Abstract

Research that advances the practice of poultry nutrition covers a wide range, including fundamental studies that explore mechanisms in cells or a small number of birds and applied research that determines product yield in authentic production facilities. Ultimately, the applied research informs the economic analysis necessary for implementation of novel nutritional strategies or products. Nutritional controversies arise from applied research experiments that were not designed and interpreted based on the realities learned from mechanistic work. The experimental design and measurements selected for comparing nutrient sources or setting nutrient recommendations should be informed by underlying mechanistic information, such as the relative priorities of cells and tissues for the nutrient and the shape of the dose response relationship across a wide range of added dietary levels of the nutrient. Integrating mechanistic and applied research provides more robust results that can be used across wider ranges of diets and husbandry conditions.

Primary Audience: Nutritionists

DESCRIPTION OF PROBLEM

Nutritional research facilitates continual advancement of feeding practices that drive ever increasing efficiency of production of poultry products [1]. Optimally, there is a continuum of research on novel nutritional questions ranging from fundamentally mechanistic to distinctly applied (Table 1). Applied research done in actual or simulated production facilities verifies the efficacy and cost efficiency of new concepts or products and facilitates their widespread adaptation in the industry. However, predictable application across wide-ranging diets and husbandry conditions requires a deeper understanding of the underlying mechanisms by which the nutritional concept or product works. These mechanistic underpinnings require research at the genomic, molecular, cellular, and organismal level of investigation. In addition to providing perspective on the limitations and optimal applications, mechanistically oriented research often uncovers new concepts and potential products that can be developed for continued progress in poultry nutrition.

Table 1.

Overview of research questions and measurements in nutrition

Research type  Experimental unit (replication)  Measurement 
Genomic  Genome (n = 1)  Genes, control elements 
Mechanistic, cellular  Cell line or primary cultures, usually from 1 bird  mRNA, metabolites, enzyme activities, transport rates 
Mechanistic, organismal  Individual birds, with replication (n = 3–10)  mRNA, metabolites, enzyme activities, transport rates 
Translational, highly controlled  Pens of birds (n = 5–15)  Growth, intake, efficiency, morphometrics 
Translational, field  Complexes of birds (n = 3–5)  Growth, intake, efficiency, morphometrics 
Economic  Producer (n = 1)  Net cost or savings 
Research type  Experimental unit (replication)  Measurement 
Genomic  Genome (n = 1)  Genes, control elements 
Mechanistic, cellular  Cell line or primary cultures, usually from 1 bird  mRNA, metabolites, enzyme activities, transport rates 
Mechanistic, organismal  Individual birds, with replication (n = 3–10)  mRNA, metabolites, enzyme activities, transport rates 
Translational, highly controlled  Pens of birds (n = 5–15)  Growth, intake, efficiency, morphometrics 
Translational, field  Complexes of birds (n = 3–5)  Growth, intake, efficiency, morphometrics 
Economic  Producer (n = 1)  Net cost or savings 

Genomic and Mechanistic Research

At the most fundamental level, nutritionists can explore the genome of poultry. Genomic analysis permits an understanding of the genes, and consequently the proteins present in poultry, and permits comparisons and analogies with exhaustively studied species, such as mice and humans. Although nutritionists often consider genomics to be the domain of geneticists and animal breeders, this level of analysis has many important applications to applied poultry nutrition. For example, interrogation of the published genomes of chicken and turkey reveals the absence of the gene that codes for the taste receptor that detects sucrose and other sweet molecules (Tas1r2); the gene for the taste receptor that detects amino acids (Tas1r1) is present, however. Such information is fundamental for understanding feedstuff preferences as well as developing strategies to optimize intake of underused feedstuffs. The developing field of nutragenetics explores how intraindividual variability in the genome causes variation in dietary responses and should help increase the uniformity of poultry flocks. The field of nutrigenomics explores the effect of diet on gene expression, the metabolome and, ultimately, the fate of consumed nutrients.

Mechanistic research at the cellular or organismal level builds on genomics and provides a framework for understanding the role of nutrients in metabolism and, ultimately, their utility for optimizing the productivity of a bird. Mechanistic research helps us understand how nutrients are digested and absorbed, how they are transported throughout the body, and how they are metabolized into the appropriate form for their required functions. For example, research on gene expression during a lysine deficiency has illuminated the cell populations (e.g., thymocytes) that are least able to compete for this nutrient and therefore become deficient before more routinely monitored tissues such as skeletal muscle [2, 3]. Such mechanistic information should inform the selection of endpoints examined for nutritional adequacy. Ideally, we should select endpoints that represent metabolic processes or cell types that are of the lowest priority and not those that are of high priority when examining nutrient requirements or the relative bioavailability of nutrient sources or nutrient-releasing enzymes.

Translational Research

Translational research examines the effect of a dietary manipulation on performance parameters and other endpoints of nutritional adequacy. This is done using facilities and husbandry that provide tight environmental control (e.g., temperature, disease status), the capacity for large numbers of repetitions, and uniformity so that very low variation exists among replicates. Such facilities permit the detection of small differences (<5%) with high statistical confidence. Translational research is also done in authentic production facilities where the highly variable environmental differences provide results that are relevant to an actual set of husbandry and geographic conditions. These results can then be used to predict economic returns that should result from adopting a novel concept or product in that set of conditions.

Integrating Mechanistic and Translational Research

Results from each level of research inquiry described in Table 1 should inform the optimal experimental design and measurements collected in the other levels of inquiry. When attention to other levels of inquiry is inadequate, results may be misinterpreted or even invalid due to inappropriate endpoints, experimental design, and statistical analysis. This is illustrated by the many controversies surrounding the bioavailability of different forms of nutrients (e.g., organic vs. inorganic minerals, l-methionine vs. analogs). We know from mechanistic research that a nutrient’s bioavailability depends upon its physical and chemical properties (e.g., solubility or reactivity), its digestive and absorptive routes, its propensity for interactions or antagonisms with other dietary constituents, and its metabolic pathway to the location and form needed for its essential functions. Often each of these properties has a different dose-response characteristic across dietary levels (Figure 1). Furthermore, each of these relationships may differ between nutrient sources due to their different physical properties, absorptive routes, metabolic fates, and so on. Thus, when we compare the bioavailability of 2 different nutrient forms, we must select our experimental design and endpoints so that realistic and defendable results are obtained.

Figure 1.

The amalgamation of 5 underlying processes determines the relationship between the dietary concentrations of 2 different nutrient sources versus product yield (e). In this hypothetical example, one nutrient source is represented by solid lines and a second source is shown in dashed lines. The (a) rates of digestion, (b) absorption, (c) utilization, and (d) interference due to antagonisms and (f) antagonists of each nutrient source follow different Michaelis-Menten kinetics. In each case, the dose-response relationship between dietary nutrient concentration and (e) product yield is a reflection of all 5 of these digestive and metabolic relationships. Note that, even when Michaelis-Menten kinetics occur for all underlying processes, no region of the dietary response is linear. Also note that comparison of the 2 nutrient sources at low dietary levels results in very different conclusions of relative bioavailability than comparisons at levels near the requirement.

Figure 1.

The amalgamation of 5 underlying processes determines the relationship between the dietary concentrations of 2 different nutrient sources versus product yield (e). In this hypothetical example, one nutrient source is represented by solid lines and a second source is shown in dashed lines. The (a) rates of digestion, (b) absorption, (c) utilization, and (d) interference due to antagonisms and (f) antagonists of each nutrient source follow different Michaelis-Menten kinetics. In each case, the dose-response relationship between dietary nutrient concentration and (e) product yield is a reflection of all 5 of these digestive and metabolic relationships. Note that, even when Michaelis-Menten kinetics occur for all underlying processes, no region of the dietary response is linear. Also note that comparison of the 2 nutrient sources at low dietary levels results in very different conclusions of relative bioavailability than comparisons at levels near the requirement.

Problems occur when experiments are designed based on habit or convenience and not on sound nutritional knowledge. For example, bioavailability comparisons are commonly examined at dietary levels that are well below requirements so that a linear response can be approximated and relatively simple statistics can be applied to calculate ratios of slopes, which provides an estimate of relative bioavailability [4]. This estimate is then routinely applied to dietary levels at, or above, the requirement. This approach has become the gold standard, even though no biological reason exists to expect that the dose response of a nutrient is linear at any range of dietary levels or that the relative bioavailability of 2 nutrient sources at frankly deficient dietary levels should be predictive of bioavailability at sufficient dietary levels. Thus, nutritional controversies arise from experiments that were not designed and interpreted based on the realities learned from mechanistic work. In other words, nutritional progress is greatly improved when endpoints and experimental designs are tailored to the biology of the specific question being asked instead of forcing questions into historical experimental designs that have become the convention. When recent developments at the mechanistic level are used to help design and interpret applied research, the results should be more valuable and less controversial.

CONCLUSIONS AND APPLICATIONS

  1. Nutritional controversies arise from applied research experiments that were not designed and interpreted based on the realities learned from mechanistic work.

  2. For bioavailability or nutrient requirement studies, the physical and chemical properties of a nutrient, its digestive and absorptive routes, its propensity for interactions or antagonisms with other dietary constituents, its metabolic pathway to the location, and form needed for its essential functions are important mechanistic parameters that should inform the experimental design and endpoints measured.

1
Presented as a part of the Informal Nutrition Symposium “From Research Measurements to Application: Bridging the Gap” at the Poultry Science Association’s annual meeting in San Diego, California, July 22–25, 2013.

REFERENCES AND NOTES

1
Gous
R. M.
Predicting nutrient responses in poultry: Future challenges.
Animal
 , 
2007
, vol. 
1
 (pg. 
57
-
65
)
[PubMed]
2
Humphrey
B. D.
Kirsch
S.
Morris
D.
Molecular cloning and characterization of the chicken cationic amino acid transporter-2 gene.
Comp. Biochem. Physiol. B Biochem. Mol. Biol.
 , 
2008
, vol. 
150
 (pg. 
301
-
311
)
[PubMed]
3
Humphrey
B. D.
Stephensen
C. B.
Calvert
C. C.
Klasing
K. C.
Lysine deficiency and feed restriction independently alter cationic amino acid transporter expression in chickens (Gallus gallus domesticus).
Comp. Biochem. Physiol. A Mol. Integr. Physiol.
 , 
2006
, vol. 
143
 (pg. 
218
-
227
)
[PubMed]
4
Ammerman
C. B.
Baker
D. H.
Lewis
A. J.
Bioavailability of Nutrients for Animals
1995
San Diego, CA
Academic Press