(Received for publication, September 27, 1996, and in revised form, January 4, 1997)
From Human lactoferrin (hLF) is an iron-binding
protein involved in host defense against infection and severe
inflammation. Transgenic mice were produced harboring either hLF
cDNA or genomic hLF sequences fused to regulatory elements of the
bovine Based on these observations, we anticipate that (unsaturated) rhLF and
natural hLF will exert similar, if not identical, antibacterial and
anti-inflammatory activity in vivo.
Human lactoferrin (hLF)1 is a
single-chain metal binding glycoprotein of Mr
77,000 belonging to the transferrin family (1, 2). Lactoferrin consists
of two highly homologous lobes, designated the N- and C-lobe, each of
which can bind a single ferric ion concomitantly with one bicarbonate
anion (2). The amino acid sequence of hLF has been determined by
protein and DNA sequencing (1, 3) and its tertiary structure elucidated
by x-ray crystallography (2). N-Linked glycosylation of hLF
is of the sialyl N-acetyllactosaminic type (4). Human LF was
first isolated from milk (5) but has also been found in many other
external secretions such as tears, saliva, pancreatic juice, vaginal
secretion, and seminal fluid, as well as in the specific granules of
neutrophils (6). Extensive in vitro and some in
vivo evidence indicates that hLF is involved in host defense
against infection and severe inflammation, most notably at mucosal
surfaces such as those of the gastrointestinal tract (6). In
vitro antimicrobial activities of LF include bacteriostatic
activity (7), probably residing in the capacity for high affinity
binding of iron; bactericidal activity (8) probably mediated through
direct binding of LF to bacterial cell wall components such as
lipopolysaccharide (LPS) (9, 10); synergy with other antibacterial
milk components such as human lysozyme (10) and immunoglobulin (11);
and fungicidal activity (12). Human LF has been shown to promote the
growth of Bifidobacterium species (13), the predominant
bacteria of the intestinal flora of healthy breast-fed infants. The
in vitro anti-inflammatory activities of LF include
inhibition of the formation of hydroxyl radicals by scavenging free
iron (14), inhibition of cytokine production (15), and binding to the
inflammatory mediator LPS (9, 10, 16). Lactoferrin has been shown to
promote the growth of intestinal cells in vitro (17). The
function of specific LF receptors on intestinal cells and lymphocytes
is as yet unclear (6).
The prospect of a wide variety of potential applications in human
healthcare has motivated several investigators to study the feasibility
of large-scale production of functional recombinant hLF. Here we
describe the characterization of hLF secreted in milk of transgenic
mice. Transgenic hLF was found to be identical with natural hLF derived
from human milk by immunological and functional criteria. Structurally
it differed only in glycosylation and the degree of iron
saturation.
CNBr-activated Sepharose 4B, S-Sepharose,
concanavalin A-Sepharose, heparin-Sepharose CL-6B, chelating Sepharose
fast flow and blue Sepharose CL-6B were obtained from Pharmacia Biotech Inc. (Uppsala, Sweden). Calf thymus single-stranded DNA-agarose was
purchased from IBF Biotechnics (Villeneuve-la-Garenne, France). Ulex europaeus agglutinin (UEA) type I-agarose was purchased
from Sigma.
Mammary gland-specific
expression vectors based on regulatory elements from the bovine
A quantitative RIA for hLF
with anti-hLF monoclonal antibody (mAb) 13.17 coupled to Sepharose was
performed essentially as described (23). Briefly, mAb 13.17-Sepharose
suspension was incubated for 16 h by head-over-head rotation with
test samples. A minor modification was that serial dilutions of natural
hLF (standard) and mouse milk were made in phosphate-buffered saline, 0.1% Tween 20, 5 M NaCl to disrupt electrostatic
interactions of hLF with murine milk components. Sepharose beads were
then washed and hLF bound to Sepharose was detected by incubation with polyclonal rabbit 125I-anti-hLF for 6 h. After
washing, Sepharose-bound radioactivity was measured. Results were
expressed as percentage binding of the total amount of labeled
antibodies added.
Male and female nontransgenic mice,
cDNA hLF (line 33) and genomic hLF transgenic mice (line 937) were
given repeated intraperitoneal injections with 50 µg of purified
natural hLF, either in complete Freund's adjuvant (first injection) or
incomplete Freund's adjuvant (booster injections: 3, 6, and 9 weeks
after the first injection). Essentially the same hyperimmunization
protocol has previously been used to prepare murine mAbs (24). Ten days
after the fourth injection, blood samples were obtained, and sera were
tested for the presence of specific antibody by a RIA, in which
goat-anti-mouse immunoglobulin coupled to Sepharose was incubated with
serial dilutions of mouse sera together with 125I-hLF. The
anti-hLF antibody responses of individual mice were expressed as
percentage of the mean response, arbitrarily defined as 100%, in the
nontransgenic mice.
Freshly prepared
iron-nitrilotriacetic acid solution was used to saturate hLF purified
from fresh human milk (designated as "natural hLF") with iron as
described (16). Saturation of natural hLF with iron (designated as
"iron-saturated natural hLF") was complete as assessed by
absorbance measurement and nonreduced SDS-PAGE of nonboiled samples
(25).
Nonreduced SDS-PAGE was performed as described
(16). In some cases, samples were boiled for 5 min in nonreducing SDS
sample buffer (26) prior to SDS-PAGE analysis to achieve concomitant denaturation and desaturation of hLF (25). Reduced SDS-PAGE was
performed after boiling of samples for 5 min in SDS sample buffer
containing 5% (v/v) Absorption coefficients of
A0.1%280 = 1.1 and
A0.1%280 = 1.4 were used for
unsaturated and iron-saturated hLF, respectively, and
A0.1%465 = 0.058 for
iron-saturated hLF (27).
The concentration of iron-saturated hLF (in mg/ml) in purified hLF was
calculated by dividing the A465 by 0.058. The
concentration of unsaturated hLF (in mg/ml) was calculated as:
(A280 Iron-saturated natural hLF, cDNA transgenic hLF, and
genomic transgenic hLF were diluted 7-fold in buffers of varying pH to final concentrations of 4.2, 3.9 and 4.3 mg/ml, respectively. Samples
were incubated at 20 °C and the A465 measured
continuously. Results are expressed as percentage desaturation by
reference to iron-saturated natural or transgenic hLF diluted in 0.15 M NaCl. Buffers were 0.1 M MES, 0.15 M NaCl of pH 6.1, 0.1 M sodium acetate, 0.15 M NaCl of pH 5.1, pH 4.5, pH 4.0, and of pH 3.7, and 0.1 M glycine-HCl, 0.15 M NaCl of pH 3.1, pH 2.5, and of pH 2.1. Complete removal of iron from transgenic and
iron-saturated natural hLF was performed by dialysis of preparations
against 0.1 M glycine-HCl, 0.15 M NaCl, pH 2.5 for 2 h at 30 °C.
Serial dilutions of purified hLF (1 µg/ml)
in phosphate-buffered saline, 0.1% Tween 20 were incubated with ligand
coupled to agarose as described above for the RIA of hLF. Bound hLF was detected by subsequent incubation with polyclonal
125I-anti-hLF antibodies. The response with natural hLF in
each RIA was arbitrarily defined as 100%. From the parallel
dose-response curves, the reactivity of iron-saturated natural hLF,
cDNA transgenic hLF, and genomic transgenic hLF to each ligand was
expressed as a percentage by reference to natural hLF. All RIAs were
performed with 0.5 ml of agarose suspension (2 mg of agarose/ml of
phosphate-buffered saline, 0.1% Tween 20, 0.02% sodium azide).
Ligand-agarose was either commercially obtained or prepared as follows:
anti-hLF mAbs E1 and E2 were partially purified from hybridoma culture supernatant by precipitation with 50% ammonium sulfate and coupled to
Sepharose (20 mg of protein to 1 g of CNBr-activated Sepharose 4B); 15 mg of purified rabbit anti-hLF was coupled to 1 g of
Sepharose; Fe2+-Sepharose was prepared by incubating a
solution of FeSO4 with chelating Sepharose fast flow;
Salmonella minnesota Re595 LPS-Sepharose was prepared as
described (16). The specificity of hLF interaction with solid-phase
ligand was demonstrated by RIAs in which Sepharose to which no ligand
had been coupled ("glycine-Sepharose") was used (16).
Transgenic mice expressed hLF in a mammary gland-specific and
lactation-restricted fashion (18) at levels in milk up to 13 mg/ml.
This paper describes the characterization of the transgenically produced protein. Fig. 1 shows dose-response curves of
natural hLF and milk samples of genomic hLF (A) and cDNA
hLF (B) transgenic mice in the RIA for hLF. The identical
slopes and maximal responses in the RIA indicate that the antigenic
determinants for monoclonal and polyclonal anti-hLF antibodies in
natural and transgenic hLF are equally well accessible. Natural and
transgenic hLF thus appear immunologically identical, and levels of hLF
in transgenic mouse milk can be calculated by reference to natural
hLF.
The immunological identity of transgenic hLF with natural hLF was
confirmed by immunizing of transgenic mice with natural hLF (Fig.
2). No anti-hLF response was detectable (lower than 0.25% of that in controls) in 9 out of 11 transgenic mice, whereas a
strong antibody response was observed in all six nontransgenic mice
(40,000- fold diluted sera still showed anti-hLF activity). These
results indicate that natural hLF is considered "self" by the
immune system in the transgenic mice.
Analytical chromatography of transgenic mouse whey
and human whey on a Mono S cation-exchange column (16) revealed that transgenic and natural hLF had completely bound to the column and that
all mouse and human whey proteins other than hLF had eluted at a salt
concentration of 0.3 M NaCl. Therefore, transgenic hLF as
well as natural hLF were isolated by batchwise extraction with
S-Sepharose from milk to which 0.4 M NaCl had been added (16). Unless stated otherwise, "cDNA transgenic hLF" and
"genomic transgenic hLF" refer to rhLF isolated from pooled milk of
transgenic mouse lines 33 and 937, expressing hLF in milk at about 0.2 and 2.5 mg/ml, respectively.
Purified transgenic and natural hLF were then subjected to Mono S
chromatography to exclude that N-terminal degradation of hLF had
occurred during the purification process (16). Both cDNA (Fig.
3) and genomic transgenic hLF (not shown) as well as natural hLF (see Ref. 16) eluted as single peaks at exactly the same
position (0.7 M NaCl), i.e. the position
characteristic of uncleaved hLF as was confirmed by N-terminal protein
sequence analysis. This analysis also revealed that the bovine
The
binding of metal ions to hLF is accompanied by a change in its
conformation (2). The SDS-PAGE migration of iron-saturated hLF, which
has a more globular structure (2), is faster than that of unsaturated
hLF, if the analysis is carried out under nonreducing conditions with
freshly diluted, nonboiled samples (25). Boiling in the presence of SDS
brings about denaturation of the iron-saturated molecule with release
of iron (25). Boiled samples of purified genomic and cDNA
transgenic hLF migrated as two doublets of protein bands on nonreduced
SDS-PAGE (Fig. 4, lanes 1 and 2),
whereas only two bands were observed with natural hLF (lanes
3 and 4). The ratio of the minor to the major band in
natural hLF was about 1 to 9 (lanes 3 and 4, see
also Ref. 16), whereas the ratios of the minor to the major doublet in genomic and cDNA transgenic hLF were about 1 to 9 and 4 to 6, respectively (lanes 1 and 2). Results with
nonboiled samples suggested that most of genomic and cDNA
transgenic hLF was saturated with iron (lanes 5 and
6). Quantitation of iron in purified transgenic hLF
preparations (28) yielded results identical to calculations of the iron
content based on absorbance measurement (see below). On reduced
SDS-PAGE, the mobility of the two bands in natural hLF and the two
doublets in transgenic hLF decreased to the same extent (lanes
9-12). The ratios of bands and doublets representing iron-saturated hLF in nonboiled samples of natural and transgenic hLF
(lanes 5-8) were identical to the ratios found in
respective hLF samples from which iron had been released upon boiling
(lanes 1-4) or reduction (lanes 9-12). SDS-PAGE
analysis of deglycosylated natural and transgenic hLF showed a single
protein band with the same mobility for each hLF species, indicating
that glycosylation heterogeneity accounts for the presence of the
distinct protein bands in glycosylated natural and transgenic hLF (Fig.
5).
Absorbance spectrum analysis of cDNA transgenic hLF revealed an
absorption peak at 465 nm (Fig. 6), indicating that the
metal ion bound to transgenic hLF is indeed iron (27). Genomic
transgenic hLF showed a similar absorbance spectrum (results not
shown). The A280/A465
ratio of natural hLF was 573.1, whereas those of iron-saturated natural
hLF, cDNA, and genomic transgenic hLF were 24.2, 27.3, and 25.8, respectively. Calculation of the degree of saturation of hLF with iron
indicated that natural hLF was 3% saturated, whereas iron-saturated
natural hLF, cDNA transgenic hLF, and genomic transgenic hLF were
100, 86, and 92% saturated, respectively. The
A410/A465 ratios of
iron-saturated natural hLF, cDNA, and genomic transgenic hLF were
0.80, 0.85, and 0.82, respectively, indicating that preparations were
pure (22).
Iron is released from iron-saturated
hLF under acidic conditions. Iron release from hLF begins at about pH
4.0 and is complete at about pH 2.5 (29). To obtain information on the
iron binding characteristics of transgenic hLF, we compared the
pH-mediated iron release from this protein with that of iron-saturated
natural hLF, monitoring the decrease in A465
upon incubation in buffers over the pH range 6.1-2.1. The results
indicated that the release of iron under acidic conditions from
transgenic hLF is not different from that of iron-saturated natural hLF
(Fig. 7). The pH values at which 50% of hLF molecules
were desaturated (after 24 h) were 3.6, 3.7, and 3.6 for
iron-saturated natural hLF, cDNA, and genomic transgenic hLF,
respectively.
The binding of iron by transgenic and iron-saturated natural hLF that
had been desaturated by dialysis against pH 2.5 was then studied by
titration with iron. The results indicated that both desaturated
transgenic and natural hLF could be completely resaturated with iron
(Fig. 8). Human LF with iron at molar ratios of 0.4 and
0.8 appeared to be about 20 and 40% saturated, respectively (Fig. 8,
lanes 3 and 2, cDNA transgenic hLF;
lanes 7 and 6, natural hLF), whereas saturation
appeared complete at an iron to hLF molar ratio of 2 to 1 (lanes
1 and 5). This indicates that both transgenic and
natural hLF can bind two iron ions. Moreover, the results suggest that
an hLF molecule that has bound one iron ion almost immediately binds a
second one, i.e. cooperativity in iron binding of hLF N- and
C-lobes appears to occur.
Human
LF has been shown to bind to a large variety of molecules, and some of
the interactions may be of physiological relevance. The results in Fig.
9 indicate that binding of transgenic hLF to DNA, LPS,
and heparin is not different from that of natural hLF. The identical
binding properties of the distinct hLF species to the lectins
concanavalin A (30) and Ulex Europaeus agglutinin (affinity for
L-fucose (31)) indicates that transgenic hLF is N-glycosylated and that the extent of fucosylation is
similar as that in natural hLF. Both natural, iron-saturated hLF and
transgenic hLF appeared to bind equally well to chelated iron ions. The
binding was completely abrogated after selective modification of
histidine residues with diethyl pyrophosphate (32), suggesting that
surface-exposed histidine residues in hLF can bind metal ions (data not
shown). The identical binding properties of hLF species to polyclonal rabbit anti-hLF antibody as well as to anti-hLF mAbs E1, E2, and 13.17, which bind to different conformational determinants in hLF,2 suggest that the tertiary structure of transgenic hLF
is similar if not identical to that of natural hLF.
Recombinant expression of hLF enables studies of the relation
between protein structure and function. So far, rhLF has been expressed
in non-human (22) and human cell lines (16), in yeast (20), fungi (33),
tobacco plant cells (34), and in transgenic mice (18). Recombinant
expression also allows large-scale production for studies of potential
applications in human health care. For this reason we have genetically
modified cattle to obtain expression of hLF in the bovine mammary gland
(35). In this paper we demonstrate the feasibility of producing
functional rhLF in the milk of transgenic mice.
The results of double antibody immunoassays (Figs. 1 and 9) and
hyperimmunization of transgenic mice with natural hLF (Fig. 2) indicate
that the murine and rabbit immune systems do not distinguish transgenic
from natural hLF. The tolerance for natural hLF in male and female
transgenic mice suggests that the expression of transgenic hLF is not
completely restricted to lactation and may have occurred in other
tissues during maturation. Transcriptional activity of the
The strongly positively charged N terminus of hLF plays an important
role in its bactericidal activity (37), in its binding to LPS (9) and
glycosaminoglycans (38), as well as in its clearance from the
circulation (39). Mono S chromatography and N-terminal sequencing (Fig.
3) revealed equal cationic properties and intact N-terminal sequences
of natural and transgenic hLF.
Purified transgenic hLF appeared to be about 90% saturated with iron,
whereas natural hLF was saturated for about 3% (Figs. 4 and 6).
Western blots of transgenic mouse milk samples showed transgenic hLF
was already saturated with iron prior to purification.2
Expression of recombinant hLF by cell lines or microorganisms also has
been shown to yield hLF saturated with iron (16, 20, 22, 33) and
Cu2+ (20) probably due to the presence of excess metal ions
in the culture medium (16, 20). We anticipate that transgenic hLF expressed in the bovine mammary gland will show a degree of iron saturation lower than 10% based on analyses of natural hLF that had
been added to cow milk.2
Transgenic hLF did not differ from natural hLF in its ability to
release bound iron with decreased pH (Fig. 7), nor in its ability to
bind two iron atoms concomitantly and maintain its iron-saturated
conformation in the presence of SDS (Fig. 8). Identical iron binding
properties imply that unsaturated transgenic and natural hLF will
behave identically with respect to biological effects mediated through
binding of iron, e.g. bacteriostasis (7), inhibition of
hydroxyl radical formation (14), and transport of iron from the mother
to the newborn (6, 27).
Analysis of untreated and enzymatically deglycosylated transgenic and
natural hLF (Fig. 5) indicated that the different mobility of
transgenic and natural hLF on SDS-PAGE results from differences in
N-glycosylation between murine and human mammary gland
epithelium. Spik et al. (4) have shown that glycosylation of
lactoferrin varies between animal species, its site of production, and
even at a single production site. The relative densities of the two doublets of transgenic hLF were different in cDNA and genomic transgenic hLF (Fig. 4), most likely relating to the presence of a
cysteine in stead of a glycine at position 404 in cDNA transgenic hLF, since this amino acid change results in increased glycosylation at
the optional glycosylation site Asn624 (16, 26).
The role of glycosylation in the biological actions of hLF is not
known, but may serve to protect the molecule against rapid degradation
in the gastrointestinal tract (16, 40). The identical binding
properties of natural and transgenic hLF to concanavalin A and UEA I
suggests that transgenic hLF like natural hLF (4) contains biantennary
glycan chains. Thus, hLF expressed in the mammary gland is likely to be
more similar to natural hLF than hLF produced in yeast or fungi
(33).
Some of the biological activities of hLF arise from its capacity to
bind to other molecules. For example, direct binding of hLF to LPS of
Gram-negative bacteria with disturbance of cell wall stability has been
implicated in its bactericidal action (9, 10). The interaction of hLF
with the lipid A moiety of LPS (9) may account for the reduced
production of cytokines upon challenge with LPS (15) as well as for the
inhibition of neutrophil priming (41). The interaction with DNA of hLF
translocated to the nucleus may alter the transcriptional activity of
certain genes, which may contribute or explain for the immunomodulatory effects of hLF (42). The neutralization of the anticoagulant activity
of heparin by hLF released from neutrophils at an infection site may
serve to localize the infection through encapsulation with fibrin (43).
The affinities for each of these possibly physiologically relevant
ligands did not differ between transgenic hLF, iron-saturated hLF, and
natural hLF (Fig. 9).
In conclusion, the similar properties of unsaturated transgenic hLF and
natural hLF suggest that hLF produced in bovine milk will exert
similar, if not identical, antibacterial and anti-inflammatory activities in vivo.
We thank Marianne Kroos (Erasmus University,
Rotterdam, The Netherlands) for performing the N-terminal sequencing
and Dr. Peggy Neville (Dept. of Physiology and Medicine, University of Colorado School of Medicine, Denver, CO) for critically reading the
manuscript. Dr. B. Schoepen (Lichthaart, Belgium) is acknowledged for
helpful discussion.
Pharming BV,
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
FOOTNOTES
ACKNOWLEDGEMENTS
REFERENCES
S1 casein gene. Recombinant hLF expressed
in the milk of transgenic mice (transgenic hLF) was compared with
natural (human milk-derived) hLF. Immunological identity of the two
forms was shown by double antibody immunoassays and the absence of an
anti-hLF antibody response in transgenic mice on hyperimmunization with
natural hLF. Mono S cation-exchange chromatography and N-terminal
protein sequencing of transgenic and natural hLF revealed identical
cationicity and N-terminal sequences. SDS-polyacrylamide gel
electrophoresis and absorbance measurements of purified transgenic hLF
showed this protein was 90% saturated with iron, whereas natural hLF
is only 3% saturated. The pH-mediated release of iron from transgenic
hLF was not different from that of iron-saturated natural hLF.
Unsaturated transgenic hLF could be completely resaturated upon
addition of iron. Slight differences in mobility between transgenic and
natural hLF on SDS-polyacrylamide gel electrophoresis were abolished by
enzymatic deglycosylation. Binding of transgenic and natural hLF to a
range of ligands, including bacterial lipopolysaccharide, heparin,
single-stranded DNA, Cibacron blue FG 3A, and lectins, was not
different.
Reagents
S1 casein gene and either hLF cDNA (3) or the
genomic hLF sequences were introduced into the murine germline as
described (18). Signal sequences were derived from the bovine
S1 casein gene (in case of cDNA-based vectors) or the hLF gene (genomic vectors). Comparison of the coding sequence of
genomic hLF and published hLF sequences (19-22), revealed that the
amino acid sequence of mature genomic transgenic hLF represents the
most frequently occurring sequence for
hLF.2 The cDNA encoding transgenic hLF
contains two unique differences, Ile130
Thr and
Gly404
Cys (3). A detailed description of the
construction of hLF expression vectors, as well as of the differences
in hLF milk concentrations between distinct categories of hLF
transgenics, is beyond the scope of this paper and will be published
separately.3 Milk from lactating females
was obtained as described (18).
-mercaptoethanol.
1.4 × (A465/0.058))/1.1. The percentage of iron
saturation of purified hLF preparations was calculated as the ratio of
iron-saturated hLF to total hLF. The UV-VIS absorbance spectra of
purified hLF preparations were recorded using an Ultrospec II
spectrophotometer (LKB, Bromma, Sweden)
Immunological Identity of Transgenic and Natural
hLF
Fig. 1.
Dose-response curves of natural hLF and
transgenic hLF in the RIA for hLF. Serial dilutions of natural hLF
(10 µg/ml; A and B, ), of nontransgenic
mouse milk (B, ×), of genomic hLF transgenic mouse milk
(A,
,
, and
), and of cDNA hLF transgenic mouse
milk (B,
,
, and
) were incubated with anti-hLF mAb
13.17 coupled to Sepharose. Bound hLF was detected by subsequent
incubation with polyclonal 125I-anti-hLF antibodies.
Results are expressed as percentage binding of the
125I-anti-hLF antibodies added. The volume of the
experimental sample tested (µl) is indicated on the
abscissa. Human LF concentrations in the genomic hLF
transgenic mouse milk samples were: 3.8 mg/ml (
, line 937), 0.3 mg/ml (
, line 984), 1.3 mg/ml (
, line 984). The hLF
concentrations in the cDNA hLF transgenic mouse milk samples were:
0.8 mg/ml (
, line 649), 1.7 mg/ml (
, line 649), 0.4 mg/ml (
,
line 660).
[View Larger Version of this Image (18K GIF file)]
Fig. 2.
Antibody response in nontransgenic and hLF
transgenic mice after repeated injections with natural hLF.
Nontransgenic mice (n = 6), cDNA hLF
(n = 6), and genomic hLF transgenic mice (n = 5) were repeatedly injected with purified natural
hLF. The antibody responses in sera of individual mice () were
expressed as percentage of the mean response (arbitrarily defined as
100%) in the nontransgenic mice. The horizontal lines
indicate the mean response in each group of mice.
[View Larger Version of this Image (15K GIF file)]
S1 casein signal sequence (cDNA hLF) and the hLF
signal sequence (genomic hLF) had been correctly and completely removed
in the murine mammary gland to yield mature rhLF.
Fig. 3.
Mono S chromatography and N-terminal protein
sequencing of transgenic hLF. Fifty µg of S-Sepharose-purified
cDNA transgenic hLF was applied to a Mono S HR 5/5 (Pharmacia)
column, and bound protein was eluted with a linear salt gradient from
0-1.0 M NaCl as described (16). N-terminal protein
sequencing (16) results are provided by the standard one-letter code
for amino acids.
[View Larger Version of this Image (14K GIF file)]
Fig. 4.
Nonreduced and reduced SDS-PAGE (7.5%)
analysis of purified hLF. Nonreduced SDS-PAGE (lanes
1-8) of purified hLF (5 µg) freshly diluted in nonreducing SDS
sample buffer and applied either directly (lanes 5-8) or
after boiling for 5 min (lanes 1-4). Reduced SDS-PAGE
(lane 9-12) of samples that were boiled for 5 min in
reducing SDS sample buffer. Gen, genomic transgenic hLF;
cDNA, cDNA transgenic hLF; Fe,
iron-saturated natural hLF; Nat, natural hLF. Proteins were
visualized by staining with Coomassie Brilliant Blue. Left-hand
numbers (Mr × 103) indicate
the migration of the protein standards.
[View Larger Version of this Image (46K GIF file)]
Fig. 5.
Reduced SDS-PAGE (10%) of deglycosylated
natural and transgenic hLF. Human LF was deglycosylated with
N-glycosidase F (26). Five µg of untreated (lanes
1-4) and deglycosylated (lanes 5-8) hLF species were
loaded. For explanation of abbreviations, see Fig. 4.
[View Larger Version of this Image (64K GIF file)]
Fig. 6.
Absorbance spectra of iron-saturated natural
hLF and cDNA transgenic hLF between 200 and 700 nm. Trace
1, iron-saturated natural hLF (16.1 mg/ml); trace 2,
cDNA transgenic hLF (10.9 mg/ml); trace 3, natural hLF
(18.6 mg/ml).
[View Larger Version of this Image (19K GIF file)]
Fig. 7.
Desaturation of iron-saturated natural hLF,
cDNA, and genomic transgenic hLF at varying pH. Purified hLF
samples were diluted in buffers of varying pH, and the absorption at
465 nm was measured after incubation for 24 h at 20 °C. Results
are expressed as percentage desaturation by reference to iron-saturated
hLF diluted in 0.15 M NaCl. The experiment was performed on
four occasions with iron-saturated natural hLF (bars
represent the mean and S.D.) and on one occasion with cDNA and
genomic transgenic hLF.
[View Larger Version of this Image (39K GIF file)]
Fig. 8.
Resaturation with iron of desaturated natural
and transgenic hLF. Iron was added to desaturated cDNA
transgenic hLF (lanes 1-4) and desaturated iron-saturated
natural hLF (lanes 5-8) at molar ratios of 2.0, 0.8, and
0.4. Samples were diluted in nonreducing SDS sample buffer and applied
to SDS-PAGE (7.5%; 10 µg of protein/lane).
[View Larger Version of this Image (68K GIF file)]
Fig. 9.
Binding of purified transgenic and natural
hLF to ligands immobilized onto agarose. The binding of distinct
hLF species to a variety of ligands was studied by RIAs as described
under "Materials and Methods." The figure shows the mean and S.D.
of three separate experiments.
[View Larger Version of this Image (65K GIF file)]
S1 casein gene has for example been shown in the thymus
(36). The strongly reduced but detectable immune response in two of the
six hLF cDNA mice (2 and 8% of control) suggests that tolerance
may be partially disrupted upon aggressive immunization of lines
containing less powerful hLF expression vectors. This idea is supported
by hyperimmunization experiments in mouse lines expressing hLF or other
human proteins in milk over a wide concentration range (data not
shown). The immunological identity of transgenic and natural hLF
suggests that the tertiary structure and glycosylation of hLF produced
in the heterologous mammary gland is very similar, if not identical, to
that of natural hLF. Thus, we do not expect that administration of
transgenic hLF to humans will elicit an immune response. Data on
immunological identity between recombinant hLF produced in various
systems (20, 22, 33, 34) to natural hLF have not been reported.
*
The costs of publication of this
article were defrayed in part by the
payment of page charges. The article
must therefore be hereby marked
"advertisement" in
accordance with 18 U.S.C. Section
1734 solely to indicate this fact.
§
To whom correspondence should be addressed: Pharming BV, Niels
Bohrweg 11-13, 2333 CA Leiden, The Netherlands. Tel.: 31-71-5247416; Fax: 31-71-52416507; E-mail: jnuijens{at}pharming.nl.
1
The abbreviations used are: LF, lactoferrin;
hLF, human lactoferrin; natural hLF, hLF from human milk;
iron-saturated natural hLF, natural hLF that has completely been
saturated with iron in vitro; transgenic hLF, recombinant
hLF secreted in milk of transgenic mice; LPS, lipopolysaccharide; mAb,
monoclonal antibody; RIA, radioimmunoassay; MES,
4-morpholineethanesulfonic acid.
2
J. H. Nuijens and P. H. C. van Berkel,
unpublished observation.
3
F. R. Pieper, manuscript in preparation.
©1997 by The American Society for Biochemistry and Molecular Biology, Inc.