1st Page

Journal of Animal Science - Animal Growth, Physiology, and Reproduction

Biological markers of neonatal calf performance: The relationship of insulin-like growth factor-I, zinc, and copper to poor neonatal growth1


This article in JAS

  1. Vol. 88 No. 8, p. 2585-2593
    Received: Oct 28, 2009
    Accepted: Mar 31, 2010
    Published: December 4, 2014

    2 Corresponding author(s):

  1. T. W. Graham 2,
  2. J. E. Breher*,
  3. T. B. Farver,
  4. J. S. Cullor,
  5. M. E. Kehrli Jr. and
  6. A. M. Oberbauer§
  1. Veterinary Consulting Services, Davis, CA 95618;
    Department of Population Health and Reproduction, University of California, Davis 95616;
    National Animal Disease Center, USDA, ARS, Swine and Prion Diseases Research Unit, Ames, IA 50010; and
    Department of Animal Science, University of California, Davis 95616


Raising a heifer calf to reproductive age represents an enormous cost to the producer. Poor neonatal growth exacerbates the costs incurred for rearing, and use of blood variables that may be associated with poorly growing calves may offer predictive value for growth and performance. Thus, the principal objective of the present study was to describe changes in serum IGF-I, zinc, and copper from birth to 90 d in Holstein calves, while accounting for sex and twin status, in poorly growing calves and calves growing well. A second objective was to test the hypothesis that an association exists between these serum variables and morphometric indicators of growth. Measurements of BW, length, and height were recorded at birth and at 30, 60, and 90 d of age. Jugular blood (12 mL) was collected from each calf on d 1 to determine serum total protein, serum IgG, packed cell volume, serum zinc, serum copper, serum IGF-I, and CD18 genotype for bovine leukocyte adhesion deficiency; serum zinc, serum copper, and serum IGF-I (predictor variables) were also determined for each calf on d 2 through 10 and on d 30, 60, and 90. Stepwise multiple regression and logistic regression analyses were used to examine the relationships between the predictor variables and the dependent variables (BW, height, and length at d 30, 60, and 90 of life). Birth weight, sex, serum IGF-I (at all ages), serum copper, and the serum copper-to-zinc ratio were associated, to varying degrees, with the dependent growth variables. Birth weight was consistently the dominant predictor. In conclusion, these results suggest that lighter birth weight, reduced serum IGF-I, and inflammation may be important causes of poor growth in neonatal Holstein dairy calves.


Identifying biomarkers indicative of malnourishment or inflammation that predict poor growth and enable early intervention would reduce overall costs associated with raising dairy replacement heifers. Malnutrition in children and calves is associated with stunted growth (Fawzi et al., 1997; Hoffman, 1997; Mourits et al., 1997), greater morbidity and mortality from pneumonia and diarrhea (Sweeny and Divers, 1990; Yoon et al., 1997), and decreased protein synthesis and breakdown (Manary et al., 1997); malnutrition delays puberty, increases the time from birth to calving, and results in increased costs of raising heifers (Goodger et al., 1989).

Insulin-like growth factor-I is associated with increased somatic growth (Zapf and Froesch, 1986; Yakar et al., 2002), enhanced T-lymphocyte activity (Gelato, 1993; Dorshkind and Horseman, 2001), and stimulation of mammary protein synthesis (Hadsell et al., 1996). Cattle on a high plane of nutrition, compared with those on a low plane, have greater serum IGF-I concentrations (Bishop et al., 1989; Radcliff et al., 2004; Brown et al., 2005), although most studies have evaluated calves in laboratory rather than field settings (Baumrucker and Blum, 1994; Baumrucker et al., 1994a,b; Skaar et al., 1994; Daniels et al., 2008). Morbidity and inflammation are negatively associated with calf growth (Waltner-Toews et al., 1986; Paré et al., 1993; Erb, 1994; Jamaludin et al., 1996; Virtala et al., 1996). Serum zinc and copper are biomarkers of inflammation (Graham et al., 1994), and serum immunoglobulin concentration, sex, and twin status may differentially affect the growth of calves (el Bushra et al., 1989; Arrayet et al., 2002), with bulls growing faster than heifers and with compensatory growth in twins allowing similar overall growth compared with singletons. The objectives of this study were to describe changes in blood variables and growth from birth to d 90 in Holstein calves and to test the null hypothesis that there were no associations between serum IGF-I, zinc, and copper and growth, while accounting for sex and twin status.


All animals were cared for as outlined in the Guide for the Care and Use of Agricultural Animals in Agricultural Research and Teaching (FASS, 2010).


Holstein bull (n = 211) and heifer (n = 209) calves between 12 and 36 h of age were brought into the calf facility (Foster Farms, Modesto, CA) after being fed approximately 4 L of colostrum at the dairy of origin. All calves were provided by Foster Farms Dairy. All cattle were enrolled during a 6-wk period in March and April. The trial was completed in July. All calves were sequentially allocated into a treatment group (n = 105/group), and a color-coded identification tag was placed in the left ear of each calf, as described previously (Arrayet et al., 2002). Each calf was injected intramuscularly with vitamins A and D (500,000 IU, Boehringer Ingelheim Animal Health, St. Joseph, MO, and 75,000 IU, Butler Co., Dublin, OH, respectively). Calves were housed in individual hutches for the duration of the trial and fed 1.89 L of milk twice daily until they were weaned at 90 d of age. At wk 1 and 4, calves were vaccinated for infectious bovine rhinotracheitis and parainfluenza type 3 with an intranasal vaccine (Nasalgen, Schering-Plough, Omaha, NE). Bull calves were implanted at d 30 with 36 mg of zeranol (Ralgro, Pitman-Moore Inc., Kansas City, KS) and were surgically castrated at d 60. Calves were dehorned, and heifers had permanent identification tags placed in both ears at 4 wk of age. Calves were revaccinated for infectious bovine rhinotracheitis and parainfluenza type 3; vaccinated for bovine virus diarrhea, bovine respiratory syncytial virus, leptospirosis, and clostridia (Cattlemaster 4 + L5, Pfizer Animal Health, Exton, PA; Ultrabac 8, Pfizer Animal Health); and given vitamins A and D (500,000 and 75,000 IU, respectively) at wk 10 of age.


Calves were fed approximately 3.8 L/d of whole pasteurized waste milk containing 1 of 4 zinc supplements: 40 mg of zinc/kg of DM (that is, no additional zinc; control group) or 100 mg of zinc/kg of DM [control milk with 60 mg of additional zinc added as zinc sulfate (A. L. Gilbert Company, Oakdale, CA), zinc methionine (Zinpro, Eden Prairie, MN), or zinc lysine (Zinpro)]; the details of the milk diet have been published previously (Arrayet et al., 2002). Calves were offered a grain concentrate free choice from birth. The basal concentrate diet contained 40 mg of zinc/kg of DM. Parallel to the zinc supplementation in the milk, the concentrate diets were likewise either 40 or 100 mg of zinc/kg of DM with the same zinc forms as noted above. The diet formulation has also been published previously (Arrayet et al., 2002).

Data Collection

On arrival at the calf ranch, approximately 12 to 36 h after birth and colostrum ingestion, calves were weighed and morphometric measures were determined. Calf BW, length, and height were repeated at 30, 60, and 90 d of age. More detailed morphometric measures from these animals were also obtained and were published in Arrayet et al. (2002). Jugular blood [collected by venipuncture (10 mL), allowed to clot, centrifuged at 800 × g for 15 min at 25°C, and aliquoted into 2-mL microfuge tubes and frozen at −20°C; another 2 mL was collected in EDTA] was collected from each calf on d 1 to determine serum total protein, IgG, packed cell volume, serum zinc, serum copper, serum IGF-I, and CD18 genotype [bovine leukocyte adhesion deficiency (BLAD)]. Jugular blood [collected by venipuncture (10 mL), allowed to clot, centrifuged at 800 × g for 15 min at 25°C, and aliquoted into 2-mL microfuge tubes and frozen at −20°C] was also collected from each calf on d 2 through 10 and on d 30, 60, and 90 for the determination of serum zinc, serum copper, and serum IGF-I.

Blood Variables

The CD18 genotype (homozygous normal or heterozygous for the D128G allele that causes BLAD when present in a homozygous-affected calf) was identified as described previously (Arrayet et al., 2002). Plasma immunoreactive IGF-I was measured in a nonequilibrium assay after removal of IGFBP by acid ethanol cryoprecipitation as described previously (Breier et al., 1991; Chow et al., 1994) using iodinated human recombinant IGF-I (Collaborative Research Inc., Bedford, MA; Hodgkinson et al., 1989). The polyclonal antiserum (UB3-189, provided by L. Underwood and J. J. Van Wyk; University of North Carolina, Chapel Hill) was obtained through the National Institute of Diabetes and Digestive and Kidney Diseases National Hormone and Pituitary program (Bethesda, MD). The samples were run in 24 assays, with the same reference control used in each assay. Intraassay variation was 4.5% and interassay variation for controls was 6%. Serum zinc and copper values were measured using published techniques (Graham et al., 1994).

Statistical Analyses

With the data collected, the following variables were examined in the model: twin status; sex; BLAD status; 4 (average over the first 10 d and d 30, 60, and 90) serum zinc, copper, and IGF-I measurements; 4 (same days as the serum zinc and copper measurements) copper-to-zinc ratios; packed cell volume on d 1; serum IgG on d 1; serum total protein on d 1; and the morphometric measurements described above (BW, height, and length on d 0, 30, 60, and 90). The statistical program BMDP 2D (Statistical Solutions, Saugus, MA) was used to examine the descriptive information regarding each variable. Variables with large skewness and kurtosis factors were transformed using either the log or square root function, and the transformation yielding the smallest skewness and kurtosis factors was selected. The following transformations were made: log (d 30 copper-to-zinc ratio), log (d 60 copper-to-zinc ratio), log (serum IGF-I averaged over the first 10 d), log (d 30 serum IGF-I), and square root (d 60 serum IGF-I). No other variables were transformed.

For each statistical run (for each dependent variable), the sample was broken into quintiles, and the bottom 40% (representing poorly growing calves) were compared with the top 40% (representing calves growing well). The statistical programs BMDP 2R and BMDP PR (Statistical Solutions) were used to perform stepwise addition multiple regression and logistic regression analyses, respectively, on the data set. Nine statistical runs were made with each program, using each morphometric variable (BW, length, and height) at d 30, 60, or 90 as the dependent variable. The other nonmorphometric variables, as well as BW at birth, were used as the independent variables.


Of the 211 Holstein bull calves, 35 were twins. A much smaller percentage of the heifer calves were twins (17 twins out of 209 heifer calves evaluated). There was no effect of zinc supplementation on any measures of growth in the current trial (Arrayet et al., 2002). Descriptive data for the calves at different ages stratified by sex and twin status for serum zinc, copper, and the copper-to-zinc ratio are presented in Table 1. Bull calves tended to have reduced serum zinc (P < 0.06) and greater serum copper (P < 0.08) than heifer calves. This sex effect was lost when the data were expressed as the ratio of copper to zinc. Singletons had significantly less serum zinc (P < 0.001), greater serum copper (P < 0.001), and a greater copper-to-zinc ratio (P < 0.001) than twin calves. Although serum zinc decreased with increasing age, the opposite was observed for serum copper and the copper-to-zinc ratio (P < 0.001). Serum IGF-I concentrations (Table 2) were greater in singletons than twins (P < 0.001) and in bull calves than heifers (P < 0.001), and they increased with increasing age (P < 0.001), with heifer singletons exhibiting the greatest increase over time.

Table 1.

Please see the pdf to view this table.

Table 2.

Please see the pdf to view this table.


The results from the stepwise addition multiple regression analyses for BW, body length, and height are presented in Tables 3, 4, and 5, respectively. These results demonstrate a strong relationship between birth weight and later BW, length, and height in the Holstein calves. For example, in the case of d 30 BW, birth weight accounted for 56% of the variation. The magnitude of the birth weight contribution to the variation diminished with increasing age, but of the independent variables measured, birth weight represented the greatest contribution. Sex and twin status also contributed to the variation observed for BW, length, and height, but those independent variables became less predictive as the animals aged. By d 90 of age, in addition to birth weight, the serum concentration of IGF-I at d 90 showed a distinct relationship to BW, length, and height.

Table 3.

Please see the pdf to view this table.

Table 4.

Please see the pdf to view this table.

Table 5.

Please see the pdf to view this table.


The coefficients for the logistic regression models for BW, length, and height are presented in Tables 6, 7, and 8, respectively. Results from logistic regression models demonstrate, similar to the stepwise addition multiple regression models, that birth weight was the greatest predictor of whether a calf would be in the bottom 40% for BW, overall length, or height. This was true for all 3 ages evaluated (d 30, 60, and 90 of age). Additional associated coefficients for these dependent growth variables were sex, serum IGF-I (at all ages), and the serum copper-to-zinc ratio. For body length, serum copper and heterozygous BLAD status were also associated with poorly growing calves. Table 9 presents the values used to determine the adjusted-risk odds ratio. Values of a hypothetical poorly growing calf were calculated from the calves in the bottom 40% for growth. Similarly, averages of the values from all the calves in the top 40% for growth were used as the reference values. Adjusted-risk odds ratios for the reference group, compared with the poorly growing group, showed greater birth weight, greater serum IGF-I concentrations, and less serum copper than those of the poorly growing group. The reference group also consisted of more male calves than female calves.

Table 6.

Please see the pdf to view this table.

Table 7.

Please see the pdf to view this table.

Table 8.

Please see the pdf to view this table.

Table 9.

Please see the pdf to view this table.



Birth weight, serum IGF-I, serum copper, and the serum copper-to-zinc ratio were associated with risk of poor growth. Greater birth weight and serum IGF-I concentrations were contraindicative of the poor growth rate of a calf, whereas greater serum copper concentration and serum copper-to-zinc ratio were associated with an increased risk of a calf growing poorly. Birth weight was the principal predictor variable associated with the dependent growth variables of BW, body length, and height. The strength of association is most likely attributable to greater genetic potential, with larger animals have larger offspring; however, birth weight also accounts for growth differences attributable to sex and twin status (Arrayet et al., 2002).

Reasons for an association between serum IGF-I and growth have been described previously in mice and rats (Zapf and Froesch, 1986; Hadsell et al., 1996) as well as livestock (Bell et al., 1998; Lammers et al., 1999; Radcliff et al., 2004; Brown et al., 2005). This association is most likely due to an enhanced stimulation of long bone growth, increased nutrient availability (Lowe, 1991), and muscle tissue accretion (Fiorotto et al., 2003) by IGF-I. Further, IGF-I enhances immune function under periods of stress; stress raises immunosuppressive glucocorticoids, and under this condition, IGF-I maintains lymphocytic activity (Dorshkind and Horseman, 2001). A reduction in IGF-I is therefore correlated with diminished immune function while under stress. Malnutrition is a documented stressor that raises glucocorticoids (Dorshkind and Horseman, 2001). The nutrient deprivation also results in depressed IGF-I concentrations, although there is a shift in the remaining IGF-I utilization to support the immune system (Davis, 1998). Thus, periods of stress or malnutrition reduce IGF-I, and the remaining IGF-I is redirected away from stimulating growth.

Inflammation or the nonspecific immune response-induced changes in serum zinc and copper have been described over the past several decades. Under the influence of IL-1, IL-6, tumor necrosis factor, and other inflammatory mediators, serum copper is increased as ceruloplasmin is exported from the liver, causing serum copper concentrations to increase. Similarly, zinc is sequestered or excreted, although the mechanism for this has not been well described (Keen and Graham, 1989; Graham et al., 1995; Polberger et al., 1996). The negative association of greater serum copper concentrations and greater copper-to-zinc ratios with growth can most likely be attributed to inflammation (Graham et al., 1994). Serum zinc and copper are direct biomarkers of inflammation (Graham et al., 1994), with decreased serum zinc and increased serum copper present during episodes of inflammation. The poorly growing calves had almost 175% greater serum copper concentrations in comparison with the reference group of calves growing well. Likewise, the copper-to-zinc ratio was increased in the poorly growing calves. Beyond reduced feed intake when a calf is ill, inflammation utilizes energy stores in the calf that could otherwise be used for growth, compounding the lack of nutrient intake.

The present data indicate that by avoiding exposure to those factors that would cause decreased serum IGF-I and increased inflammation, the risk of a calf growing poorly is decreased. These data also indicate that BW at birth is a very strong predictor variable of growth (multiple regression R2 values ranged from 37 to 56%, 27 to 43%, and 14 to 22% at d 30, 60, and 90 of age, respectively). Future research efforts should be directed toward devising methods (nutritional, managerial, and pharmacological) to increase serum IGF-I and decrease inflammatory events in neonatal calves. Breeders and producers should also consider selecting for animals with heavier birth weights, but with body conformations that allow for calving ease. Through these techniques, dairy calves may experience less malnutrition and greater growth rates, thus increasing the economic returns to the producer.

It is well documented that excessive prepubertal growth rates reduce future milk production (reviewed in Mourits et al., 2000). In fact, in large-breed dairy cattle, the optimal prepubertal growth rate has been modeled at 0.7 kg/d. Although this rate increases calf rearing time, thereby increasing calf costs, the reduction in future milk produced would not compensate for this short-term cost savings (Mourits et al., 2000). However, recent studies have shown that growth rate per se does not differentially affect mammary parenchymal development (Daniels et al., 2009). Thus, replacement heifer producers must balance a fine line between optimal prepubertal growth and suboptimal growth caused by malnutrition. The present study, however, focused on the need to identify, during the early neonatal life, calves susceptible to malnutrition and its associated impaired growth and immune function.

To our knowledge, prospective cohort studies of this magnitude studying IGF-I in newborn calves to d 90 of life have not been reported previously in cattle, although a positive association between IGF-I and growth has been reported for older beef cattle (Davis and Simmen, 2006, 2010; Lancaster et al., 2008) and dairy heifers (Brickell et al., 2009a,b,c). This study described changes in serum IGF-I from birth to d 90 in Holstein calves, accounting for sex and twin status. This study also found a relationship between birth weight, serum IGF-I, serum zinc, and serum copper and BW, length, and height at d 30, 60, and 90 of age. The data indicate that lighter birth weight, reduced serum IGF-I, and inflammation may be important causes of poor growth in neonatal Holstein dairy calves. This study targeted dairy calves; however, it is likely that the information from this trial can be used to some degree to interpret human infant malnutrition. Whether similar patterns of change and associations between growth and serum IGF-I occur in human infants requires further study and definition.




Be the first to comment.

Please log in to post a comment.
*Society members, certified professionals, and authors are permitted to comment.