Genetic engineering of milk composition: modification of milk ...


Dec 10, 2012 (5 years and 5 months ago)


.1111 J C/ill Vitir l993:58(suppl):299S-306S.Printed in USA.© 1993 American Society for Clinical Nutrition
Genetic engineering of milk composition:modification of milk
components in lactating transgenic animals13
Heng-Chc’nl Yolli and Robert D Brc’mc’l
ABSTRACF Recent progress in recombinant DNA tech-
nology as well as in embryo manipulation and transfer has made
the introduction of specific genes into the germline of animals
relatively commonplace.With appropriate genetic constructs
expression of the inserted genes in transgenic animals can be
controlled in a tissue-specific and in a differentiation-specific is now possible to consider alteration of the
composition of milk produced by a lactating animal in any of
a variety ofways.There is a growing list offoreign milk proteins
that have been expressed,and one can envisage placing almost
any protein gene of interest under the control of the ci.s’-acting
promoter and enhancer elements ofa milk protein gene.Mod-
ifleation of milk composition can be extended not only to the
proteins of commodity value but manipulation of key
metabolic fat.lactose,and other minerals in
milk..#{149}1?nJ (‘liii Niur 1993:58(suppl):299S-306S.
KEY WORDS Transgenic animal,genetic engineering,gene
expression.promoter.milk protein,milk fat,lactose
In the future,food chemists will be able to design the char-
aeteristics of food materials at the level of production in either
plants or this technology develops there is a
need for food chemists to identify new and desirable character-
isties of milk ( I ) and other commodities.The ability to model
protein structures with computers is advancing rapidly and is
widely used in pharmaceutical development (2).Its application
to food-protein design is in some ways a simpler application
because the types ofphysieal changes might be more predictable
than the subtle changes imposed during the design of a drug.
For food proteins it is often possible to extract useful structure
and function information at a less-refined level without concern
for detailed molecular structure ( I).
Regulation of gene expression of milk proteins is a complex
process that involves peptide and steroid well as
interactions ofeell to cell and cell to substratum (3).No mam-
mary cell line has been developed that can mimic the situation
in vivo for the elucidation ofthis process.Consequently.the use
oftransgenic animals is an attractive model for the understanding
ofregulation ofmilk-protein gene expression.It was shown that
it is possible to express rat whey acidic protein (WAP) in mice
(4).rat/3-casein in mice (5,6),sheep/3-lactoglobulin in mice
(7).bovine a-lactalbumin in mice (8),and human proteins of
potential pharmaceutical value in the milk ofmiee (9.10),rabbits
(1 1),and sheep (12,13).Advantages ofthe mammary synthesis
ofproteins include the ability to produce them in large quantities
at low cost (7,14.15) and the potential to carry out the necessary
and appropriate posttranslational modifications,ie,phosphor-
ylation or glycosylation.required to obtain a biologically active
form of the protein ( 14).Given that mammary tissue cultures
that are fully differentiated are not available,the model of mam-
mary expression in transgenie animals is not only exciting for
its basic research potential hut also has a direct application for
biotechnology when information can be used in food and nu-
trition areas.
In the eases so far investigated.the genes for the foreign pro-
teins,once incorporated into germline.are transmitted as Men-
delian dominant characteristics and the animals produce the
proteins only in their milk during lactation.Transgenic rabbits
( 1 1,16),pigs ( 16- 18),sheep ( 12.17),and cattle ( 19) were re-
ported.Once further advances in gene transfer in this area are
made,one can easily envisage the propagation of new genes in
the dairy cattle population by use ofembryo transfer and artificial
insemination,which are currently widely used.This article will
focus on what types oftechnological capabilities currently exist
and are being developed and it will consider several issues whose
effect must be considered if the technology is to become com-
mercially viable.
Production of novel milk proteins
Studies have used the mouse (4-10).the rabbit ( I I),and the
sheep ( 12.13) to test the expression ofproteins in the mammary
gland.These are summarized in Table 1.Many kinds of proteins,
from bacteria to humans.are represented in this list.Murine
WAP promoter has been used for several studies with a consid-
erable variation of expression (9,2 1).The results from using
promoters from sheep (7) and cattle ( 10) are promising and par-
ticularly relevant for use in livestock species.The transgenie
mice carrying the genomie sheep 13-laetoglobulin gene secreted
sheep/(-laetoglobulin into their milk at as much as 3.6 times
the rate of endogenous/3-laetoglobulin expression (7).The
amount of expression of the transgenes seems to result from
I From the Department of Dairy Science.University of Wisconsin.
2 Supported by the Wisconsin Milk Marketing Board.
I Address reprint requests to RD Bremel.Department ofDairy Science.
University ofWisconsin.Madison.WI 53706.
by guest on December 9, 2012ajcn.nutrition.orgDownloaded from
Summary of proteins expressed in mammary glands of transgenic animals?
Promoter and
size Signal peptide
Gene (coding region)
size 3’ Region and size
animal Expression Reference
kb kb kb
Mouse WAP.Mouse c-inve,4.8 Mouse c-nice 7.3 Mouse 10% of endogenous 20
2.5 (including 3’)
Mouse WAP Human Ha-ras
7.4 Mouse 2% ofendogenous 21
Mouse WAP.Human P52 Human P52 (eDNA).Mouse WAP 8.0 Mouse 2.5% ofendogenous or 22
2.5 0.49
200 nmol/L (1.5 mg/
Mouse WAP.Human tPA Human tPA (eDNA) SV4O 4.9 Mouse 3-l4 nmol/L (I 14-460 9
2.6 zg/L)
0.6-1500 nmol/L (20-
50000?zg/L)or 5% of
Rat?t-casein.Bacterial CAT SV4O Mouse Expressed in mammary 6
2.3 or 0.5 including exon I and
intron A (0.49 kb) of
extract and thymus 23
Rat?l-casein.Rat?t-casein Rat?-casein (genomic),Rat Lt-casein,3.0 14.0 Mouse 0.01-1% ofendogenous 5
3.5 7.5 6
Rat WAP.0.95 Rat WAP Rat WAP (genomic).
Rat WAP.1.4 4.3 Mouse 1-95% ofendogenous 4
Rabbit 1-cascin.Human IL-2 Human lL.2 (genomic).Human lL-2 7.3
Rabbit 3-28 nmol/L (50-430 I I
Sheep BLG.4.0 Sheep BLG Sheep BLG (genomic).
Sheep BLG.7.3 or
16.2 or 10.5 Mouse 0.2-1.3 mmol/L (3-23 g/
Sheep BLG.4.0 Human FIX Human FIX (eDNA).
1.55 + 4.9 (BLG)
Sheep BLG,l.6 12.05 Sheep 0.004% ofhuman plasma
or 0.5 nmol/L (25?g/
Sheep BLG.4.0 Human i1-AT
Human (51-AT (eDNA).
.3 + 4.9 (BLG)
Sheep BLG.1.6 1 1.8 or 14.0 Sheep 120 nmol/L (S mg/L) 12
Bovine ALA.Bovine ALA Bovine ALA (genomic).Bovine ALA,0.336 3.1 Mouse 0.2-30 Mmol/L (2.5-450 8
0.75 2.0
Bovine (SSI-CN.Human UK Human UK (genomic).Bovine ast-CN.2.0
Mouse l8-36 Mmol/L (1-2 gIL) 10
21 7.5
* WAP,whey acidic protein:P52.breast cancer protein.estrogen-inducible secretory polypeptide:tPA.tissue plasminogen activator.CAT,chloramphenicol acetyltransferase:
lL-2.interleukin 2:BLG.?-Iactoglohulin:FIX.antihemophilic Factor IX:a1-AT.s1-antitrypsin:ALA.a-lactalbumin.c?s1-CN.?sucaSein:UK.urokinase.
many factors.which include cis-acting of inte-
gration of transgene on a chromosome,and possibly as-yet-un-
identified trans-acting factors.There is no reason to think that
similar results could not be attained with cows once technical
details have been worked out (24).The areas in which techno-
logical progress is required include mieroinjeetion.embryo eul-
ture and transfer,isolation of embryo-derived stem cells of do-
mestic animals,in vitro maturation of embryos.development
of viral vectors,development of strong mammary-specific vec-
tors,more-defined information on promoters and enhancers or
their interaction,and development of a mammary cell-culture
system to test amount ofexpression ofa construct.With present
technology the generation interval and the costs associated with
embryo injection and transfer in large animals are likely to pre-
elude their use.From what is known ofthe mammary function
in different mammals and the successes with cross-species
expression of milk genes.there is no reason to believe that the
technology will not be portable to cattle during this decade.
Characteristics of milk protein genes
It is apparent that the major regulatory elements that confer
mammary,the production ofproteins only in the
milk oflactating animals,reside in the upstream and downstream
flanking regions and internal regions ofa milk protein gene (4,
6).This location is not particularly surprising.because this work
draws on a large body of similar information obtained from
other molecular genetic research.What is striking,however,is
the high conservative homology of the 5’- and 3’-flanking se-
quences of the genes expressed in the mammary glands across
various species ofanimals (25-27).The amino acid sequence of
signal peptides for caseins from various species is highly con-
served,showing substantial homology among them (28).Their
intraspecies homology is as high as 100% at the amino acid level
(28) although somewhat lower at the nucleotide level (25) because
of silent mutations in which a change in a base does not give
rise to an amino acid substitution.The signal sequences of the
various lactoproteins share common features:high hydropho-
bicity and clustered hydrophobic amino acid residues (28).
Bovine casein genes are located on a single chromosome as a
gene cluster (29),probably chromosome 6.For all ofthe species
so far examined,the only caseins found are homologous to one
or another of four archetypal bovine caseins (25).From gene-
sequencing work on bovine and rat caseins.some common se-
quences have already been determined (30,3 1).The mutation
rate in the flanking regions is lower than that in the coding region
ofthe gene (32).This difference implies stronger functional re-
quirements for these regions throughout evolution than for the
structural coding region of casein itself.The leader peptide and
phosphorylation sites of three calcium-sensitive caseins (a51 ‘ a?2,
by guest on December 9, 2012ajcn.nutrition.orgDownloaded from
0.6-1.7 g/L
BSA 0.4g/L
and 1?) are conserved at the amino acid and nucleotide levels.
At the mRNA level,the untranslated regions are short at the 5’
end and long at the 3’ end,with the 5’-untranslated region’s
being conserved within each casein (25).The sequence homology
between bovine ssicasein and guinea pig f3-easein constitutes
90.5% for the 5’-nontranslated region,82.2% for the signal pep-
tide region.64% for the main polypeptide region.and 72% for
the 3’-noncoding region (26).The conserved sequences in the
5’-flanking region ofbovine 13-easein are identical or homologous
to the potential binding sites for nuclear factors and for gluco-
corticoid and progesterone receptors (33).The comparison of
casein mRNA sequences ofdifferent animals indicated that the
3’-untranslated region is more conserved compared with the
coding portion (31,34).
The sequences responsible for a tissue-specific and develop-
mental stage-specific expression reside in not only the 5,3’-
flanking region but also certain intragenie sequences (4,6).Ev-
idenee indicates that transgenes containing genomie sequences
were expressed more efficiently than those with eDNA (see Table
1).The effects ofintrons on gene expression were demonstrated
when 10- to 100-fold more mRNA was produced from the in-
tron-containing construct (35).The first intron of the human
collagen al(I) gene contains several positively and negatively
acting elements (36).As an increasing amount ofgene sequence
data becomes available,more commonality will probably be
found between the sequences in the different species.
Recent advances in cloning vectors for human genome with
the yeast artificial chromosome (YAC) now allow one to clone
?400 kb (37).Through adaptations of this technology it may
be possible to clone all the casein genes in a single vector and
to insert them into transgenic animals.Recently a procedure
was developed for introducing exogenous DNA into mouse eggs
by injection of chromosome fragments that facilitate incorpo-
ration ofvery large (> 10 megabases) DNA fragments into cells
and embryos without the need for cloned sequences (38).This
procedure can be applicable in cases in which it is desirable to
introduce a gene cluster or a gene that spans a great distance.
Potential changes in milk
The relative contents ofthe proteins in cow milk is shown in
Figure 1.There are two broad classifications of milk proteins,
casein and whey fractions.which result from their relationship
to cheese manufacturing.The classification ofmilks from various
animals is based on the fraction of the total proteins that pre-
cipitate at pH 4.6.The optimum pH for precipitation of caseins
from various species differs from pH 4.6 and thus the catego-
rization as casein and whey must be viewed with some skepti-
cism.For example,the optimum pH for precipitation of casein
from rodent milk is pH 4.0.From a manufacturing point of
view,the caseins are important because they form the curds in
cheese production and constitute?80% ofthe protein in milk.
The remainder are the so-called whey proteins because they do
not become associated with the curds in cheese.Both the caseins
and whey proteins provide several critical functions for pro-
cessing and handling offluid milk and manufactured milk prod-
ucts.Their functions include fat-globule emulsification,ionic
and colloidal mineral stabilization.pH buffering.cheese-curd
formation,heat-stability regulation.viscosity and gelation de-
velopment in cultured and sterile milk products,foam expansion
and stabilization of frozen dairy products,and control of ice
and lactose crystallization in frozen milk and ice cream products
FIG 1.Composition and concentration ofcow’s milk proteins.s-CN.
?i-lactoglobulin:ALA.(5-lactalbumin:BSA.bovine serum albumin:IgGs,
immunoglohulin Gs.The concentration ofeach protein was from Eigel
et al (39).
(40).Because single changes in amino acid composition can
alter functionality of protein markedly.any well-defined alter-
ation in one milk protein system is likely to produce changes in
the functions mentioned above.
Although there are hundreds of individual proteins in milk,
the proteins with economic value are really quite limited.As
shown in Figure 1.bovine caseins comprise four main classes:
(5??(40%).a52 (10%),/3 (40%),and K (10%).They are all phos-
phorylated to varying degrees:a52 ( 10-1 3P).?S (8-9P).[3 (5P)
and K ( 1P) (39).They are rich in proline.thus resulting in a
random coil conformational state that is resistant to heat-induced
denaturation but imparts a strong tendency for them to undergo
polymerization by hydrophobic,ionic,and Ca2?bonding (40).
The caseins have never been crystallized.Their uneven distri-
bution of polar and apolar domains makes them amphiphilie,
giving them particularly useful properties as emulsifiers ( I).
An obvious change in milk would be to selectively increase
one ofmilk protein components that is already present in milk.
One way that this change might be achieved would be by intro-
ducing extra copies of existing genes into the normal bovine
genome.For example.the increased proportion of normal and
engineered a51-easein will result in larger micelles,which may
alter cheese-curd characteristics in addition to enhancing the
rate of desirable textural development ( 1).Another modification
that might be considered would be a production of modified
a51-casein by introduction ofa chymosin-sensitive region through
site-directed mutagenesis ( I ).which can be achieved easily by
the polymerase chain reaction (PCR) with synthetic oligonuele-
otides.By incorporation ofadditional K-easein the easein micelle
size might be reduced.increasing the thermal stability to decrease
the degree ofcoagulation during sterilization process.Addition-
ally,removal ofsome phosphate groups ofeasein peptide may
yield a softer cheese ( 1).
by guest on December 9, 2012ajcn.nutrition.orgDownloaded from
H iie?i’ J)rolc’ins
A smaller portion of the milk proteins is the whey fraction,
which consists of/3-laetoglobulin (50%),a-laetalbumin (20%),
serum albumin ( 10%),and other minor proteins shown Figure
1.Whey proteins have a nutritional value in bottled milk but
represent a waste product for the cheese industry.Heat processing
of milk before rennet treatment (the initial step in cheese man-
ufacturing,when coagulation takes place) causes sufficient whey
protein denaturation and whey protein interaction with casein
micelles to inhibit the action of rennin.This reduced suseepti-
bility of s-easein to rennin is due to its formation of intermo-
lecular disulfide bonds between (3-lactoglobulin and K-easein (40).
For this reason it might be desirable to selectively eliminate the
(3-laetoglobulin fraction from milk.It is only present in the milk
ofruminant animals and sea mammals such as the dolphin (two
types) and the manatee (4 1) and is apparently not required for
lactation.Its presence in milk confers some undesirable man-
ufacturing properties (40),and thus its elimination could provide
an opportunity for new types of manufacturing practices and
development of novel milk products.
Au/k liii
As discussed above,the genetic engineering of milk compo-
nents is not limited to the commodity proteins.Through either
reduction or extinction of key enzymes in the synthesis of milk
fat,the fat content can be modified.An example might be ex-
tinetion ofacetyl coenzyme A (CoA) earboxylase,which regulates
the rate ofde novo fat synthesis from acetate within the mam-
mary gland,in which 50% of milk fats are made.It is a large
and highly complex enzyme whose genetic sequence was deter-
mined (42).A reduction in the amount of this enzyme would
lead to a dramatic reduction in the fatty acid content of milk
with a concomitant reduction in the energy requirements of the
animal producing the milk.Table 2 summarizes the economic
effects on dietary nutrition requirements for dairy cattle for milk
production if fat content were reduced from 3.8% to 2.0% by
specific inhibition of mammary fat synthesis.A similar strategy
for a removal of a-lactalbumin is also possible and will be dis-
cussed in the next section because it is coupled with lactose
The lactose content of cow milk is?5%,it has limited sol-
ubility,and its crystal is responsible for the sandiness defect in
ice cream (43).Because of lactose intolerance resulting from
absence of its hydrolyzing enzyme.?-galactosidase,many people
in the world cannot consume milk (43).The problem might be
overcome either by removal ofa-laetalbumin or by introduction
of an enzyme such as (3-galactosidase into milk.Either of these
strategies might be useful.
Unlike f3-lactoglobulin and acetyl CoA carboxylase,(5-lact-
albumin is necessary to the secretory process because lactose is
critical to the movement ofwater and thus most probably other
milk constituents through secretory cells (44).It is one subunit
oflactose synthetase that catalyses the synthesis oflactose.Once
secreted,lactose cannot permeate the luminal membrane of the
mammary alveolus and thereby establish an osmotic gradient
across the secretory cells.
There are several possible strategies through which these com-
ponents might be removed.A premature stop codon might be
Dietary requirements for production ofa hypothetical low-fat milk by
reduction offat synthesis in the mammary gland ofdairy cattle
3.8% Fat 2.0% Fat
Forage (% ofdry matter) 58.8 83.2
Concentrate (% ofdry matter) 41.2 16.8
Forage(%offeedcost) 44.4 74.7
Grain (% of feed cost) 48.9 20.8
Protein supplements (% of feed cost) 4.4 0.0
Feed cost (‘T of milk cost) 38.8 37.1
Milkprice($/kg) 0.23 0.19
Feed cost ($/kg of milk) 0.09 0.07
inserted into the a-lactalbumin gene,leading to abortive synthesis
ofthe gene product through site-specific mutagenesis within the
eukaryotic genome.A second strategy is by the introduction of
an antisense sequence (45) coupled with a mammary-specific
promoter.The transcriptional product of the antisense sequence
ofthe a-lactalbumin gene is an mRNA whose complementarity
with endogenous a-laetalbumin mRNA would give rise to an
RNA-RNA blocking translation ofthe endogenous mRNA.
.4 niibodj’ production in,ni/k
It is possible to consider intervention in human and animal
health problems through the modification of milk.As Figure 1
indicates,bovine milk normally contains a considerable amount
ofantibodies [immunoglobulins G (IgG) types 1 and 2,A,and
M:ref39].Pathogenic organisms in milk and other parts of the
food supply continue to be a problem and the transgenic tech-
niques described can be envisioned as having application in this
area as well.A gene coding for a monoelonal antibody against
a certain antigen was expressed in other (non-antibody-produc-
ing) cells under the control of a specific promoter.Fragments
of IgG involved in antigen binding (Fab) were expressed in
E.scheric/iia (‘0/i to give specific binding to antigen and anti-
phosphoryleholine antibody was produced in transgenic mice
(46).Therefore production of monoclonal antibodies in the
mammary gland might be reasonably achieved by introduction
of the genes coding for antibodies under the control of a mam-
mary-specific promoter.Likewise,antibodies against enteric
pathogens such as salmonella,lysteria,or others could be pro-
Production n/human,ni/k proteins
The composition ofhuman milk is significantly different from
cow milk.It has a protein content among the lowest (0.9%) in
mammals and whey proteins are the major protein,in contrast
to bovine milk,in which easeins predominate (47).By use of a
transgenie model with a mammary gland with bovine genes,
human milk proteins can be produced in milk by the mammary
glands offarm animals to substitute for human milk or to make
components for an improved infant formula.Whey protein
concentrates and modified whey products are used as functional
ingredients in infant formula (40).For this purpose,a similar
strategy as that for bovine caseins can be applied to increase
exclusively bovine whey fraction in milk or to produce a human
whey protein such as a-laetalbumin.Another example for pro-
duetion ofa human milk protein component might be lactofer-
rim (48).
by guest on December 9, 2012ajcn.nutrition.orgDownloaded from
Chemical modifications of milk protein have provided the
food scientist with insights on structure-function relationships
in many food systems (49).Genetic engineering ofmilk proteins
can provide new insights into the production of novel food
products:however.many challenges remain ifthe composition
of milk is to be effectively altered.More detailed information
on the actions of promoters and enhancers in eukaryotic genes
has to be accumulated.and cis-acting and trans-acting elements
must be defined.These will permit transgene constructs with
strong,highly efficient,mammary-specific promoters and en-
hancers,which will direct their high expression in mammary
gland.The promoters that are currently used for the production
oftransgenie animals and cellular gene-transfection experiments
are well characterized and are effectively used in various cell-
culture systems.Rather limited information is available on in-
teraction among mammary-specific promoters.enhancers,lran.s-
acting elements,and intragenic sequences.As shown in Table
1.the high expression of sheep (3-lactoglobulin in mouse milk
by use ofregulatory elements ofsheep/3-lactoglobulin (7) is en-
couraging because it demonstrated the development of mam-
mary-specific targeting under the control of milk protein pro-
moters.In other work,human urokinase expression by use of
long 5’ upstream control sequences of bovine easein to drive
mammary-specific expression demonstrates that this technique
may have general applicability to proteins foreign to mammary
glands ( 10).
Probably the greatest hindrance to the immediate commercial
application of the transgenie technology is the relatively low
efficiency ofproduetion oftransgenie offspring (mice 1-2%.sheep
0.5%).The improvement of the efficiency of gene transfer is
required for the process to be economically practical in agri-
cultural systems in which the techniques can be much more
complicated than those ofexperimental models.With the evo-
lution of methods for introduction of DNA constructs into so-
matic cells it might be possible to produce mammary cells with
modified functions (transgenic mammary glands) without the
necessity ofprodueing transgenic animals carrying genes in their
germs cells.For example,if DNA constructs could be introduced
directly into the mammary cells in nonlactating animals it would
then obviate the necessity to produce transgenic animals.
There is a need for improved specificity of gene insertion.
This can also be accomplished by gene targeting through ho-
mologous recombination in which viral promoters can be flanked
by sequences of multiple-copy genes at which transgenes can be
inserted by a loop-in mechanism.Taken together.the aecu-
mulated data show that the expression oftransgenes in transgenie
animals is not proportional to the copy number inserted.For
commercial application in domestic livestock,the improvement
in the efficiency of transgenie production must occur concur-
rently with an increased specificity ofgene insertion.The location
ofinsertion into the genome and amount ofexpression are critical
to any scenario for introduction of genes into a population of
animals.So far one ease has supported such an argument:the
human 13-globin gene was expressed in a tissue-specific manner
in transgenie mice at a rate directly related to its copy number
yet independent of its position of integration (50).Transgenes
appear to segregate as Mendelian dominants.An agriculturally
useful system of genetic constructs must be correlated with
quantitative traits.A transgenic cow with a significantly lower
production of milk would be of little economic value to a com-
mercial dairy producer.Likewise breeding schemes must be de-
veloped so that the gene can be maintained in the population.
One ofthe limiting steps in the production oftransgenie an-
imals is the ability to quickly screen potential transgenic offspring
that may have only one copy of the gene of interest in their
genome.The use ofPCR in which two oligonucleotides are used
to amplify specific sequences between them (5 1 ) is encouraging
because it allows one to quickly screen the positive transgenie
animals.Also,once sequences are known it allows one to obtain
a large amount of DNA without cloning in a vector.Figure 2
shows an example of PCR screening of transgenic mice con-
taming bovine a51-casein genes.It takes only 2-3 d instead of
weeks to screen a large number of animals.
As mentioned earlier,considerable progress was reported for
cloning a big fragment ofgenomie DNA with YACs (37).Ad-
vances in generation of large combinatorial phage libraries of
immunoglobulin repertoires in X-phage demonstrated that genes
for antibodies primed by specific antigens can be easily cloned
(52).All ofthese technologies can be applied to facilitate engi-
neering of transgenes to be expressed in mammary glands of
transgenie animals.
There are other ways that we might be able to combine mo-
lecular genetic technology with a more traditional animal breed-
ing scheme.Within the past several years milk analysis has shifted
from the traditional Babcock test to the use of multispectral
infrared analyzers that make it possible to estimate the various
individual components in milk.Widespread use of this tech-
nology in the management of dairy cattle has made it possible
to screen several million animals and to select animals in the
population that have milk with unique characteristics.An ex-
ample ofthis is shown in Figure 3.These data are the result of
a preliminary screen of?650 000 Holsteins in Wisconsin whose
milk was tested for both fat and protein content.To be included
in the data set the animal had to have been tested on three
separate occasions.The points in the graph represent the unusual
animals in the population.Traditional animal-breeding programs
have not evaluated these animals.Because the protein compo-
sition in milk is encoded by only a few genes.careful analysis
ofthe genes either by restriction fragment-polymorphism studies
(53) or gene sequencing should be ofeonsiderable interest.It is
likely that a mutation in one of the genes might give rise to
elevated expression.
The protein content of the milk of laboratory animals is typ-
ically > 10%.Ifthe protein content ofthe milk ofdairy animals
could be increased to similar concentrations through genetic
engineering then changes would be required in dairy-cattle diets.
This situation is not limited to dairy cattle.Laboratory animal
diets may have to be evaluated for protein sufficiency if additional
proteins are to be expressed in their milk.Were it possible to
specifically inhibit fat example of how the nutri-
tional requirements might be expected to change can be seen in
Table 2.Approximately halfthe fat found in milk is synthesized
de novo from 2C and 4C precursors in the mammary gland.If
we were to inhibit acetyl CoA carboxylase through gene targeting
what would then happen to the nutritional requirements of the
animals?The system used to feed cattle is based on the as-
sumption that fat and protein in milk are highly coupled and
by guest on December 9, 2012ajcn.nutrition.orgDownloaded from
0 1 2 3 4 5 6 7 8 9 101112
FIG 3.Protein and fat percentages of milk from unusual Holstein
cows in a Wisconsin population.
1353 -
1078 -
872 -
603 -
FIG 2.Screening of transgenic mice by the polymerase chain reaction (PCR).Lane M.molecular weight marker
[liac Ill-cleaved?X-I 74 restriction fragment (RF) DNA].The notations on the left indicate the sizes ofthe molecular
weight marker ( I 353,1078.872.and 603 bp).Lane Cl.without mouse DNA:lane C2.with nontransgenic mouse tail
DNA:lane 1- I 3,tail DNA ofoffspring from a transgenic founder.The products ofthe amplified transgene are indicated
by a thick arrow on the right (584 bp).Lanes I.2,5.6,8,9,10,and 1 1 show the amplified transgene.
therefore need not be considered independently.To simulate
this situation the fat and protein coefficients were uncoupled
and rations were generated with the help of a ration-balancing
program from normal feedstuffs with linear-programming tech-
niques.maintaining the protein content at a similar value to
that normally found but decreasing the fat content (54).What
arises from this simulation is that the diet of the animal will
change dramatically.The model predicts that the same milk
volume and milk protein production could be maintained while
suppressing fat on a diet consisting predominantly of forage.
The model predicts that the forage level in the diet could exceed
80%.We recognize that this prediction is totally hypothetical,
but the potential effect could be enormous.
Any system introducing new genes into the population will
have potential ramifications on animal breeding.The high pro-
ductivity of the dairy-cattle population is testament to the
breeding strategies that have been developed and implemented
over the past several decades.A number of scenarios for the
introduction of new proteins into milk were outlined above.If
the method ofintroduetion ofthese genes is through transgenic
animals then it will be necessary to consider how these genes
will be propagated in the population.Scenarios such as those
described depend very heavily on the technological capabilities
available when the scenario is developed.It will be possible to
evaluate some of the changes in milk in laboratory animals so
that there is a higher probability of a desirable outcome when
introduced into lactating animals.The production of rare bio-
logicals with sufficiently high economic values will most likely
drive the development of the technological capabilities described
in this article.After this it will become feasible to develop small,
specialized populations of dairy cattle.sheep.and goats for the
production of specialty milks.
Various aspects for the modification of milk components by
use of lactating transgenie animals were discussed.At present,
the efficiency ofproduetion is low and the amount of expression
is not commercially feasible.However,as further information
both at the gene level and on transfer technology accumulates
the application of mammary expression in transgenic livestock
will have a substantial effect on biotechnology and related in-
dustries,such as farming.processing.and nutrition.The major
challenge for scientists from related areas is to economically
produce modified or engineered milk proteins with improved
function that pose no health hazard to producing animals or the
consuming public and retain a high nutritional quality.[3
by guest on December 9, 2012ajcn.nutrition.orgDownloaded from
We are grateful to George Shook for fat and protein data for Wisconsin
dairy cattle.We also thank Terry Howard for use ofhis ration-balancing
program to produce the hypothetical dairy cattle rations.
1.iimenez-Flores R.Richardson T.Genetic engineering ofthe caseins
to modify the behavior of milk during processing:a review.i Dairy
Sci 1988:71:2640-54.
2.Committee on Computer-Assisted Modeling,Board on Basic Biology,
Commission ofLife Sciences,National Research Council.Computer-
assisted modeling:contributions of computational approaches to
elucidating macromolecular structure and function.Washington.
DC:National Academy Press,1987.
3.Eisenstein RS,Rosen iM.Both cell substratum regulation and hor-
monal regulation of milk protein gene expression are exerted pri-
marily at the posttranscriptional level.Mol Cell Biol 1988:8:3 183-
4.Bayna EM.Rosen JM.Tissue-specific.high level expression of the
rat whey acidic protein gene in transgenic mice.Nucleic Acids Res
5.Lee K-F.DeMayo I.Atiee SH.Rosen JM.Tissue-specific expression
of the rat/3-casein gene in transgenic mice.Nucleic Acids Res
6.Lee K-F,Atiee SH.Henning Si.Rosen JM.Relative contribution
of promoter and intragenic sequences in the hormonal regulation
ofrat f3-casein transgenes.Mol Endocrinol 1989:3:447-53.
7.Simons JP.McClenaghan M,Clark Ai.Alteration ofthe quality of
milk by expression ofsheep?i-lactoglobulin in transgenic mice.Na-
ture 1987:328:530-2.
8.Vilotte J-L.Soulier S.Stinnakre M-G,Massoud M.Mercier J-L.
Efficient tissue-specific expression ofbovine a-lactalbumin in trans-
genie mice.Eur i Biochem 1989:186:43-8.
9.Gordon K.Lee E.Vitale IA.Smith AE.Westphal H,Hennighausen
L.Production of human tissue plasminogen activator in transgenic
mouse milk.Biotechnology (NY) 1987:5:1 183-7.
10.Meade H.Gates L,Lacy E.Lonberg N.Bovine asicasein gene se-
quences direct high level expression of active human urokinase in
mouse milk.Biotechnology (NY) 1990:8:443-6.
1 1.Buehler TA.Bruyere T.Went DF.Stranzinger G,Buerki K.Rabbit
13-casein promoter directs secretion ofhuman interleukin-2 into the
milk of transgenic rabbits.Biotechnology (NY) 1990:8:140-3.
12.Simons JP.Wilmut I.Clark AJ,Archibald AL,Bishop JO,Lathe
R.Gene transfer into sheep.Biotechnology (NY) 1988:6:179-83.
I 3.Clark Al,Bessos H,Bishop al.Expression of human anti-
hemophilic factor IX in the milk oftransgenic sheep.Biotechnology
(NY) 1989:7:487-92.
14.Bremel RD.Yom H-C.Bleck GT.Alteration ofmilk composition
using molecular genetics.I Dairy Sci 1989:72:2826-33.
I 5.Pittius CW,Hennighausen L.Lee al.A milk protein gene
promoter directs the expression of human tissue plasminogen acti-
vator eDNA to the mammary gland in transgenic mice.Proc NatI
Acad Sci USA 1988:85:5874-8.
16.Brem G.Brenig B.Goodman al.Production of transgenic
mice.rabbits and pigs by microinjection.Zuchthygiene 1985:20:
25 1-2.
I 7.Hammer RE.Pursel VG,Rexroad CE al.Production of trans-
genie rabbits.sheep and pigs by microinjection.Nature I 985:315:
I 8.Petters RM.Shuman RM.Johnson BH.Mettus RV.Gene transfer
in swine embryos by injection ofcells infected with retrovirus vectors.
I Exp Zool 1987:242:85-8.
19.Roschlau K,Rommel P.Andreewa L,et al.Gene transfer experi-
ments in cattle.I Reprod Fert [Suppl] 1989:38:153-60.
20.Schoenenberger C-A,Andres A-C,Groner B,van der Valk M.
LeMeur M.Gerlinger P.Targeted c-?ni’c gene expression in mam-
mary glands of transgenic mice induces mammary tumours with
constitutive milk protein gene transcription.EMBO I 1988:7:169-
21.Andres A-C,Schoenenberger C-A,Groner B,Hennighausen L.
LeMeur M,Gerlinger P.Ha-ras oncogene expression directed by a
milk protein gene promoter:tissue specificity,hormonal regulation,
and tumor induction in transgenic mice.Proc NatI Acad Sci USA
22.Tomasetto C,Wolf C,Rio M-C,et al.Breast cancer protein PS2
synthesis in mammary gland oftransgenic mice and secretion into
milk.Mol Endocrinol 1989:3:1579-84.
23.Lee KF,Atiee SH,Rosen iF.Differential regulation ofrat/3-casein-
chloramphenicol acetyltransferase fusion gene expression in trans-
genie mice.Mol Cell Biol 1989:9:560-5.
24.Simons JP.Land RB.1987.Transgenic livestock.J Reprod Fert
[Suppl] 1987:34:237-50.
25.Bonsing I,Mackinlay AG.Recent studies on nueleotide sequences
encoding the caseins.J Dairy Res 1987:54:447-61.
26.Gorodetskii SI,Zakhar’ev VM,Kyarshulite DR.Kapelinskaya TV,
Skryabin KG.Cloning of nucleotide sequence of eDNA for bovine
a51-casein.Biokhimiia 1986:51:1641-8.
27.Stewart AF.Bonsing I.Beattie CW.Shah F,Willis IM,Mackinlay
AG.Complete nucleotide sequence of bovine?S2?and/3-casein
cDNAs:comparisons with related sequences in other species.Mol
Biol Evol 1987:4:231-41.
28.Gaye PP.Mercier JC.The functional significance of signal sequence:
their role in the transfer of lactoproteins across the endoplasmic
reticulum membrane.Reprod Nutr Dev 1981:21:199-208.
29.Matyukov VS.Urnyshev AP.Linkage of ass-./3-’ s-casein loci in
cattle.Genetika 1980:16:884-6.
30.Hobbs AA,Rosen IM.Sequence of a- and -y-casein mRNAs:evo-
lutionary comparison of the calcium-dependent rat casein multigene
family.Nucleic Acids Res 1982:10:8079-98.
3 1.Stewart AF,Willis IM,Mackinlay AG.Nucleotide sequences of bo-
vine?and s-casein cDNAs.Nucleic Acids Res 1984:12:3895-
32.Yu-Lee L.Richter-Mann L.Couch CH.Stewart AF,Mackinlay
AG.Rosen JM.Evolution ofthe casein multigene family:conserved
sequences in the 5’ flanking and exon regions.Nucleic Acids Res
33.Gorodetsky SI.Tkaeh TM.Kapelinskaya TV.Isolation and char-
acterization of the Bos taurus (3-casein gene.Gene 1988:66:87-96.
34.Bayev AA,Smirnov 1K.Gorodetsky SI.The primary structure of
bovine/3-casein eDNA.Mol Biol (Mosk) 1987:21:255-65.
35.Brinster RL,Allen JM.Behringer RR,Gelinas RE,Palmiter RD.
lntrons increase transcriptional efficiency in transgenic mice.Proc
NatI Acad Sci USA 1988:85:836-40.
36.Bornstein P.Mckay I.Liska DJ,Apone S.Devarayalu S.Interactions
between the promoter and first intron are involved in transcriptional
control of al(I) collagen gene expression.Mol Cell Biol 1988:8:
485 1-7.
37.Brownstein BH.Silverman GA,Little al.Isolation of single-
copy human genes from a library of yeast artificial chromosome
clones.Science 1989:244:1348-51.
38.Richa I.Lo CW.Introduction of human DNA into mouse eggs by
injection ofdissected chromosome fragments.Science 1989:245:175-
39.Eigel WN.Butler JE.Ernstrom al.Nomenclature of proteins
ofeow’s milk:fifth revision.I Dairy Sci 1984:67:1599-631.
40.Mon CV.Functionality ofheated milk proteins in dairy and related
foods.I Dairy Sci 1985:68:2773-81.
4 1.Pervaiz S.Br#{149}?K.Purification and characterization of the major
whey proteins from the milks of the bottlenose dolphin ( Tursiop.s
iruncalus),the florida manatee ( Tricheclius inanalu.s’ latirosiris),
and the beagle (Canisf#{224}iniliari.s).Arch Biochem Biophys 1986:246:
by guest on December 9, 2012ajcn.nutrition.orgDownloaded from
42.Lopez-Casillas F.Bai DH,Luo X.Kong IS.Hermondson MA,Kim
KH.Structure of the coding sequence and primary amino acid se-
quence ofacetyl-eoenzyme A carboxylase.Proc Natl Acad Sci USA
43.Kang Y.Jimenez-flores R.Richardson T.Casein genes and genetic
engineering ofthe caseins.In:Evans JW,Hollaender A.eds.Genetic
engineering of animals:an agricultural perspective.New York:
Plenum Press,1986:95-111.
44.Kuhn NJ.The biosynthesis of lactose.In:Mepham TB,ed.Bio-
chemistry oflactation.Amsterdam:Elsevier Science Publishers,1983:
45.Nishikura K.Murray JM.Antisense RNA ofproto-oneogene c-fos
blocks renewed growth ofquiescent 3T3 cells.Mol Cell Biol 1987:7:
46.Pinkert CA.Manz I,Linton P-i.Klinman NR,Storb U.Elevated
PC responsive B cells and anti-PC antibody production in transgenic
mice harboring anti-PC immunoglobulin genes.Vet Immunol Im-
munopathol 1989:23:321-32.
47.Bounous G.Kongshavn PAL,Taveroff A.Gold P.Evolutionary
traits in human milk proteins.Med Hypotheses 1988:27:133-40.
48.Mercier i-C.Genetic engineering applied to milk producing animals:
some expectations.In:Smith C.King JWB,Mckay iC,eds.Ex-
ploiting new technologies in animal breeding.London:Oxford Uni-
versity Press,1986:121-31.
49.Richardson T.Chemical modification and genetic engineering of
food proteins.J Dairy Sci 1985:68:2753-62.
50.Grosveld F,van Assendelft GB,Greaves DR.Kollias 0.Position-
independent.high-level expression of the human/3-globin gene in
transgenic mice.Cell I 987:51:975-85.
5 1.Saiki RK,Gelfand DH,Stoffel S.Scharf Si,Higuchi R,Horn GT.
Primer-directed enzymatic amplification ofDNA with a thermostable
DNA polymerase.Science 1988:239:487-91.
52.Huse WD,Sastry L.Iverson al.Generation of a large com-
binatorial library ofthe immunoglobulin repertoin in phage lambda.
Science I 989:246:1275-81.
53.Cowan CM.Dentine MR.Ax RL,Schuler LA.Structural variation
around prolactin gene linked to quantitative traits in an elite Holstein
sire family.Theor AppI Genet 1990:79:577-82.
54.Howard WT,Shaver RD.Wallenzien AC.WISPLAN TMR FIXED.
Madison,WI:University of Wisconsin-Extension,WISPLAN
Computing Service.1992.
by guest on December 9, 2012ajcn.nutrition.orgDownloaded from