results

3
Introduction

• virtually no standardization exists in the
  industry today
• major areas requiring standards
• experimental design
• diet formulation
• production techniques
• type and purity of feed ingredients
• culture conditions
• analytical methodologies

4
Introduction

• Several authors have made suggestions
• some of the first were made by New (1976), but
  have had little impact
• he suggested the development of a standard
  reference diet
• these semi-purified diets were largely developed
  using crustaceans tissues as protein sources
• species Homarus americanus, Panulirus argus,
  various penaeids

5
Introduction

• Part 1 Experimental Culture Systems
• Part 2 Selection of Experimental Organisms
• Part 3 Experimental Diets
• Part 4 Evaluation of Nutrient Requirements
• Part 5 Recommended Guidelines

6
I. Experimental Culture Systems

• Water Source and Quality
• first step to proper research
• holding system must not interfere with the
  response to the nutritional factor under
  investigation
• water supplied might profoundly affect results
• a major decision in planning

7
I. Experimental Culture Systems

• Water Source and Quality (continued)
• contaminants (pesticides, heavy metals, ammonia,
  nitrite) can cause stress and depress growth
• also avoid high levels of chlorine , dissolved
  iron, hardness and alkalinity
• avoid high levels of suspended or dissolved
  nutrients in water (they could serve as
  sufficient source of nutrition)
• you would fail to recognize deficiency response

8
I. Experimental Culture Systems

• Water Source and Quality (continued)
• growth enhancing effect of pond water well
  established (Moss, 1992)
• ultimately causes substantial variation in the
  level of treatment
• must include filtration and, often, UV
• summary no standardization of water leads to
  erroneous conclusions

9
I. Experimental Culture Systems

• Experimental Conditions
• physiochemical factors affecting response
  variables include temp, salinity, photoperiod
• need to be controlled within normal criteria for
  growth, monitored, reported
• probably should be standardized for each
  developmental stage of each species
• goal physiologically optimal conditions

10
I. Experimental Culture Systems

• System Design
• flow-through systems usually assure constancy of
  chemical composition
• that is, unless external conditions fluctuate
  substantially
• recirc systems are acceptable, but require more
  careful and frequent monitoring

11
I. Experimental Culture Systems

• System Design (continued)
• must also minimize intraspecific behavioral
  interactions
• these can produce differential growth rates
• solution individual containers (Port A)
• water should not have come in contact with other
  individuals (lobsters, Hedgecock and Nelson,
  1978)
• caveat could be required in some cases
  (schooling species)

12
I. Experimental Culture Systems

• System Design (continued)
• benefits to housing animals individually include
• reduced cannibalism (molt and die)
• increased sample sizes (each indivi-dual is
  considered a single, independent observation
• large sample size is required for those species
  that exhibit a high degree of variation in size
• growth, molting, behavioral responses can be
  observed and recorded for the duration of the
  experiment

13
I. Experimental Culture Systems

• System Design (continued)
• size of culture containers is also critical
• most species exhibit density-dependent growth
  (even for one individual per container)
• weight increase is influenced by surface area and
  volume of container
• can reduce molt frequency (growth)
• may wind up with high growth rate in tanks with
  higher mortality!!

14
II. Selection of Experimental Organisms

• Must know the feeding history of animals to be
  used
• inconsistent results increase when not known
• often resolved by feeding a preconditioning
  control diet, prior to study
• length of period depends on type of nutritional
  investigation
• faster growing juveniles need less conditioning
• conditioning less when studying nutrients that
  are readily metabolized (COH, AAs, etc.)

15
II. Selection of Experimental Organisms

• Preconditioning period increases the probability
  of obtaining well-defined responses to different
  levels of a dietary nutrient
• in general, animals of similar size
  (weight/length) should be chosen
• however, might be misleading depends on history
  of disease, nutrition, etc.
• small shrimp are not necessarily the same age
  (e.g., RDS)

16
III. Experimental Diets

• Standard Reference Diets
• not all species have been that well researched
  (crustaceanltfishltterrestrials)
• most work on small terrestrials gives insight
  into research on aquatics
• SRDs for mice were only established 25 years ago
• differences in nutrition of animals can result in
  up to 5-fold differences in LC50 data for some
  species/toxicants (Bengston et al., 1985)

17
III. Experimental Diets

• Standard Reference Diets
• SRDs have been developed due to a need to
  minimize variation in ingredient composition of
  nutritional test diets
• SRD1 vitamin-free casein (Conklin et al., 1983)
  ? reasonable growth
• SRD2 crab protein concentrate (Castell et al.,
  1989b) ? reasonable growth
• TAMU-SMP reference diet

18
Effect of Lecithin on Cholestrol in Lobster (Baum
et al., 1990)

• Ingredient dry weight
• casein 31.0 0.0 31.0 0.0
• egg albumin 4.0 0.0 4.0 0.0
• isolated crab protein 0.0 35.0 0.0 35.0
• soy lecithin (refined) 10.0 10.0 0.0 0.0
• cellulose 12.2 12.2 22.2 22.2
• corn starch 24.0 same same same
• lipid mix 6.0 ? ? ?
• wheat gluten 5.0
• vitamin mix 4.0
• mineral mix 3.0
• cholestrol 0.5
• vit E 0.1
• vit A 0.1
• vit D 0.1

19
Requirement for Lysine by Penaeus vannamei (Fox
et al., 1992)

• Ingredients Experimental Diets ( lysine)
• 1.50 1.90 2.10 2.30 2.50 3.00
• wheat gluten 31.15 25.37 21.86 18.36 14.86
  6.10
• lysine gluten 0.00 5.73 9.20 12.66 16.13 24
  .79
• fish meal 27.00 27.00 27.00 27.00 27.00 27.00
• wheat starch 27.51 27.56 27.60 27.64 27.67 27.77
• menhaden oil 1.66 1.66 1.66 1.66 1.66
  1.66
• cholestrol 0.25 0.25 0.25 0.25 0.26
  0.25
• lecithin 0.50 0.50 0.50 0.50 0.50
  0.50
• cellulose 2.74 2.74 2.74 2.74 2.74
  2.74
• diatomaceous earth 2.69 2.69 2.69 2.69
  2.69 2.69
• ascorbic acid 0.50 0.50 0.50 0.50
  0.50 0.50
• vitamin premix 2.00 2.00 2.00 2.00
  2.00 2.00
• mineral premix 4.00 4.00 4.00 4.00
  4.00 4.00

20
III. Experimental Diets

• Composition
• should contain as few ingredients as possible
• minimizes availability and compositional change
  issues (I.e., SRD diets must be reproducible)
• must use purified diets (composed primarily of
  refined ingredients)
• ingredient nomenclature must follow published
  guidelines (International Union of Nutritional
  Sciences, 1978)

21
III. Experimental Diets

• Composition
• caveat use of ingredients with greater than 99
  purity is cost prohibitive (even for research)
• however, more pure more control
• changing levels of supplemented micronutrients by
  changing levels of feed ingredients leads to
  concomitant alteration in levels of
  micronutrients

22
III. Experimental Diets

• Preparation
• careful attention must be paid when mixing
  dietary ingredients
• factors particle shape, density, static charge,
  hygroscopicity, adhesiveness
• satisfactory mixing time required for all diets
  (affected by quality of ingredient and mixer)
• we use V-type mixers

23
III. Experimental Diets

• Preparation
• main concern generation of heat followed by
  loss of heat labile nutrients
• cold extrusion rather than steam pelletizing or
  extrusion is preferred
• air drying in forced air oven at low temps also
  preferred
• best drying freeze drying

24
III. Experimental Diets

• Binders
• as mentioned, crustaceans find food by
  chemoreception, not sight
• some are real picky and may not consume feeds for
  hours (rarely)
• somehow stability must be standardized
• DAbramo and Castell (1997) recommend 16 hours
  stability
• glutens appear useful, but are pretty bad in
  terms of leaching of CAAs, ascorbic acid

25
III. Experimental Diets

• Interactions
• regards the possible effect of each ingredient
  on bioavailability of other nutrients in diet
• alginate binds with phytic acid and reduces
  availability of calcium
• zinc bioavailability reduced by calcium,
  phosphorus, phytic acid
• some COHs affect uptake of proteins and amino
  acids

26
III. Experimental Diets

• Digestibility Studies
• as mentioned, digestibility determinations are
  undertaken primarily with Cr2O7
• may not pass homogenously through gut due to
  sequestering (probably in HP)
• Does concentration in fecal material increase
  with time? If so, you have a problem.
• Other suggestion check each ingredient or
  nutrient source being tested at 15 and 30
  levels (remainder is a base formulation)

27
III. Experimental Diets

• Digestibility Studies
• all animals should be fed a conditioning diet for
  one week prior to dig. studies
• remove all feces prior to start
• feed, remove uneaten feed after one hour
• do this for three days and then pool

28
III. Experimental Diets

• Processing and Storage
• identical formulations and feed ingredient
  sources do not guarantee that diets will have
  identical nutritional value
• ingredient particle size, grinding, mixing,
  processing technique, temperature, moisture
  content affect nutritional value
• conditions and length of storage do also
• 44-66 loss of vit C occurs as a result of high
  temp processing (1/2 this via cold processing)

29
IV. Evaluation of Nutrient Requirements

• Methods of Evaluation
• studies should be repeated at various times to
  confirm reproducibility
• proves experimental design
• indices weight gain, survival, tissue levels,
  biochemical indices
• minimum of 4 dietary levels of the nutrient being
  studied

30
IV. Evaluation of Nutrient Requirements

• Methods of Evaluation
• less than 4 levels means curve fitting is
  difficult
• level of nutrient must be confirmed in each diet
  (i.e., dont just calculate it!)
• also check out actual availability of the
  nutrient to the organism
• measured dietary requirements generally assume
  100 availability

31
IV. Evaluation of Nutrient Requirements

• Methods of Evaluation
• water soluble nutrients leach, especially when
  feed particles are handled over long periods of
  time
• possible impact of leaching on nutrient levels is
  virtually impossible to quantify
• nutrients must eventually be provided in water
  insoluble form (e.g., microencapsulation)

32
IV. Evaluation of Nutrient Requirements

• Experimental Period
• how long should be ascertained in advance via
  tissue deposition studies
• without this, you dont know if levels are higher
  or lower than when you started
• if you have high levels at the start, you may not
  ever see a significant increase
• if period is too short, you may never see
  deficiency signs

33
IV. Evaluation of Nutrient Requirements

• Comparison of Growth Rate Estimates
• growth and survival are typical indices
• for crustaceans, the time of molting represents
  very rapid weight gain
• intermolt relatively slow growth
• point weight gain analysis in inverts much more
  discontinuous, complex than verts
• you not only have to compare final weight, but
  also initial weight (solution ANCOVA)
• ANCOVA tests possible affect of initial weight
  on results, initial weight is the covariate
• Did initial weight have a significant influence
  on final weight???

34
IV. Evaluation of Nutrient Requirements

• Comparison of Growth Rate Estimates
• other methods IGR (instantaneous growth rate),
  includes initial weight in growth calc
• IGR (weight gain over time period/initial
  weight) x 100 all divided by duration of time
  period (usually days)
• IGR percent weight gain/time unit
• also useful with crustaceans to measure weight
  gain only for intermolt individuals

35
IV. Evaluation of Nutrient Requirements

• Survival
• another typical index used however, highly
  influenced by environmental stress
• death can occur just due to handling
• at onset of trial, replace dead for a period of
  one week
• replacements should have been fed same control
  diet and be acclimated to system
• must also consider frequency of handling

36
IV. Evaluation of Nutrient Requirements

• Survival
• for crustaceans, sampling throughout the feeding
  trial is virtually impossible
• the cause of mortality might not be evident
• need to check HP in shrimp, see if hepatosomatic
  index is normal
• survival of less than 80 should be suspect
• have a look at Japanese studies in the 80s

37
IV. Evaluation of Nutrient Requirements

• Statistical Analysis
• an expression of the degree of variation of
  measured responses for each dietary treatment
  within an experiment should be presented
• standard error (SE) is usually the appropriate
  statistic for comparing variation among
  treatments
• if error variances are homogenous, then the
  degree of variation should be expressed as pooled
  standard error for all experimental animals

38
IV. Evaluation of Nutrient Requirements

• Statistical Analysis
• as mentioned, survival should not be less than
  80 in any treatment
• less than 80 reduces the number of observations,
  reduces accuracy of statistical analysis
• coefficient of variation should be less than 15
• higher values could yield incorrect conclusions

39
IV. Evaluation of Nutrient Requirements

• Estimating Nutrient Requirements
• different methods have been used
• usually linear regression from a series of
  selected points
• each point represents a response (e.g., weight
  gain) to a set level of dietary nutrient
• graph showing all responses should increase
  linearly with dose/level and then plateau
• the requirement is where the sloped line
  intersects with the plateau/flat line

40
IV. Evaluation of Nutrient Requirements

• Estimating Nutrient Requirements
• however, this can sometimes be misleading,
  especially if dose/levels are widely distributed
  (e.g., 0, 100, 200, 300, 400, 500 mg/kg diet vs.
  200, 250, 300, 350, 400, 450 mg/kg diet)
• best method broken-line analysis
• also known as spline interpolation, fits a series
  of linear regression to determine exact point of
  inflection of growth curve