Funct Integr Genomics (2009) 9:197217 DOI
10.1007/s10142-008-0103-x
ORIGINAL PAPER
Insulin regulates milk protein synthesis at multiple levels in
the bovine mammary glandKarensa K. Menzies & Christophe Lefvre
& Keith L. Macmillan & Kevin R. Nicholas
Received: 21 August 2008 / Revised: 25 November 2008 / Accepted:
25 November 2008 / Published online: 24 December 2008 #
Springer-Verlag 2008
Abstract The role of insulin in milk protein synthesis is
unresolved in the bovine mammary gland. This study examined the
potential role of insulin in the presence of two lactogenic
hormones, hydrocortisone and prolactin, in milk protein synthesis.
Insulin was shown to stimulate milk protein gene expression, casein
synthesis and 14C-lysine uptake in mammary explants from late
pregnant cows. A global assessment of changes in gene expression in
mammary explants in response to insulin was undertaken using
Affymetrix microarray. The resulting data provided insight into the
molecular mechanisms stimulated by insulin and showed that the
hormone stimulated the expression of 28 genes directly involved in
protein synthesis. These genes included the milk protein
transcription factor, ELF5, translation factors, the folate
metabolism genes, FOLR1 and MTHFR, as well as several genes
encoding enzymesElectronic supplementary material The online
version of this article (doi:10.1007/s10142-008-0103-x) contains
supplementary material, which is available to authorized users. K.
K. Menzies (*) : C. Lefvre : K. R. Nicholas Department of Zoology,
University of Melbourne, Parkville, VIC 3010, Australia e-mail:
[email protected] C. Lefvre Victorian Bioinformatics
Consortium, Monash University, Clayton, VIC 3080, Australia K. K.
Menzies : K. L. Macmillan School of Veterinary Science, University
of Melbourne, Werribee 3030, Australia Present address: C. Lefvre :
K. R. Nicholas Institute of Technology Research and Innovation,
Deakin University, Geelong 3217, Australia
involved in catabolism of essential amino acids and biosynthesis
of non-essential amino acids. These data show that insulin is not
only essential for milk protein gene expression, but stimulates
milk protein synthesis at multiple levels within bovine mammary
epithelial cells. Keywords Insulin . Milk protein . Mammary gland .
Microarray Abbreviations AA amino acid EAA essential amino acid
NEAA non-essential amino acid BCAA branched chain amino acid HEC
hyperinsulinemic-euglycemic clamp I insulin F hydrocortisone P
prolactin NH no hormone IPA ingenuity pathways analysis
Introduction Historically, insulin has been attributed a role in
proliferation and maintenance in mammary tissue, but now an
important role in milk protein synthesis is emerging. In vitro
studies in the mouse and rat have shown that there is an absolute
requirement for insulin, in the presence of prolactin and
hydrocortisone, for the induction of milk protein gene expression
(Bolander et al. 1981; Choi et al. 2004; Kulski et al. 1983;
Nagaiah et al. 1981). However, the role of insulin in milk protein
synthesis in the bovine mammary gland remains equivocal. Mammary
culture experiments in the 1970s and 1980s indicated that
198
Funct Integr Genomics (2009) 9:197217
induction of milk protein gene expression required the
complement of prolactin, hydrocortisone and insulin (Andersen and
Larson 1970; Choi et al. 1988; Djiane et al. 1975; Servely et al.
1982). In contrast, more recent studies by Sheehy et al. (2000)
suggested milk protein genes could be expressed in cultured mammary
explants from pregnant cows in the absence of insulin. In vivo
studies in cows employing the hyperinsulinemic euglycemic clamp
(HEC) technique, which elevated circulating insulin levels
fourfold, increased milk protein yield by 1015% (Griinari et al.
1997; Mackle et al. 1999b, 2000b). This milk protein response was
enhanced to 25 30% in the presence of an additional exogenous amino
acid (AA) supply (Griinari et al. 1997; Mackle et al. 1999b,
2000b). In these HEC studies and that of McGuire et al. (1995),
infusion of either AA or branched-chain AA (BCAA) alone did not
enhance milk protein production (Griinari et al. 1997; Mackle et
al. 1999b, 2000b; McGuire et al. 1995). Not all HEC studies in cows
have shown an increase in milk protein production (Metcalf et al.
1991), and insulin did not enhance milk protein yield when
administered locally to the mammary gland using an intramammary
technique (Mackle et al. 2000a). However, in both bovine and murine
mammary tissue, insulin has been shown to stimulate milk protein
synthesis in vitro (Park et al. 1979; Wang and Amor 1971).
Collectively, these studies suggest that insulin may play a role in
milk protein production, and it is conceivable that this role may
be partly a direct effect on the mammary gland. Milk protein
synthesis may be regulated at multiple levels within the mammary
epithelial cells including transcription, post-transcription,
translation and amino acid supply. Therefore, it is conceivable
that insulin has a role in the synthesis of milk proteins and not
simply expression of the milk protein genes. Hydrocortisone and
prolactin play a role in both transcription of the rodent casein
gene and stabilisation of the mRNA, whilst insulin is thought only
to be involved in transcription of the casein gene (Bolander et al.
1981; Kulski et al. 1983; Nagaiah et al. 1981). More recent
investigations in the mouse have shown that insulin and prolactin
synergistically lengthen the poly(A) on -casein mRNA by increased
phosphorylation of the cytoplasmic polyadenylation element binding
protein, resulting in enhanced translation of the protein,
presumably by affecting accumulation of casein mRNA (Choi et al.
2004). In the cow, a phosphorylation role for insulin has been
identified in translation. Barash (1999) found that the combination
of insulin and prolactin synergistically promoted the
phosphorylation of eIF4E-binding protein 1 (4E-BP1), an initiation
factor-binding protein, in cow mammary tissue. In addition, 4E-BP1
was shown to be phosphorylated by MAP kinase in response to insulin
treatment in muscle and adipose
tissues. Phosphorylation of 4E-BP1 results in the release of an
initiation factor for translation, eIF4E, and makes the latter
available for translation (Lin et al. 1994; Pause et al. 1994). The
elongation phase of protein synthesis in eukaryotes has also been
reported to be stimulated by insulin through phosphorylation of
eEF1 (Myers et al. 1994) and dephosphorylation of eEF2 (Redpath et
al. 1996). The dephosphorylation of eEF2 results in increased eEF2
activity and rate of peptide elongation (Proud 1994). Furthermore,
the activity levels of eEF2 are high in bovine lactating mammary
tissue compared to that in heifers (Christophersen et al. 2002). An
increase in eIF4E transcripts in vivo in bovine lactating tissue
(Long et al. 2001; Toerien and Cant 2007) and eEF2 synthesis in NIH
3T3 mouse embryonic fibroblast cells (Levenson et al. 1989)
suggests that transcription and phosphorylation of translation
factors could be an important regulatory mechanism for synthesis of
milk proteins. A role for insulin in transcription of genes for
translation factors in the bovine mammary gland remains to be
elucidated. The extraction rate of AA from the blood by the mammary
gland is very high and the overall efficiency of mammary
utilisation of AA for milk protein synthesis, assessed by mammary
uptake to output ratios, exceeds 80% in the dairy cow (Clark et al.
1978; Mackle et al. 2000a, b). However, the uptake of some
essential AA (EAA), most notably arginine, lysine and the BCAAs, is
greater than their output in milk (Mackle et al. 2000b; Mepham
1982). Metabolism of these particular EAA is central for the
formation of the intracellular pool of non-essential AA (NEAA)
supply (Davis and Mepham 1976; Wohlt et al. 1977), which is
inadequate to account for the needs of milk protein synthesis
(Clark et al. 1978; Mackle et al. 2000b). Although in vivo studies
involving short-term treatment with insulin in cows have found
little effect on mammary uptake of AA (Metcalf et al. 1991), a
sustained elevation in plasma levels for several days utilising HEC
technique increased the fractional extraction of BCAA by 48% and
arginine and lysine by 20% (Mackle et al. 2000b), as well
increasing milk protein yield (Griinari et al. 1997; Mackle et al.
1999a, b; Mackle et al. 2000b; McGuire et al. 1995). Furthermore,
in vitro studies in cultured bovine mammary acini have shown that
insulin stimulates uptake of the nonmetabolisable BCAA cycloleucine
in a dose-dependent manner (Park et al. 1979). Insulin has been
shown to upregulate the Y+ transport system for the cationic AA,
lysine and arginine in mouse mammary culture experiments (Kansal et
al. 2000). A direct role for insulin to stimulate the uptake of
cationic AA in bovine mammary tissue remains to be confirmed. In
addition, the underlying molecular mechanism by which insulin may
stimulate mammary epithelial cell uptake of the neutrally charged
BCAA by the L transporter system (Park et al. 1979), and
potentially of arginine and lysine by the Y+ transport
Funct Integr Genomics (2009) 9:197217
199
system, remains to be elucidated. It is also conceivable that
insulin may play a role in the cellular metabolism of EAA to meet
the requirements of NEAA for milk protein synthesis, and this also
remains to be examined. Most ruminant studies to date investigating
the role of insulin on milk protein production (Griinari et al.
1997; Mackle et al. 1999b, 2000b; McGuire et al. 1995; Metcalf et
al. 1991) have been empirical and complicated by the potential
effects of whole body metabolism. Tissue culture models permit the
study of the direct effects of hormones on the regulation of milk
protein gene expression, providing a greater understanding in the
transcriptional regulation of genes. These models also allow the
study of milk protein synthesis which may be regulated at the
levels of posttranscription, translation and amino acid supply. The
current study has exploited the mammary explant culture model to
investigate the direct role/s of insulin in milk protein synthesis
at multiple levels in the bovine mammary gland. The use of
Affymetrix microarray has allowed a global assessment of mammary
gene expression and offers potential insight into the molecular
mechanisms underlying the insulin-stimulated milk protein
synthesis.
for extraction of total RNA were stored at 80C until further
analysis. Northern blot analysiscasein gene expression Total RNA
was extracted using an RNeasy Lipid Kit (Qiagen, Sydney,
Australia). Total RNA (10 g) was electrophoresed in MOPS 1% agarose
gels at 100 V/cm using 10% MOPS buffer, transferred to Zeta-Probe
GT membranes and prehybridised for 4 h at 42C in 30% formamide
hybridisation buffer. Membranes were hybridised overnight at 42C
with a 32 P--s1-casein cDNA probe (4.5105 cpm/mL) labelled with
[-32P]dCTP (3,000 Ci/mmol; Perkin Elmer, Boston, MA, USA) using
DECAprime II random priming DNA labelling kit (Ambion, Melbourne,
Australia). Membranes were washed once at room temperature with 2%
SSC and 1% SDS, twice at room temperature with 1% SSC and 0.1% SDS
and twice at 42C in 0.1% SDS and 0.1% SSC. The 32 P-labelled cDNA
was detected using a phospho-imager (Bio-Rad, Molecular Dynamics
Typhoon Scanner). Synthesis of casein Mammary explants to be
analysed for casein synthesis were incubated with 33P (3,000
Ci/mmol; GE Health Care, Melbourne, Australia) at 5 Ci/mL medium
during the final 6 h of culture. The incorporation of 33P into the
casein micelle was measured by calciumrennin precipitation of the
casein micelle as previously described (Juergens et al. 1965).
Casein synthesis is expressed as decay per minute per milligram
tissue for each treatment. Lysine uptake Lysine uptake in mammary
explants was determined following incubation with L-[U-14C]lysine
(322 mCi/mmol; MP Biomedicals, USA) at 0.5 Ci/mL of media for the
final 6 h of culture. Explants were weighed, washed immediately by
vortexing in ice-cold acetone twice, followed by two washes in
ice-cold 5% trichloric acid. Explants were dissolved in 0.5 mL 500
mM NaOH at 50C for 4872 h and the level of radioactivity determined
after the addition of 10 mL Ultima Gold scintillator fluid
(Bio-Rad, Melbourne, Australia). Amino acid uptake was expressed as
disintegrations per minute per milligram tissue. Statistical
analysis of lysine uptake Linear mixed models fitted by GenStat
10th Edition were used to analyse the two culture experiments of
14C-lysine uptake in mammary explants. For 14C-lysine uptake in
mammary explants cultured over a 48-h time period, the model was:
lysine=constant+treatment+time+treatment
Materials and methods Animals Pregnant HolsteinFreisan cows no
later than 30 days prior to parturition were slaughtered at the
abattoir according to the abattoir guidelines and the udders
retrieved from the production line. Mammary tissue samples were
excised under sterile conditions and transported to the laboratory
in Medium 199 with Earles salts (Gibco BRL Life Technologies,
Melbourne, Australia) for mammary explant culture. The stage of
pregnancy was determined from farmers records of cow insemination
date and ultrasound pregnancy tests. Tissue culture Mammary
explants from pregnant animals were prepared and cultured in Medium
199 with Earles salts (Gibco BRL Life Technologies) containing 5%
foetal calf serum as described previously (Nicholas and
Tyndale-Biscoe 1985), but without the addition of bovine serum
albumin. Explants were placed on siliconised lens paper which was
floated on 5 mL of media. Explants were incubated at 37C in 5% CO2
and the media changed every second day. Hormones were added at the
following concentrations unless indicated: bovine insulin (I), 100
ng/mL (Sigma-Aldrich, Melbourne, Australia); hydrocortisone (F), 50
ng/mL (Sigma-Aldrich, Melbourne, Australia) and ovine prolactin
(P), 200 ng/mL (National Hormone and Pituitary Program, USA).
Explants
200
Funct Integr Genomics (2009) 9:197217
time+cow+cowtreatment+ where lysine, the observed lysine level;
time, the effect of 0, 24 or 48 h and treatment, the effect hormone
treatment (i.e. NH, I, FP or IFP) are the fixed effects and cow,
the effect of cow; cowtreatment, the effect of explant and , the
residual random error are the random effects. Overall significance
tests were conducted with REML F tests. Data is graphed as the
meanSEM for uptake of 14C-lysine (decay per minute per milligram of
tissue) in cultured mammary explants. For 14C-lysine uptake in
mammary explants cultured in different hormone treatments for 3
days after an initial 2-day culture in the absence of exogenous
hormones, the following linear mixed model was used to analyse the
data: lysine= constant+treatmenttime+cow+cowtreatment+ where the
terms are defined as in the previous experiment. Note, the
treatmenttime term corresponds to the seven-level factor, the
levels being FP, I, IFP(i), IFP(ii), IFP(iii), NH (day 2) and NH
(day 3). Overall significance tests were conducted with REML F
tests. Data is graphed as the meanSEM for uptake of 14C-lysine
(decay per minute per milligram of tissue) in cultured mammary
explants. Affymetrix microarray Mammary tissue from four cows was
used for microarray analysis. Mammary explants from each cow were
initially cultured in media with NH for 5 days and then in media
containing FP or IFP for 3 days and total RNA extracted. The RNA
from two cows was pooled (#1 and #2, #3 and #4) for each culture
treatment FP, IFP and NH, and the RNA samples analysed using
Affymetrix GeneChip bovine genome arrays under contract to the
Australian Genomics Research Foundation. Microarrays were performed
in duplicate for each culture treatment FP, IFP and NH using a
total of six Affymetrix GeneChips. Signal intensities were
normalised using the robust multi-array average (RMA) function in
Bioconductor (http://www.bioconductor.org) (Bolstad et al. 2003;
Irizarry et al. 2003). Differential gene expression was then
assessed
using the differential expression analysis function of the limma
package (Smyth 2004). Genes that were differentially expressed in
mammary explants cultured in FP compared to IFP formed the
insulin-responsive (I-responsive) dataset, and genes that were
differentially expressed in mammary explants cultured in NH
compared to IFP formed the IFP-responsive dataset. Version 2 of the
bovine Affymetrix chip annotation was used to assign gene
descriptions to the differentially expressed probe sequences. The
bovine Affymetrix chip is not completely annotated and genes that
were described as similar to x gene were designated that particular
genes description (Table 1). To focus specifically on mammary
function, both the IFP- and I-responsive genes were assigned to
cellular function categories using information from NCBI Gene
Entrez and manual searching of the literature using the NCBI Pubmed
and Entrez gene databases. Where appropriate, genes were designated
additional cellular functions. To validate the mammary explant
culture as an appropriate model to investigate the molecular
mechanisms underlying milk protein synthesis, the IFP-responsive
genes were identified with key genes and cellular processes
reported in the literature to be differentially regulated during
lactogenesis in the mammary gland. This included a comparison with
datasets from three studies reporting key regulatory genes of
lactogenesis in the mouse (Naylor et al. 2005; Ramanathan et al.
2007; Rudolph et al. 2003) and referencing to bovine studies.
I-responsive genes important for protein synthesis were identified
using the information collected from the NCBI database and computer
software that identified metabolic pathways within the dataset
related to amino acid metabolism, as described below. Analysis of
amino acid metabolic pathways of I-responsive gene dataset
Canonical Pathway Analysis within Ingenuity Pathways Analysis (IPA)
software (Ingenuity Systems; http://www. ingenuity.com) was used to
identify metabolic pathways
Table 1 The annotation summary of differentially expressed genes
in IFP- and I-responsive genes in cultured mammary explants
Treatment Total IDs Annotated Fully Insulin up Insulin down Total
IFP up IFP down Totala
Hypothetical and transcribed locus Similar to 63 45 108 172 183
355 28 30 58 113 85 51
Un-annotated
Mapped to Unigenesa 107 86 190
Mapped to IPAa 101 81 182
Mapped to IPAKa 90 65 155
125 139 264 439 477 916
47 45 92 138 164 302
10 19 29 26 45 69
Included for the I-responsive dataset is the numbers of genes
that mapped to Unigenes, Ingenuity Pathways Analysis (IPA) and to
IPA Knowledge Base (IPAK) for Canonical Pathways Analysis
Funct Integr Genomics (2009) 9:197217
201
that were significantly associated with the I-responsive
dataset. The IPA software does not interpret bovine Affymetrix or
Unigene IDs, so genes in the I-responsive dataset were assigned
corresponding human, mouse or dog Unigenes, which resulted in 97%
of the annotated genes Iresponsive genes being eligible for IPA
analysis (Table 1). The corresponding Unigene IDs for the
I-responsive dataset were uploaded into IPA software as up- and
down-regulated gene datasets, and the genes mapped to Ingenuity
Pathway Knowledge Base for analysis. The numbers of genes mapped to
Ingenuity Pathway Knowledge Base for each dataset is outlined in
Table 1. Preliminary functional analysis of the I-responsive
datasets by IPA (which identified biological functions
significantly associated with a dataset) showed only the positively
regulated genes that were significantly associated with protein
synthesis (data not shown) and, therefore, only the up-regulated
genes were subject to Canonical Pathway Analysis. This identified
metabolic pathways from the IPA library of canonical pathways that
were significant to the dataset in two ways: (1) A ratio of the
number of genes from the dataset that map to the pathway divided by
the total number of genes that map to the canonical pathway is
displayed. (2) Fishers exact test was used to calculate a P value
determining the probability that the association between the genes
in the dataset and the canonical pathway is explained by chance
alone. Canonical pathways that were related to amino acid
metabolism and were identified by IPA to be associated with
I-responsive dataset were selected. The relevance of these genes
with amino acid metabolism was further assessed using the linked
literature references to the gene from IPA and manual searching of
literature using the NCBI Pubmed and Entrez gene databases. Genes
within the I-responsive dataset that were not associated with
Ingenuity Pathways Knowledge Base were manually screened and
assessed using the literature of NCBI Pubmed and Entrez gene
databases to determine if relevant to amino acid metabolism.
Reverse transcriptase PCR First strand cDNA synthesis was performed
using Superscript III Reverse Transcriptase (Invitrogen Life
Technologies, Melbourne, Australia). In a 20-L reaction volume, 1 g
total RNA was used to generate cDNA. The polymerase chain reaction
(PCR) was performed using the GoTaq Green PCR kit (Promega, Sydney
Australia), and in a 20-L reaction volume, 0.2 L of the first
strand reaction was used as a template. Annealing temperature used
for all the primers of the genes of interest was 50C, and for the
UXT primers, 60C. PCR for the housekeeping gene, UXT (Bionaz and
Loor 2007), was performed at 20, 25, 30 and 35 cycles of
amplification. PCR using ELF5, EIF4E and
SLC7A5 primer sets was performed using 28 cycles and 35 cycles
for the FOLR1 primer set. PCR reactions were electrophoresed
through 1% TAE agarose gel with SYBR Safe (Invitrogen, Sydney,
Australia) DNA gel stain, the PCR products visualised using UVP
BIODoc-it System (Pathtech, Australia) and saved by PCTV Vision
software. The primer sequences and PCR product sizes are outlined
in Table 2. To confirm correct amplification with FOLR1, ELF5,
EIF4E and SLC7A5 primer sets, the amplified products were gel
isolated and the sequence reactions and subsequent analysis by an
ABI Prism Genetic Analyser were performed by the Pathology
Department of The University of Melbourne. Alignment of the
resulting nucleotide sequences was performed using BLAST, National
Centre for Biotechnology Information (NCBI) database
(http://www.ncbi.nlm.nih.gov).
Results Milk protein gene expression To examine if insulin is
essential for milk protein gene expression, mammary explants from
four late pregnant cows were initially cultured for 5 days in media
with no hormones (NH) to allow the effects of endogenous hormones
to subside, and then either hydrocortisone and prolactin (FP) or
insulin, hydrocortisone and prolactin (IFP) was added to the
culture media for 3 days. Analysis of gene expression by Affymetrix
microarray showed minimal induction of -s1-, -s2-, - and -casein
gene expressions in mammary explants cultured in NH for 5 days
(Fig. 1a). In explants cultured in FP for 3 days, there was no
change in expression level of -s2- and -casein genes compared to
explants cultured in NH. There was some increase in the expression
level of -s2- and -casein genes in explants cultured in FP compared
to NH. Maximum expression of all four casein genes in mammary
explants required insulin
Table 2 PCR primer sequences Gene FOLR ELF5 EIF4E SLC7A5 UXT
Primer 5 3 5 3 5 3 5 3 5 3 GCTGTGCCTTTTAGTGTGTGTG
TGGGCTTCTATGCTGGTGTT CATCCGCTCACAAGGTTACTC TCTTCCTTTGTCCCCACATC
GCGGCTCCACCTAAAA ACAAGACAAAGGCGAATGAGA TTCACTTCACCCTCACGTCTC
CCCCAACAAAACACAAAACTC GGTTGTCGCTAAGCTCTGTG TGTGGCCCTTGGATATGGTT
Size (bp) 183 287 196 204 101
202
Funct Integr Genomics (2009) 9:197217
A
16000
*-S1-casein g -S1-casein -S2-casein g-S2-casein -casein
K-casein
*
12000
8000
* * * * * * * * *
4000
0 NH FP IFP
hormone treatment
B
5000
-lactalbumin -lactoglobulin
* * *
4000
BLG required insulin in the presence of FP in the culture media.
The increase in -s1-casein gene expression in response to insulin
in the presence of FP in mammary explants was confirmed by Northern
analysis. Northern analysis showed -s1-casein transcripts were
undetectable in explants cultured in NH for 5 days and in explants
cultured in NH and FP for a further 3 days (Fig. 2a). Incubation of
explants with I together with FP for 3 days induced -s1-casein gene
expression. The level of milk protein gene expression was
correlated with the concentration of I in the presence of FP (Fig.
2b). Northern analysis showed there was no detectable expression of
the -s1-casein gene in explants cultured for 3 days in NH and
subsequently after a second incubation for 3 days in NH. Minimal
amounts of -s1-casein gene transcripts were observed when I was
included in the media at a concentration of 12.5 ng/mL, and the
level of gene expression increased progressively to maximum levels
when I was present at 1,000 ng/mL of media. Casein synthesis
intensity intensity
3000
2000
1000
* * *NH FP IFP
0
hormone treatmentFig. 1 Milk protein gene expression in response
to insulin in cultured mammary explants. Mammary explants were
cultured with no hormones (NH) for 5 days before culture in
hydrocortisone and prolactin (FP) or insulin, hydrocortisone and
prolactin (IFP) for 3 days. a Maximum expression of the major milk
casein genes in explants required I in the presence of FP. Minimal
expression of the casein genes is observed in explants cultured in
NH and some induction of -s2- and -casein gene expression occurred
in explants cultured with FP. b Maximum expression of the major
whey protein genes required I in the presence of FP in cultured
mammary explants. Minimal expression was observed in explants
cultured in NH and FP. Data is from the microarray of total RNA,
pooled for cows #1 and #2 and #3 and #4 and analysed using two
Affymetrix GeneChips for each hormone treatment. Gene intensity
levels are presented as meanrange between two GeneChips.
Significant differences in gene expression between hormone
treatments are: *P0.001, NH and FP; **P0.001, FP and IFP;
***P0.001, NH and FP
To investigate if insulin can prime the mammary gland for
I-independent milk protein gene expression, mammary explants from
seven cows were initially cultured in media with IF for 4 days
prior to 3 days culture in media with FP and IFP. There was no
expression of -s1-casein gene in explants cultured in media with IF
for 4 days or subsequent 3 day culture in IF (Fig. 3a). Induction
of -s1-casein gene expression in mammary explants cultured in only
FP occurred in explants from three cows. Transcripts of -s1casein
gene were observed in explants from all seven cows cultured in IFP,
and in mammary explants from cows #2 and #3, the amounts -s1-casein
gene transcripts were similar to that in explants cultured in FP
(Fig. 3a). Subsequent analysis of casein synthesis in explants from
the same treatment groups showed synthesis of casein proteins
occurred in explants cultured in IF for 4 and 7 days (254 and 205
dpm/mg tissue, respectively) (Fig. 3b). Minimal synthesis of casein
protein also occurred in explants cultured in FP for 3 days (205
dpm/mg tissue) whereas maximum synthesis occurred in explants
cultured in the complement of IFP (490 dpm/mg tissue). Lysine
uptake Two experiments were performed to investigate the potential
of insulin to stimulate uptake of 14C-lysine in cultured mammary
explants. The first experiment addressed whether the late pregnant
mammary gland responded to insulin for lysine uptake or acquired
the capacity to respond to insulin for lysine uptake. Mammary
explants were cultured for 48 h in either NH, I, FP or IFP and
14C-lysine
(I) in the presence of FP in the culture media. Minimal
induction of the whey protein genes, -lactalbumin (LALBA) and
-lactoglobulin (BLG), was observed in mammary explants cultured in
NH for 5 days and FP for 3 days (Fig. 1b). Maximum induction of
both LALBA and
Funct Integr Genomics (2009) 9:197217
20318S 28S
A
cow # 1
cow # 2
cow # 3
cow # 4
1172
NH5 NH8
FP
IFP
NH5
NH8
FP
IFP
NH5 NH8
FP
IFP
NH5
NH8 FP
IFP
Bcow # 7 cow # 8 cow # 9
18S 28S
1172
NH3 NH6 (i)
(ii)
(iii)
(iv)
(v)
NH3 NH6 (i)
(ii)
(iii)
(iv)
(v) NH3 NH6
(i)
(ii)
(iii)
(iv)
(v)
Fig. 2 The effect of insulin on -s1-casein gene expression in
cultured mammary explants. Total RNA (10 g) was assayed by Northern
analysis using an S1-casein labelled probe. Upper panels show equal
loadings as determined by visualisation of 28S and 18S ethidium
bromide stained ribosomal bands. a Mammary explants were cultured
with no hormones (NH) for 5 days before culture in hydrocortisone
and prolactin (FP) or insulin, hydrocortisone and prolactin (IFP)
for 3 days. The expression of -s1-casein (1,172 bp) is observed in
explants cultured in IFP and not in FP or NH. b Mammary
explants were cultured with no hormones (NH) for 3 days and then
in different insulin concentrations in the presence of FP for 3
days. Titration of insulin concentrations used in the mammary
explant culture medium are (i) I=12.5 ng/mL; (ii) I=25 ng/mL; (iii)
I=50 ng/ mL; (iv) I=100 ng/mL; (v) I=1,000 ng/mL. No induction of
-s1casein (1,172 bp) gene expression occurs in explants cultured in
NH. Minimal induction of -s1-casein is observed in 12.5 ng/mL of
insulin and maximum induction of -s1-casein occurs in explants
cultured in 1,000 ng/mL of insulin
uptake was measured at 06 h (referred to as time 0 h), 24 30 h
(referred to as 24 h) and 4854 h (referred to as 48 h). The rate of
14C-lysine uptake did not increase significantly in NH, I or FP
within 24 h (all P>0.05) (Fig. 4a), but did increase
significantly in explants cultured with IFP (P0.05). In contrast,
explants cultured in media with either I or IFP showed an increase
in 14C-lysine uptake (both P0.05). In the second experiment,
mammary explants from three cows were initially cultured in media
with NH for 2 days and then for 3 days in media with either I (100
ng/mL) alone, FP alone or I (50, 100, 1,000 ng/mL) with FP. The
rate of 14C-lysine uptake in explants cultured in NH for 2 days
represents a baseline value for lysine uptake (Fig. 4b). Subsequent
culture in NH or FP for 3 days did not change the rate of
14C-lysine uptake in the explants (both P>0.05). Insulin alone
at a concentration of 100 ng/mL stimulated uptake of 14C-lysine in
explants (P0.05). 14C-lysine uptake increased progressively in
explants cultured in FP with incremental concentrations of I to a
maximum rate when I was present at 1,000 ng/mL of media (P
