Nucleoside analogs (NAs) conjugated with galactosyl terminating
peptides selectively enter hepatocytes via the asialoglycoprotein receptor and, after intracellular release from the carrier, partly exit
from these cells into the bloodstream, resulting in higher concentrations in liver blood than in systemic circulation. Therefore, conjugates of anticancer NAs can be exploited to accomplish a loco-regional noninvasive treatment of liver micrometastases. In the
present experiments we studied whether the enhancement of drug levels
in liver blood achieved when NAs are given in the coupled form depends
on the rate of drug elimination from the bloodstream. Three NAs,
adenine arabinoside (ara-A), 5-fluoro-2'-deoxyuridine (FUdR), and
2',2'-difluorodeoxycytidine, were coupled with lactosaminated human albumin, a galactosyl terminating carrier. In rats that received
an intravenous bolus injection of these conjugates, we compared the
drug concentrations in liver blood to those in the systemic
circulation. We found that enhanced levels of NAs in liver blood were
only achieved by administering the conjugates of the drugs (ara-A and
FUdR), which are rapidly cleared from the bloodstream. Increased drug
levels also were obtained when ara-A and FUdR conjugates were slowly
infused (a way of administration often used for anticancer drugs). The
experiments also showed that galactosyl terminating conjugates of NAs
might have the potential to produce a therapeutic effect only when the
coupled drugs are active at low blood concentrations, since the amounts
of drugs introduced into hepatocytes and released by these cells in the bloodstream cannot be increased when the receptor for the hepatic uptake of galactosyl terminating peptides is saturated.
 |
Introduction |
To
obtain a selective delivery of nucleoside analogs (NAs) to hepatocytes
and reduce their side effects in the treatment of chronic viral
hepatitis, these drugs were conjugated with galactosyl terminating
peptides, which specifically enter hepatocytes through the
asialoglycoprotein receptor (Fiume et al., 1979
; 1997). The validity of
this approach was supported by clinical studies in hepatitis B
virus-infected patients, which demonstrated an increase in the
chemotherapeutic index of adenine arabinoside (ara-A) after conjugation
with lactosaminated human albumin (L-HSA) (Torrani Cerenzia
et al., 1996; Zarski et al., 2001).
It was found that NAs, after the intracellular splitting of the bond
with the carrier, partly exit from hepatic cells into bloodstream
(Fiume et al., 1998; Di Stefano et al., 1999). Although this release
reduces the efficacy of hepatocyte targeting, it has a useful
consequence, since it can result in higher NA concentrations in hepatic
blood than in systemic circulation (Di Stefano et al., 2000).
Therefore, conjugation of anticancer NAs with galactosyl terminating
carriers might be a way to expose neoplastic cells fed by liver
sinusoids to enhanced drug levels and accomplish a loco-regional
noninvasive treatment of hepatic micrometastases (Di Stefano et al.,
2000, 2002).
In the regional chemotherapy performed by local infusion, the advantage
of drug delivery to the target region depends on the rate of drug
elimination from the rest of the body (Ensminger and Gyves, 1983). In
the present experiments we studied whether the local drug exposure
advantage achieved in liver when NAs are given as galactosyl
terminating conjugates is similarly affected by the rate of elimination
from bloodstream of the drugs released by liver cells. Moreover, we
investigated the effect on this enhancement of the means of
administration of the conjugates (by bolus injection or slow infusion).
We conjugated FUdR, ara-A, and dFdC with L-HSA. ara-A
and FUdR are rapidly cleared from blood (Ensminger et al., 1978;
Preiksaitis et al., 1981; Di Stefano et al., 2000), whereas dFdC slowly
disappears from bloodstream, at least in rats (Shipley et al., 1992).
In rats injected with ara-A, FUdR, and dFdC, administered in the free
or conjugated form, either by bolus injection or by slow infusion, we
determined the concentrations of these NAs in hepatic veins and in
inferior vena cava. These concentrations are a measure of the drug
levels in liver blood and in systemic circulation, respectively.
| |
Materials and Methods |
Conjugate Preparation.
Lactosaminated albumin (L-HSA) was
prepared as described by Wilson (1978). The lactose/HSA molar ratio was
24. Conjugation of ara-A, [3H]FUdR, and
[3H]dFdC was obtained via the imidazolides of
their 5'-phosphoric esters (Fiume et al., 1993). Using this procedure,
NAs are linked to lysine residues of L-HSA by a phosphate
bridge. The drug/L-HSA molar ratio, determined spectrophotometrically,
ranged from 13 to 16. [3H]FUdR and
[3H]dFdC were obtained from Moravek
Biochemicals (Brea, CA). Before conjugation, they were diluted with the
unlabeled compound to obtain a specific activity of 5.5 × 104 dpm/µg. A conjugate
L-[14C]HSA-FUdR, radioactive in the
carrier moiety, was obtained by labeling HSA with
[14C]formaldehyde (56 mCi/mmol) (PerkinElmer
Life Sciences, Boston, MA), according to the method of Jentoft and
Dearborn (1983). The specific activity was 1.7 × 103 dpm/µg, which corresponded to 1 [14C]formaldehyde per HSA molecule.
Animals.
Male Wistar rats weighing 200 to 220 g were
used. They were obtained from Harlan Italy (Udine, Italy) and were
maintained in an animal facility at the Department of Experimental
Pathology, University of Bologna, receiving humane care in accordance
with European Legislation. The protocols of the experiments were
approved by the ethical committee of the University of Bologna. Animals were fed a standard pellet diet ad libitum.
Determination of Drugs in Inferior Vena Cava and in Hepatic
Veins.
Free and conjugated drugs were administered intravenously
either by bolus injection or by slow infusion. Bolus injection was performed via the dorsal vein of the penis under ether anesthesia. Infusion was performed via the femoral vein in animals anesthetized by
intraperitoneal administration of ketamine/xylazine. The compounds were
given in a volume of 0.87 ml, administered in 1 h using an Orion
Sage M361 (Expotech, Houston, TX) infusion pump.
Blood sampling from inferior vena cava and from hepatic veins was
performed at different times after the bolus injection or at the end of
the 1-h infusion period. The procedure described by Di Stefano et al.
(2000) was followed. ara-A was measured according to the method of
McCann et al. (1985); [3H]FUdR and
[3H]dFdC were determined using the procedure of
isotopic dilution applied to HPLC. To identify the elution time of
[3H]FUdR and [3H]dFdC
in the HPLC chromatograms and to evaluate the recovery, 15 µg of the
unlabeled drug were added to each plasma sample (400 µl), kept cold
in ice. In the plasma samples containing
[3H]FUdR, proteins were removed by the addition
of 30 µl of trichloroacetic acid (80%) and centrifugation at
2-4°C. After diethyl ether extraction to eliminate trichloroacetic
acid, 200 µl of supernatant were chromatographed on a Spherisorb ODS2
(Waters, Milford, MA), equilibrated, and eluted according to the method
of McCann et al. (1985). In the plasma samples containing
[3H]dFdC, proteins were removed by addition of
1 ml of ethanol and centrifugation. Supernatant was dried under vacuum,
dissolved in 300 µl of H2O, and analyzed by
HPLC as described for [3H]FUdR. Radioactivity
eluting at the position of the unlabeled markers was counted, and the
plasma concentration of [3H]FUdR or
[3H]dFdC was calculated, taking into account
the recovery of the marker (measured by UV absorbance) and the specific
activity of the injected drug. The recovery of the markers was
85 to 95%. The specific activity of free or conjugated drugs was
5.5 × 104 dpm/µg. When radioactivity of
the drug peak was lower than 200 dpm, the plasma concentration of the
drug was considered to be below a measurable level. With a 90%
recovery of the markers, the lowest measurable plasma concentration was
about 20 ng/ml.
In rats receiving a bolus injection of [3H]FU
(Amersham Biosciences AB, Uppsala, Sweden) (see below), the drug was
extracted from plasma as described for [3H]dFdC
and measured by the procedure of isotopic dilution; HPLC analysis was
performed using two Spherisorb ODS2 columns connected in series and
isocratically eluted with 20 mM sodium tetraborate, pH 7.5. The limit
of detection was 20 ng/ml.
| |
Results |
Figure 1 shows that after
administration of ara-[3H]A or
[3H]dFdC the levels of radioactivity were
similar in liver, spleen and intestine. The finding of higher amounts
of radioactivity in kidney of
[3H]dFdC-treated rats is in agreement with
previous data (Shipley et al., 1992). In animals injected with
[3H]FUdR or [3H]FU [FU
is a precursor of FUdR widely used in the chemotherapy of colorectal
cancer (Heriot and Kumar, 1998)], the levels of radioactivity in liver
were significantly higher than those in spleen, intestine, and kidney.
The radioactivity in kidney was higher than in spleen and intestine.
View larger version (57K):
[in this window]
[in a new window]
|
Fig. 1.
Radioactivity (dpm/specific activity/gram) in organs
of rats 5 min after the i.v. injection of the 3H-labeled
drugs. Experiments were performed as described by Fiume et al. (1987).
Values are means from four animals; bars indicate standard errors. The
data were evaluated by Student's t test. In rats
treated with [3H]dFdC, the difference between the values
measured in kidney and those in the other organs was statistically
significant (+++, p < 0.001). In animals injected
with [3H]FUdR or [3H]FU, the values in
liver were higher than those in spleen, intestine (***,
p < 0.001) and kidney ( ,
p < 0.01). In the same animals the levels of
radioactivity in kidney were higher than those in spleen and intestine
(+++, p < 0.001).
|
|
The rapid disappearance of ara-A (Preiksaitis et al., 1981; Di Stefano
et al., 2000), [3H]FU, and
[3H]FUdR (Ensminger et al., 1978) from
bloodstream is confirmed by the data of Tables
1 and 2. Given to rats by bolus injection at the dose of 0.5 µg/g, these drugs
were not detected in blood of inferior
vena cava after 15 to 30 min (the limits of detection were 2 ng/ml for
ara-A and 20 ng/ml for both [3H]FU and
[3H]FUdR). On the contrary, after
administration of [3H]dFdC (0.5 µg/g), the
drug was still detectable at 8 h (limit of detection = 20 ng/ml) (Table 3).
View this table:
[in this window]
[in a new window]
|
TABLE 1
ara-A levels in inferior vena cava (IVC) and in hepatic veins (HV) of
rats intravenously injected with the free or L-HSA-conjugated drug
Data were obtained from three rats and are given as mean ± S.E.
When the difference between the drug level in IVC and HV is
statistically significant (Student's t test), it is
indicated by asterisks (***p < 0.001).
|
|
View this table:
[in this window]
[in a new window]
|
TABLE 2
[3H]FU and [3H]FUdR levels in inferior vena cava
(IVC) and in hepatic veins (HV) of rats intravenously injected with the
free or L-HSA-conjugated drugs
Data were obtained from three rats and are given as mean ± S.E.
When the difference between the drug level in IVC and HV is
statistically significant (Student's t test), it is
indicated by asterisks (***p < 0.001).
|
|
View this table:
[in this window]
[in a new window]
|
TABLE 3
[3H]dFdC levels in inferior vena cava (IVC) and in hepatic
veins (HV) of rats intravenously injected with the free or
L-HSA-conjugated drug
Data were obtained from three rats and are given as mean ± S.E.
When the difference between the drug level in IVC and HV is
statistically significant (Student's t test), it is
indicated by asterisks (**p < 0.01;
***p < 0.001).
|
|
After administration of [3H]FU and of
[3H]FUdR given by bolus injection or by 1 h infusion ([3H]FUdR), the blood levels of
drugs were several times lower in hepatic veins than in inferior vena
cava (Table 2), a result in agreement with data in humans and explained
by the ability of hepatic cells to extract fluoropyrimidines from
bloodstream (Ensminger et al., 1978; Wagner et al., 1986). In rats
administered [3H]dFdC (bolus injection or
infusion) or ara-A (infusion) (two NAs which are not selectively
extracted by liver cells) (Fig. 1), the drug levels were similar in
hepatic veins and in inferior vena cava (Tables 1 and 3).
Tables 1 and 2 show that administration of ara-A and
[3H]FUdR conjugates produces an increase in
drug concentrations in liver blood. In fact, in rats injected with
L-HSA-coupled [3H]FUdR or ara-A
(0.5 µg/g, corresponding to 13 µg/g conjugates), the ratios between
the drug levels in hepatic veins and those in inferior vena cava were
several times higher than in animals administered with the unconjugated drugs.
The enhancement of drug levels in liver blood was achieved not only in
animals which received a bolus injection of the conjugate, but also in
rats slowly infused.
In rats injected with L-HSA coupled
[3H]dFdC the increase in drug concentrations in
liver blood was much lower; it was observed only at the earliest times
after the bolus injection and in the animals infused it was
statistically significant only for the dose of 0.25 µg/g (Table 3).
After 1 h infusion of the conjugates, the blood levels of drugs
were similar in animals administered 13 µg/g/h or 26 µg/g/h conjugate (Tables 1-3). This finding can be explained by assuming that
the asialoglycoprotein receptor was saturated with the lower dose, so
that by doubling the dosage the amounts of conjugate taken up by the
hepatic cells could not be increased. This explanation is supported by
the finding that in rats injected with asialofetuin, the maximal
clearing capacity of the liver is equivalent to approximately 12 µg
of protein/g/h (Regoeczi et al., 1978). Further support comes from the
results reported in Fig. 2: after 1 h of infusion, the values of radioactivity in liver of rats
administered 13 µg/g/h L-[14C]HSA-FUdR (a conjugate
labeled in the carrier moiety) were similar to those of the animals
infused with the double dose; on the contrary, the levels of
radioactivity in plasma were significantly higher in the latter
animals.
View larger version (39K):
[in this window]
[in a new window]
|
Fig. 2.
Acid-insoluble radioactivity in liver and in plasma
of rats after 1 h of infusion of different doses of
L-[14C]HSA-FUdR (6.5 µg of conjugate
contained 0.25 µg of FUdR). Experiments were performed as described
by Fiume et al. (1987). Acid-insoluble radioactivity was expressed as
disintegrations per minute/specific activity/gram or milliliter
(dpm/SA/ml or g). Values are means from three animals; bars indicate
standard errors. The data were evaluated by Student's t
test. ***, significant difference (p < 0.001) only from rats infused with 6.5 µg/g/h conjugate. ,
significant difference (p < 0.001) from rats
infused with 6.5 or 13 µg/g/h conjugate.
|
|
| |
Discussion |
Our previous studies showed that administration by peripheral
venous route of NAs coupled to galactosyl terminating carriers can
result in higher drug levels in liver blood than in systemic circulation (Di Stefano et al., 2000). This observation suggested that
conjugation with these carriers might be a way to expose the neoplastic
cells of liver micrometastases, which are nourished by hepatic
sinusoids (Haugeberg et al., 1988), to enhanced concentrations of
anticancer NAs. By releasing the drug in liver blood, the conjugates would accomplish a regional chemotherapy similar to that performed by
the portal vein infusion of FU (Midgley and Kerr, 1999). It would have
the advantage of administration by the peripheral venous route, which
allows much longer treatment than the single 7-day cycle permitted by
portal vein infusion.
In the present study we have conjugated L-HSA with ara-A,
FUdR, and dFdC and have shown that the enhancement of drug levels in
liver blood can only be achieved by using NAs (ara-A, FUdR), which are
rapidly cleared from the bloodstream; this result is in agreement with
those of regional chemotherapy performed by a local infusion of drugs
(Ensminger and Gyves, 1983). We have also observed that conjugates of
rapidly cleared NAs leads to higher drug concentrations locally in
liver blood, not only when they are given by bolus injection, but also
when they are administered by slow infusion. This finding is relevant
since anticancer NAs usually act on DNA synthesis, and administration
by slow infusion can be useful to ensure that an increasing proportion
of tumor cells are exposed to the antineoplastic effect at a
metabolically vulnerable time.
In comparison with the local infusion of drugs, the use of
L-HSA conjugates for a regional chemotherapy in liver would
have a limitation, since it could be advantageous only when the coupled drugs are active at low concentrations. In fact, the quantities of
drugs transported by L-HSA inside the hepatocytes and
released from these cells in liver sinusoids could not be enhanced by
increasing the dose of the conjugate when the asialoglycoprotein
receptor is saturated. According to the present results, FUdR appears
to be a suitable drug for the liver regional chemotherapy approach achieved by NA coupling to L-HSA. FUdR is rapidly cleared
from the bloodstream, and in the treatment of human colorectal cancer, it is active at very low plasma concentrations (0.2 ng/ml) (Park et
al., 1988), which are several times smaller than those measured in the
systemic circulation of rats infused with L-HSA-FUdR.
This work was supported by Associazione Italiana per la Ricerca
sul Cancro (AIRC), Ministero dell'Università e della Ricerca Scientifica e Tecnologica (MURST), and University of Bologna (funds for
selected research topics).
NA, nucleoside analog;
ara-A, adenine
arabinoside;
dFdC, 2',2'-difluorodeoxycytidine;
FU, 5-fluorouracil;
FUdR, 5-fluoro-2'-deoxyuridine;
L-HSA, lactosaminated human albumin;
HSA, human serum albumin;
HPLC, high-pressure liquid
chromatography.
![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
Top
Abstract
Introduction
-D-arabinofuranosyladenine 5'-monophosphate (ara-AMP) with lactosaminated albumin in parenchymal and sinusoidal cells of rat liver.
Cancer Drug Delivery
4:
11-16[Medline].
