Download Analytical Subcellular Fractionation of Jejunal

Clinical Science and Molecular Medicine (1976) 51, 557-574.
Analytical subcellular fractionation of jejunal biopsy specimens:
methodology and characterization of the organelles in normal tissue
T. J. PETERS
Department of Medicine, Royal Postgraduate Medical School, London
(Receiued 14 May 1976)
Summary
(enterocytes) are known to have a complex structure.
Enterocytes
are implicated in the digestion, absorp1. Portions of closed jejunal biopsies were gently
tion and metabolism of dietary components and their
homogenized in isotonic sucrose or sorbitol and
properties and functions are disturbed to various
centrifuged at 800 g for 10 min to prepare a cell
degrees in most small-intestinal diseases. Afthough
extract.
there are considerable data on the properties of cer2. The extract was fractionated in a single-step
tain
organelles of the enterocyte of animal intestine,
procedure by isopycnic centrifugation on a conlittle
is known quantitatively about these organellesin
tinuous sucrose or sorbitol density gradient with the
human
intestine. Some information can be derived
Beaufay automatic zonal rotor.
from morphological and histochemical studies but
3. The subcellular organelles were located in the
analytical subcellular fractionation techniques perdensity gradient by assay of marker enzymes and
mit quantitative assessment of the properties of the
previously unassigned enzymes were localized to
individual organelles in health and disease. The
particular organelles.
application of fractionation procedures to human
4. The following organelles were characterized by
tissue has hitherto been limited by the technical
their modal equilibrium densities in sucrose density
difficulty
of processing the milligram quantities of
gradients: brush borders (1~21)~
peroxisomes (1.18),
mitochondria (1.16), endoplasmic reticulum (1016)~ tissue obtained by closed jejunal biopsy. The
development of a suitable single-step fractionation
basal-lateral membranes (1.12). At least three distechnique coupled with highly sensitive markertinct populations of lysosomes with different modal
enzyme assay procedures has permitted analytical
densities and enzyme content were demonstrated.
fractionation techniques to be applied to human
5. This analytical fraction technique can be used
biopsies for the first time.
to study the subcellular pathology of human jejunum.
Key words: brush borders, closed biopsy, cytosol,
disaccharidases, endoplasmic reticulum, lysosomes,
microsomes, mitochondria, plasma membrane,
subcellular fractionation, jejunum.
Methods
Patients
Jejunal biopsies were obtained with either a Crosby
capsule or the Debri multiple biopsy capsule. The
biopsies were collected from adult subjects of both
sexes undergoing investigation for possible gastrointestinal disease. They were shown under dissecting-microscope and histological examination to
be completely normal. The disaccharidase activities
of the tissue were within normal limits (Peters, Batt,
Introduction
The absorptive epithelial cells of the small intestine
Correspondence: Dr T. J. Peters, Department of Medicine,
Royal Postgraduate Medical School, Du Cane Road,
London W12 OHS.
557
558
T . J. Peters
Heath & Tilleray, 1976). The studies reported in this
paper have been approved by the local ethical
committee.
Tissue homogenization
A portion of the biopsy, approximately 10 mg wet
weight, was fixed in formalin-saline and processed
for routine histological examination; another,
approximately 10 mg wet weight, was collected in
2 ml of ice-cold sucrose solution (0.3 mol/l) containing disodium EDTA (1 mmol/l), pH 7.4, and
ethanol (22 mmol/l) (SVE medium). The tissue was
disrupted with ten strokes of a loose-fitting (type A)
pestle in a small Dounce homogenizer (Kontes Glass
Co., Vineland, N.J., U.S.A.) and centrifuged at 800g
for 10 min in an MSE 4L centrifuge (Measuring and
Scientific Equipment, Crawley, Sussex, U.K.). The
pellet was resuspended in a further 2 ml of SVE
medium with three strokes of the type A pestle and
centrifuged again. The supernatants were combined
(PNS fraction). The low-speed pellet, consisting of
nuclei, large brush-border fragments and interstitial
cells (N fraction), was resuspended in 2 ml of SVE
medium with a tight-fitting (type B) pestle.
homogenization and the density gradient media.
Brush borders were isolated from jejunal mucosa
by homogenization in a small Dounce homogenizer
in 3 ml of TrisIEDTA ( 5 mmol/l),pH7.4(Millington7
Critchley & Tovell, 1966) with five strokes of a loosefitting (type A) pestle. The brush borders were collected by centrifugation at 8000 g and were twice
resuspended and centrifuged in EDTA/Tris buffer.
The final pellet was suspended in 10 ml of Tris/HCI
buffer (1 mol/l), pH 7.4, and stirred gently for 1 h at
4°C. A portion (3-5 ml) of the Tris-disrupted brushborder suspension was then subjected to analytical
isopycnic centrifugation on a sorbitol density
gradient as described above.
Enzymic analyses
Table 1 shows the assay conditions used for
analysis of the density gradient fractions (Peters,
Heath, Wansbrough-Jones & Doe, 1975). Protein
was assayed fluorimetrically as described previously
(Peters, Miiller & de Duve, 1972). Maltase, sucrase
and lactase were assayed by the fluorimetric modification (Peters et al., 1976) of the technique of
Dahlqvist (1964).
Density gradient centrifugation
Inhibitor studies
Approximately 3.5 ml of PNS fraction was layered
on to 30 ml of density gradient extending linearly
with respect to volume, from a density of 1-05to one
of 1.28 and resting on a 4 ml cushion of density 1-32,
in the Beaufay (1966) automatic zonal rotor. All
solutions contained disodium EDTA (1 mmol/l),
pH 7-4, and ethanol (22 mmol/l). The rotor was
accelerated to 35 000 rev./min and run for 35 min
with an integrated force of 3.3 x 10'O rad2 s-'. The
rotor was then slowed to 8000 rev./min for automatic unloading and collection of the gradient
fractions (Leighton, Poole, Beaufay, Baudhuin,
Coffey, Fowler & de Duve, 1968). Some sixteen
fractions were collected into tared tubes, thoroughly
mixed, weighed, and then density was determined
with an Abbe refractometer (de Duve, Berthet &
Beaufay, 1959). Certain enzymes, including alkaline
phosphatase, the dehydrogenases, D-amino acid
oxidase and neutral 8-galactosidases, were assayed
on the same day as the fractionation: the samples
were then stored at -20°C. Sucrose was always used
as the density gradient medium but purified sorbitol
(Messer & Dahlqvist, 1966) was also used both for
For certain enzymes, analysis of the gradient fractions was carried out with selective inhibitors.
Parallel incubations were carried out under identical
conditions with and without the inhibitor. 5'-Nucleotidase was assayed in the presence of nickel chloride
(5 mmol/l), a selective inhibitor of nucleotidase
activity (Ahmed & Reiss, 1958; Campbell, 1962) or
in the presence of glycine buffer (0.1 mol/l), a selective inhibitor of alkaline phosphatase (Bodansky,
1946). a-Glucosidase was assayed in the presence of
D-turanose (32 mmol/I), an inhibitor of lysosomal
glucosidase activity (Lejeune, Thinks-Sempoux &
Hers, 1963) or with zinc chloride (2.8 mmol/l), an
inhibitor of neutral microsomal activity (Suzuki &
Kushida, 1973). 8-Galactosidase was assayed with
and without p-chloromercuribenzoate (0.18 mmol/l),
an inhibitor of non-brush-border-enzyme activity
(Asp & Dahlvqist, 1972).
Presentation of enzyme distribution results
Results are expressed as frequency-density histograms. The results in the lightest fractions, up to
Oxaloacetate, NADH
Pyruvate, NADH
a-Ketoglutarate, NADH
(NH4)804
['4C]Tryptamine
Hydrogen peroxide
Succinate,
2-(p-iodophenyl)-3-(p-nitrophenyl)-5phenyltetrazolium chloride
D-Proline
4-Methylumbelliferylphosphate
4-Methylumbelliferyl-aD-glucopyranoside
4-Methylumbelliferyl-aD-glucopyranoside
Leucyl-2-naphthylamide
Malate dehydrogenase
Lactate dehydrogenase
Glutamate dehydrogenase
Succinate dehydrogenase
Alkaline phosphatase
a-Glucosidase
Leucyl-8-naphthylamidase
a-Glucosidase
D-Amino acid oxidase
Monoamine oxidase
Catalase
Substrate
Enzyme
7.1
8.0
60
9-2
8.0
7.4
7.4
7.4
7.4
7.4
7.4
PH
0.1 mol/l sodium phosphate
0.1 mol/l sodium phosphate'')
0 1 mol/l sodium borate'')
10 mmol/l MgCI'
0.1 mol/l sodium acetate'')
0.3 moll1 Tris/HCI buffer
Peters et al. (1975)
Peters et al. (1975)
Peters et al. (1975)
Guilbault, Brignac & Zimmer
(1968)
Peters et al. (1975)
0.1 mol/l sodium phosphate buffer")
Lowry, Roberts &
1 mmol/l dithiothreitol, 1 g/l bovine serum albumin Kapphahn (1957)
0.1 mol/l sodium phosphate buffer'')
Lowry et al. (1957)
1 mmol/l dithiothreitol, 1 g/i bovine serum albumin
0.05 mol/l triethanolamine/HCl buffer")
Ellis & Goldberg (1972)
1 mmol/l ADP
Lowry et al. (1957)
0-26 mol/l sodium phosphate buffer
Wurtman & Axelrod (1963)
20 mmol/l imidazole/HCI buffer
Peters et al. (1972)
1 g/l bovine serum albumin
0.3 mol/l sodium phosphate buffer
Pennington (1961)
Incubation medium
TABLE
1. Enzyme-assay conditions
8-Glucosidase
5'-Nucleotidasc
8-Glucuronidase
Acid diesterase
Cathepsin C
N-Acetyl-8-galactosidase
N-Acetyl-/3-glucosaminidase
4-MethyIumbelliferyl-~D-glucopyranoside
60 mmol/l piperazine/HCl buffer
24 mmol/l MgCll
12 mmol/l2-glycerophosphate
0.1 mol/l sodium phosphate
9.0
5.9
0.1 mol/l sodium acetate(')
1 mmol/I dithiothreitol
10 mmol/l NaCl
0.1 mol/l sodium acetate(')
3.8
4.7
Peters er 01. (1975)
0.1 mol/l sodium acetate(')
Peters er al. (1975)
Douglas er al. (1972)
Peters er al. (1975)
Peters el 01. (1972)
Vanha-Perttula, Hopsu,
Sonninen & Glenner (1965)
Peters et al. (1975)
Peters er al. (1975)
0.1 mol/l sodium acetate(')
0.1 mol/l sodium acetate(2)
Peters er al. (1975)
Peters er al. (1975)
Szasz (1969)
Reference
~~
-~
10 mmol glycine
0.1 mol/l sodium acetate")
10 mmol/l NaCl
0.1 mol/l sodium phosphate"'
0.1 mol/l ammediol/HCl buffer")
Incubation medium
Buffers contain 0.01% (v/v) Triton XIOO.
Buffers contain 0 1 % (v/v) Triton XIOO.
4-Methylumbelliferyl
pyrophosphate diester
4-Methylumbelliferyl-8D-glucuronide trihydrate
[3H]Adenosine 5-monophosphate
5.0
5.8
5.8
4.0
3.5
4-MethylumbelIiferyl-8D-galactopyranoside
4-Methylumbelliferyl
phosphate
4-Methylumbelliferyl-2-deoxy-2acetamido-D-glucopyranoside
4- MethylumbelliferyE2-deoxy-2acetarnido-D-galactopyranoside
Glycyl-L-phenylalanyl2-naphthylamide
8-Galactosidase
Acid phosphatase
8.5
PH
y-Glutamyl-2-naphthylamide
Substrate
y-Glutamyl transpeptidase
Enzyme
TABLE
1 (continued)
2,
2
D
3
5
Subcellularfractionation of human jejunum
561
TABLE
2. Specific activities and percentages of released enyzmes in homogenized jejunal biopsies
E
of activitylmg of protein). Number of
Specific activity is expressed as mean value ~ S (munits
specimens assayed is shown in parentheses. Protein is expressed as mg+ SE. p-CMB, p-chloromercuribenzoate.
Enzyme
EC no.
Specific
activity
Activity in
PNS fraction
(%)
Alkaline phosphatase
Leucyl 8-naphthylamidase
y-Glutamyl transpeptidase
Total a-glucosidase (pH 6.0)
Zn2+-resistant a-glucosidase (pH 6.0)
Total 8-galactosidase (pH 6.0)
p-CMB-resistant 8-galactosidase (pH 6.0)
Catalase
3.1.3.1
3.4.1 1 . I
2.3.2.2
3.2.1.20
3.2.1.20
3.2.1.23
3.2.1.23
1.11.1.6
N- Acetyl-8-glucosaminidase
3.2.1.30
3.2.1.53
3.1.3.2
3.4.14.1
3.2.1.23
3.6.1 . I 1
3.2.1.31
3.2.1.21
1.1.1.37
1.4.1.2
1.4.3.4
3720f 305
7.08f0.64
41.8k6.7
81-7f 1.40
64.3+ 3.7
75.6f 3.0
1.1.1.27
3.2.1.20
3.2.1.20
387240
(5)
1.13+0.10 (14)
1.60f0.3
(3)
41.1f4.1
(8)
1.05+0.14 (25)
N-Acetyl-8-galactosidaminidase
Acid phosphatase
Cathepsin C
8-Galactosidase (pH 3.5)
Acid diesterase
8-Glucuronidase
/l-Glucosidase
Malate dehydrogenase
Glutamate dehydrogenase
Monoamine oxidase
Lactate dehydrogenase
a-Glucosidase (pH 8.0)
Zn2+-sensitive a-glucosidase (pH 6.0)
5’-Nucleotidase
Total protein
a density of 1.10, which relate mostly to soluble
activity remaining in the starting layer, were pooled
and plotted over the density interval 1.05-1.10. The
corresponding block is shaded in the histogram to
indicate that it is not part of the density distribution
proper. All calculations and plots were performed by
computer (Beaufay, Jacques, Baudhuin, Sellinger,
Berthet & de Duve, 1964; Leighton et af., 1968). All
histograms were normalized and the percentage
recoveries (sum of gradient fractions against tissue
extract injected into rotor) are given. Pooling and
averaging of several distributions were performed by
computer (Leighton et al., 1968). The method requires conversion of all the histograms into the same
pre-set density intervals with some resultant loss of
resolution.
3.1.3.5
-
64.7k6.3
4.252 0.78
12.1k1.4
9.542 1.0
7.872 1.0
1.11+0.18
0.3 13& 0.056
(IS)
(9)
(5)
36.42 2.2
53.7f 5.1
39.6f 4.0
39%+ 2. I
35.12 2.9
75.8f 3.5
32.42 6.4
79.9+5.5
(15)
82*6* 2.0
6.08 f 0.47 (14)
2.13f0.23
(4)
39.1f3.8
(7)
2.68f0.5
(5)
1.79f0.93
(4)
1.15f0.19
(5)
3.05f0.20
(7)
1.46f0.19
(3)
55.7f 3.1
62.0f 8.1
61.1f6.4
73.4f 3.9
71.5f4.8
64.Ok 7.0
66,9f 6.0
73.6f 5-4
(11)
(13)
(3)
(5)
(15)
(4)
(7)
73.8 2 4.2
55.7+ 3.1
-
50.1 If:2.4
75f2
Results
Tissue enzynzes and released actiuity
The specific activities of the various enzymes
estimated in the homogenates of the jejunal biopsies
(Table 2) form the control values against which the
results of measurements in diseased tissue will
subsequently be compared. Sufficiently sensitive
assays have been established so that the activities of
these enzymes and of protein can be determined in
the density gradient fractions. The percentageactivity
in the PNS fraction is also listed (Table 2), reflecting
the proportion of cells disrupted and the amount of
enzyme activity not sedimented at 800 g for 10 min.
Enzymes predominantly localized to the brush
border (e.g. alkaline phosphatase) are found in the
T. J . Peters
562
Alkaline ptosphatase
Dfnsi!!
FIG. 1. Isopycnic centrifugation of 8000 g-min supernatant from jejunal biopsy homogenate. Graphs show
frequency-density histograms for marker enzymes. Frequency (mean+ SD) is defined as the fraction of total
recovered activity present in the subcellular fraction divided by the density span covered. The cross-hatched area
represents, over an arbitrary abscissa interval, the enzyme remaining i n the sample layer. Th e percentages (+ SD)
of recovered activity, with numbers of experiments in parentheses, are: alkaline phosphatase, 81 18 (15);
glutamate dehydrogenase, 84+ 14 ( 5 ) ; 5’-nucleotidase, 98+ 10 ( 5 ) ; N-acetyl-Sglucosaminidase, 84+ 17 (14);
catalase, 101k I 1 (7); lactate dehydrogenase, 101+21 (4); a-glucosidase @H 8.0), 109223 (16); protein, 8 8 k IS
(17).
+
supernatant only to an extent of 35%. Similar values
are found for y-glutamyl transpeptidase, Zn2 resistant a-glucosidase and p-chloromercuribenzoateresistant 8-galactosidase. Leucyl-8-naphthylamidase
has a higher released activity than the other brushborder enzymes. Other enzymes are released into the
extract to between 50 and 75 %. Enzymes having a
major cytosol component, e.g. catalase, malate dehydrogenase and lactate dehydrogenase, are released
+
to a greater extent than the predominantly organellebound enzymes, e.g. acid hydrolases. Allowing for
the fact that certain enzymes, e.g. lactate dehydrcgenase (de Duve & Berthet, 1954) and N-acetyl-8glucosaminidase (Baccino, Rita & Zuretti, 1971),
tend to become adsorbed to other organelles,
approximately three-quarters of the cells in the
biopsies have been disrupted. As mentioned previously (Peters et al., 1975) the gentle homogeniza-
563
Subcellular fractionation of human jejunum
tion procedures employed tend to leave the nonepithelial cells of the biopsy intact, and these are
therefore sedimented in the low-speed pellet, and so
not fractionated on the sucrose gradients.
Density gradient experiments
Organelle distributions: averaged data. The distribution of the principal marker enzymes in the
sucrose density gradients include the averaged data
from several experiments (Fig. 1 and Fig. 2). Alkaline
phosphatase shows a peak with a modal density of
1.21, with skewing of activity to lower densities
(Fig. 1). 5'-Nucleotidase shows a peak with a modal
15ra-
density of 1.12 and a shoulder at density of 1.21, the
latter corresponding to that of the alkaline phosphatase. The distribution of 5'-nucleotidasein the sucrose
gradients was also determined with selective inhibitors of alkaline phosphatase and of 5'-nucleotidase,
with no significant difference in the enzyme distribution. In particular, the relative amounts of activity
at modal densities 1-11 and 1.21 were identical
although the absolute enzyme activities were, of
course, reduced by the inhibitors.
Catalase shows a major soluble component with
a particulate component at a modal density of 1.18.
D-Amino acid oxidase assayed in two experiments
was found to be associated with particulate catalase.
Acid p-qalactasidase
Glucosidase
T
0
h
I
Leucyl-p-naphthybmidase
Acid phosphatase
1
y-Glutamyl tmnspeptidase
Acid diesterase
L
1
r
M a y e dehydmqenase
I0
5
0
1.05
1-10
145
1.20
125
Densily
FIG.2. Isopycnic centrifugation of 8000 g-min supernatant from jejunal biopsy homogenate. Details are as given
for Fig. 1. The percentages ( ~ s D )of recovered activity. with numbers of experiments in parentheses, are: aglucosidase (PH6.0), 87+ 16 (7); acid j'-galactosidase, 90+ 13 (4); leucyl-&naphthylamidase, 119+ 16 (7); acid
phosphatase, El+ 19 (7); y-glutamyl transpeptidase. 98+ 15 (10); acid diesterase, 81+ 12 (5); malate dehydrogenase, 101k 19 (15); monoamine oxidase, 87k 1 1 (7).
D
T.J . Peters
564
or-Glucosidase assayed at pH 8.0 shows a peak with
a modal density of 1.16. Glutamate dehydrogenase
has a small soluble component with a sharp peak in
the density region of 1.16. A similar distribution (not
shown) was found for succinic dehydrogenase.
Monoamine oxidase is almost completely particulate
with a modal density also at 1.16. N-Acetyl-8glucosaminidase shows a soluble component with
a symmetrical peak at a modal density of 1.21 but
with an estimated median density slightly less than
that of alkaline phosphatase. Lactate dehydrogenase
is mainly localized in the soluble fractions but significant activity is found in the gradient, with
a shoulder at density 1.16, corresponding to that of
or-glucosidase (pH 8.0). Protein is also mainly re-
covered in the soluble components with a broad peak
in the region 1.15-1.18 corresponding principally to
the distributions of glutamate dehydrogenase, malate
dehydrogenase and monoamine oxidase.
a-Glucosidase assayed at pH 6.0 has a small
soluble component with the particle-bound activity
showing a bimodal peak at densities of 1.21 and 1.16,
corresponding to alkaline phosphatase and neutral
a-glucosidase (pH 8.0) respectively (Fig. 2). Leucyl/3-naphthylamidase and y-glutamyl transpeptidase
show similar distributions with a bimodal distribua-Glucosidase IpH 6.0)
1
p
Leucyi-B- nophthylomdose
a-Glucosidase (pH 8.01
y - Glutornyl tronsferose
Alkaline phosphatase
I
5'-Nucleotidose
Density
FIG.3. Isopycnic centrifugation of 8OOO g-min supernatant from homogenate of jejunal biopsy. Homogenate was
prepared and gradient formed from solutions of purified sorbitol. Details are as given for Fig. I . The percentages
of recovered activity are: maltase, 68; a-glucosidase (PH6.0). 78; leucyl-/?-naphthylamidase, 83 ; a-glucosidasc
(pH 8.0). 73 ;y-glutamyl transferase, 73 ;alkaline phosphatase, 87 ;malate dehydrogenase, 103 ;5'-nuclcotidase, 79.
Subcellular fractionation of human jejunum
tion at densities 1-21 and 1.12 corresponding to
alkaline phosphatase and 5’-nucleotidase respectively. Malate dehydrogenase has a major soluble
component with particulate activity at a density of
1.16. Acid Bgalactosidase has a broad peak of
activity at a density of 1-21and is similar to that of
N-acetyl-Bglucosaminidase.Acid phosphatase and
acid diesterase also have broad particulate components but with a significantly lower modal density
of 1.18.
Fractionation on sorbitol density gradknts. Fig. 3
shows the distribution of the various marker enzymes after collection and disruption of the biopsy
in sorbitol and density gradient centrifugation of the
tissue extract on a sorbitol gradient. The organelles
are separated in a similar manner to that achieved
on sucrose gradients although they have higher
equilibrium densities.Thus 1eucyLBnaphthyIamidase
has a modal density of 1.24, compared with a value
of 1.21 in sucrose. Maltase and sucrase (not shown)
have distinct peaks also with modal densities of 1.24.
565
Clear separation of the a-glucosidases when assayed
at pH 6.0 and pH 8-0 is achieved.
Effect of inhibitors on hyakolase distributions.
Fig. 4 shows the distribution of certain acid and
neutral hydrolases in the sucrose density gradients.
BGalactosidase assayed at pH 3.5 shows a symmetrical peak at modal density of 1.21 with a significant soluble component. Total Bgalactosidase
assayed at pH 6.0 (not shown) has a complex distribution with soluble component and broad peak in
the region 1.18-1-22. When activity is assayed in the
presence of pchloromercuribenzoate (0.18 mmolfl)
(resistant enzyme activity) a sharply de5ed peak at
a modal density of 1.22 is obtained. The pchloromercuribenzoate-sensitive Bgalactosidase activity,
obtained by plotting the difference in activity of the
enzymewhen assayed with and without the inhibitor,
shows a similar distribution to that of the acidic
Bgalactosidase, although there is a larger soluble
component.
/?-Glucosidase (Fig. 4) shows a large soluble
Density
FIO.4. hpycnic centrifugationof 8OOO g-min supernatant from jejunal biopsy homopenate. Details a e as given
for Fig. 1. The parcentages of m v m d activity are: &galactosida8e (pH3.5),79; figlucosidase, 92; figdactoeid w (PH60, p c h l o r o m e r m r i b c o z o a ~ M B ~ ~ ~ i t i71;
v e Ja-glucaidasc
,
(pH60). 73; ~gahtosidasc
[PH6.0, pchioromercuribanzoate(pCMB)-resiltant],89; a-glucosldasc(pH8.0). 103.
T. J. Peters
566
1). The activity in the soluble fractions and in the
density region 1.141.18 is strongly inhibited by this
ion. Activity deeper in the gradient is relatively
unaffected. A partial discrimination between the
high- and low-density a-glucosidase activities can be
achieved by assaying the activity in the presence and
absence of mturanose (32 mmol/l). The activity at
density 1.22 is more sensitive to this inhibitor than
the activity at density 1.15.
Fractionation of brush-border components. The
distribution of certain hydrolases in the sucrose
gradients when Trisdisrupted brush borders are
fractionated on a density gradient (Fig. 6)showsthat,
with the exception of Bglucuronidase, for all there is
a peak of activity at density 1.23, which corresponds
to a peak in the proteindistribution.BGlucuronidase
shows a broad peak in the density region 1.13-1-17,
component with activity throughout the gradient and
a distinct component at density 1.22. The distribution of Bglucosidase assayed at pH 6.0 shows
a bimodal distribution with peaks of activity at 1.22
and 1.16. BGlucosidase, assayed at pH 8.0, shows
a distinct peak at density 1.16 corresponding to the
low-density peak found when the enzyme is assayed
at pH 6.0. In addition there is a shoulder of activity
at density 1.22.
The distribution of a-glucosidaseassayed with and
without inhibitors shows that when assayed at pH
6.0 there is a distinct peak at density 1.22 with
a shoulder at 1.15 (Fig. 5), corresponding to the peak
when a-glucosidaseis assayed at pH 8 0 . Discrimination between the high- and low-density a-glucosidase
activities can be clearly achieved by assaying the
activity at pH 6.0 with and without Zn2 (2-8 m o l l
+
15r
a-Glucosidose ( p 6~0 )
:t
a-Glucosidose (DH 801
a-Glucosidase (pH 601
y
5
15t
0
a-Glucosidase
(DH 6.0)
?$d
;o
-s;e
Turanosa-resistant
1.05
k10
1-15
1-20
1.25
I
Density
FIG.5. Isopycnic centrifugation of 8OOO g m i n supernatant from jejunal biopsy homogenate. Details arc as given
for Fig. 1. The percentages of recovered activity are: a-glucosidase @H 6.0). 71 ; a-glucosidasc (pH 6.0, ZnZ+sensitive), 107; a-glucosidase (PH6.0, Zn*+-resistant), 65; a-glucosidase @H 8-0), 106; a-glucosidase @H 6.0,
turanose-resistant),66; a-glucosidase (pH 6.0, turanose-sensitive).66.
Subcellular fractwnution of human jejunum
I
Protein
I
567
8-Glucuronidase
Density
FIG.6. Isopycnic centrifugation of 8000 g-min supernatant from homogenate of isolated jejunal brush borders.
Details are as given for Fig. 1. The percentages of recovered activity are: leucyl-8-naphthylamidase, 106; yglutamyl transpeptidase, 81 ; a-glucosidase(PH 6.0), 96; S’-nucleotidase, 88; 8-glucosidase. 74; alkaline phosphatase. 64; protein, 66; ~-glucuronidase,66.
some soluble activity, but no signiscant activity in
the density 1.23 region. 5’-Nucleotidase shows
a similar distribution to that of alkaline phosphatase,
although there is a small peak of activity in the
region of density 1-12.Protein is distributed throughout the gradient with major components in the
soluble and low-density region of the gradient.
Distribution of acid hydrohes. Fig. 7 shows the
distribution of six acid hydrolases assayed in the
gradient fractions from one experiment. The distributions of the individual enzymes differ sign&
cantly from each other, soluble activity varying from
Bglucuronidase,showing the least, to acid diesterase,
showing the most. The density distributions of the
particulate components vary with /?-glucuronidase
having a modal density of 1.16 and acid /?-galactosidase a density of 1.21. More detailed analysis indicates the existence of at least three distribution patterns with the following modal densities: 1.20-1.21
(N-acetyl-Bglucosaminidase, N-acetyl-/%galactosaminidase, Bgalactosidase); 1.17-1.18 (cathepsin C,
acid diesterase); 1.16-1-20 (acid phosphatase, B
glucuronidase).
Eflects of digitonin on enzyme distributions. The
densitydistribution profiles of the various markex
enzymes after homogenization of the biopsy in SVE
d m containing digitonin (0.12 mmol/l) shows
striking differences when compared with the untreated tissue (Fig. 8). Zinc-sensitive a-glucosidase
assayed at pH 6 9 shows a modal density of 1.12,
T. J. Peters
568
,L
N - Ace1y I-p- G l u c m n i d o s e
Cuthepsin C
I0 C
c
Acid diesterase
Acid phospholose
I0
11.05
1.10
1.15
1.20
1.25
1.30
Oensily
FIG. 7. Isopycnic centrifugation of 8000 g-min supernatant from jejunal biopsy homogenate. Details are as
given for Fig. 1. The percentages of recovered activity are: N-acctyl-8-glucosaminidase,
72; 8-glucuronidase,94;
acid 8-galactosidase, 84; cathepsin C, 65; acid phosphatase, 98; acid diesterase, 109.
significantlyless than the density of 1.16 found in the
control experiments. Zinc-resistant a-glucosidase
@H 6.0), alkaline phosphatase, leucyl-pnaphthylamidase (not shown) and y-glutamyl transpeptidase
(not shown) all show a smaller but consistently
increased modal density of 1.22-1.23 compared with
1.21-1.22 in the control experiments.
The distribution of N-acetyl-pglucosaminidaseis
strikingly different,with complete loss of the particulate component with nearly all the enzyme activity
being recovered in the soluble fraction. In agreement
with this, measurements of latent N-acetyl-8
glucosaminidase(Peters et al., 1975) consistently give
values of less than 3%, compared with the value of
approximately 60% in control tissue (Peters et af.,
1975). Other acid hydrolases showing similar distributions to that of N-acetyl-Bglucosaminidase
after digitonin treatment include: Bgalactosidase,
acid diesterase, N-acetyl-@-galactosaminidase and
cathepsin C. S’-Nucleotidase shows a significant
increase in modal density compared with the control
experiments, with a shift from a density of 1.12 to
1.16. Acid phosphatase shows a greater amount of
activity remaining in the sample layer but much
particulate activity remains. 8-Glucuronidase (not
shown) behaves similarly to acid phosphatase except
that there is a disproportionately greater loss of
particulate activity in the higher density region. The
distribution of catalase and malate dehydrogenase in
the sucrose gradients is similar in both the control
and digitonin-treated experiments, apart from
a slightly higher soluble catalase activity.
Subcellular fractionation experiments were also
performed, in which digitonin (0.06 mmol/l) was
included in the homogenization medium. Only
partial solubilization of the N-acetyl-/?-glucosaminidase (latent activity approximately 20%) was
achieved and the density shifts of the membrane
bound enzymes were less than those achieved with
digitonin (0.12 mmol/l).
Possible variables in the fractionation procedures
were also studied. The tissue-homogenization procedure has been optimized to provide highest
lysosomal integrity, as assessed by latent N-acetyl-8
glucosaminidase (Peters et al., 1975). Tissue extracts
prepared by more vigorous homogenizationthan the
Subcellular fiactwnatwn of human jejunum
569
5- Nucleotidase
Zn*+-sensitive
a-glucosidase
I
Density
FIG.8. Isopycnic centrifugationof 8000 g-mh supernatant from jejunal biopsy homogenized in isotonic sucrose
(SVE medium) containing digitonin (016 mmol/l). Details are as given for Fig. 1. The percentages of recovered
activity are: a-glucosidase @H 6 0 . Zn'+-scnsitive), 106; S'-nucleotidase, 97; a-glucosidase (PH6.0. a'+resistant), 82;acid phosphatase, 102; alkaline phosphatasc, 103; catalast. 93; N-acetyl-8-glucosaminidase,93 ;
malate dehydrogenase. 8 1.
standard procedure were also subjected to centrifugation on sucrose density gradients. Essentially
similar enzyme distributions to those from the
standard procedure were found except that more
acid hydrolase and catalase activities were found in
the soluble fractions. However, the more gentle
procedure was used to preserve organelle integrity,
particularly in diseased tissues. A gentle homogenization procedure also minimizedthe contributionof the
nonepithelial cell organelles to the tissue extract.
Phasecontrast microscopy of the low-speed pellet
(N fraction) revealed many intact interstitial cells.
Tissue extracts, prepared by the standard procedure
were also subjected to centrifugation for longer
periods. After centrifugation for five times the
standard time the organelles had similar modal
densities to those after 35 min centrifugation. However, the particulate enzyme activities tended to have
more distinct and symmetrical peaks. Thus alkaline
phosphatase had a sharp peak at density 1.21-1-22
with significantly less activity in the low-density
region of the gradient. However, the standard fractionation conditions were used to keep centrifugation as short as possible, consistent with adequate
T. J . Peters
570
resolution of the organelles. More recently a titanium
version of the Beaufay rotor has achieved complete
resolution of the organelles from jejunal biopsies by
centrifugation at 50000 rev./min for 15 min. Prolonged centrifugation leads to organelle damage and
disruption, particularly with pathological tissue,
owing to the hypertonic sucrose solutions of the
gradient. Sedimentation of soluble protein into the
body of the gradient may also obscure the lowdensity organelles.
Discussion
These studies form the first descriptions of analytical
subcellular fractionation of jejunal biopsy specimens
in which all the principal organelles have been
characterized. In establishing the fractionation procedure techniques had to be devised for separating
the organelles from milligram quantities of tissue and
for assaying the very small amounts of enzyme
activity associated with these organelles. With
suitable modification the procedures developed for
study of lysosomes of normal and diseased smooth
muscle cells were used (Peters et al., 1972; Peters &
de Duve, 1974). In essence, the small-volume automatic zonal ultracentrifuge rotor of Beaufay (1966)
was used for separating the organelles and highly
sensitive enzyme assay techniques employing fluorigenic or radiolabelled substrates were used to locate
the organelles in the sucrose gradients by means of
their marker enzymes.
TABLE
3. Principal organelles, equilibrium density in sucrose and their associated enzyme acrivities
Organelle
Equilibrium
density
Basal-lateral membranes
1.12
Endoplasmic reticulum
1.16
Mitochondria
1.16
Peroxisome
1.18
Lysosomes
1.16-1.18
1.18
1.20
Brush border
It is stressed that these studies employ an analytical rather than preparative approach. Crane and his
colleagues have recently applied a preparative subcellular fractionation procedure to frozen specimens
of human jejunum mainly obtained by open surgical
biopsy (Welsh, Preiser, Woodley & Crane, 1972;
Schmitz, P r e k r , Maestracci, Ghosh, Cerda & Crane,
1973). They obtained an enriched fraction of brushborder membranes by a complex multi-step subcellular fractionation procedure. The aim of a preparative procedure is to obtain a sample of the particular organelle in a high state of purity but often
with little concern for yield. In humans the preparative approach has given valuable information on the
properties of the brush border (Maestracci, Schmitz,
Preiser & Crane, 1973; Preiser, Mennard, Crane &
Cerda, 1974; Schmitz, Preiser, Maestracci, Crane,
Troesch & Hadorn, 1974) but not on the other
organelles of the enterocyte. In addition the preparative approach to subcellular fractionation of human
jejunum is not readily applicable to pathological
tissue for the centrifugal properties of diseased
organelles may differ significantly from similar
organelles in normal tissue. In the present study
a simple single-step procedure is used to resolve the
principal organelles from the jejunal biopsy. The
activities and certain properties of all the organelles
can be determined in a single biopsy and the technique is readily applicable to diseased tissue.
The centrifugation properties of the individual
organelles and their enzyme contents as deduced
1.21
Associated enzyme activities
5’-Nucleotidase, leucyl-B-naphthylamidase (part), y-glutamyl
transpeptidase (part), alkaline phosphatase (part)
Zn2+-sensitive a-glucosidase, alkaline a-glucosidase, acid phosphatase
(part), j7-glucuronidase (part), B-glucosidase (?) (part)
Succinic dehydrogenase, glutamate dehydrogenase, monoamine oxidase,
malate dehydrogenase (part)
Catalase, D-amino acid oxidase
Acid phosphatase (part), B-glucuronidase (part)
Cathepsin C, acid diesterase
N-Acetyl-8-glucosaminidase,
acid 8-galactosidase, acid phosphatase (part),
8-glucuronidase (part)
Maltase, alkaline phosphatase, leucyl-8-naphthylamidase, y-glutamyl
transpeptidase, Zn’+-resistant a-glucosidase, p-chloromercuribenzoateresistant 8-galactosidase, 8-glucosidase (part)
Subcellular fractionation of human jejunum
from the fractionation techniques are summarized in
Table 3.
Plasma membrane (basal-lateral membranes)
The distribution of 5’-nucleotidase suggests that
it is associated with the plasma membrane. Similar
conclusions have been reached with isolated enterocytes from both rat (Douglas, Kerley & Isselbacher,
1972; Peters & Shio, 1976) and guinea-pig (Lewis,
Elkin, Michell & Coleman, 1975a). This enzyme is
also localized predominantly to the plasma membrane in a wide variety of cell types (see Solyom &
Trams, 1972). There is a second component of 5’nucleotidase, deeper in the gradient, which is
associated with the brush-border elements. Nonspecific brush-border alkaline phosphatase may be
hydrolysing the 5’-nucleotidase substrate, but this
cannot account for all the brush-border 5’-nucleotidase activity. The assay used, based on that of
Douglas et al. (1972), has a high selectivity for this
enzyme. In addition, if gradient fractionsare assayed
with alkaline phosphatase or 5’-nucleotidase inhibitors, the relativedistribution of 5’-nucleotidase in the
gradient is unchanged, indicating that the brushborder activity is not due to non-specific alkaline
phosphatase. It thus appears that 5‘-nucleotidase is
an intrinsic brush-border enzyme. Although the
brush border of the enterocyte is a highly specialized
plasma membrane, it is reasonable to suggest that
despite its specific functional adaptations it has preserved, to some extent, attributes of the normal
plasma membrane.
The increase in density of the plasma membrane
after digitonin treatment is similar to that found in
rat liver (Thinks-Sempoux,Amar-Costesec Beaufay,
& Berthet 1969; Tilkray & Peters, 1976) and in rat
embryo fibroblasts (Tulkens, Beaufay 8c Trouet,
1974), and reikcts its high concentration of digitonincomplexable cholesterol (Amar-Costesec, Wibo,
Thinks-Sempoux, Beaufay & Berthet, 1974).
As judged by their distribution profiles a small
proportion of leucyl-/?-naphthylaxnidase and yglutamyl transpeptidase activities is also associated
with the plasma membranes. A portion of the
aklaline phosphatase may be similarly located, as
confirmed by finding digitonin txeatment of the
homogenate to caw a redistribution of the plasma
membrane component to higher densities (Fig. 8).
Endophmic reticulum
This is the principal component of the ‘microsomal
571
fraction’ of differential centrifugation experiments.
Recent studies have cordinned the heterogeneous
nature of this fraction, which contains elements
derived from the plasma membrane, Golgi complex,
outer mitochondria1 membranes as well as endoplasmic reticulum granulated with ribosomes to
various degrees (Amar-Costesecet al., 1974; Tilleray
& Peters, 1976). Owing to the relatively small
amounts of endoplasmicreticulum in the enterocyte
it has not been possible to assay a range of enzymes
localized to different subcomponents of the microsomal fraction in the human jejunal biopsies. aGlucosidase, assayed at alkaline pH in order to discriminate against the very high activity of the brushborder a-glucosidase, has been used as a marker for
this organelle. The endoplasmic and cytoplasmic
a-glucosidase activities are strongly inhibited by
Zn2+ and thus by plotting the Zn2+-inhibitable
a-glucosidase activity the distribution profile of the
endoplasmic reticulum in the density gradients
can be accurately determined. The use of neutral
a-glucosidase as a marker for the endoplasmic
reticulum has been clearly demonstrated for rat liver
(Lejeune et al., 1963; Tilleray & Peters, 1976),
smooth muscle cells (Peters et al., 1972; Peters &
de Duve, 1974), skeletal muscle (Angelini & Engel,
1973), cardiac muscle (Bloomfield & Peters, 1974)
and human liver (Brown & Brown, 1965; Gamklou
& Schersten, 1972; Seymour, Neale & Peters, 1974).
The distribution of alkaline a-glucosidase in the
sucrose gradients is affected by digitonin treatment
of the biopsy homogenate. Unlike the plasma membrane and brush border, which show increased
densitits, the endoplasmicreticulum (reflected by the
distribution of a-glucosidase), shows a significant
reduction in density. The mechanism of this response
is uncertain but is presumably related to the detergent action of the digitonin and also occurs with rat
liver microsomes (Tilleray & Peters, 1976). There is
also evidence for a partial localization of acid phosphatase, Bglucuronidase and perhaps Bglucosidase
to endoplasmic reticulum of the human intestine,
these enzymes having significant particulate activities
in the same density region as a-glucosidase, at 1.16.
BGlucuronidase and Bglucosidase show similar
decreases to a-glucosidase after digitonin treatment.
Acid phosphatase does not show this effect but
similar differentialresponses of endoplasmic reticulum enzymes, apparently in the same microsomal
subfraction, OCCUT in rat liver microsomes(Tilleray &
Peters, 1976). Studiesin rat liver also show signscant
572
T. J. Peters
amounts of microsomal acid phosphatase (Neil &
Horner, 1964; Beaufay, 1972) and Bglucuronidase
(de Duve, Pressman, Gianetto, Wattiaux 8c Appelmans, 1955; Paigen, 1961; Fishman, Goldman &
Delellis, 1967) in this subcellularfraction in addition
to the lysosomal components.
1968; Peters & Shio, 1976) is consistent with the
absence of uricase from homogenates of human
jejunum (T. J. Peters, unpublished result). In rat liver
the cores of the peroxisomes contain most of the
uricase and contribute to their high density (de Duve
& Baudhuin, 1966; Leighton et al., 1968).
Mitochondria
Lysosoms
These have a modal density of 1.16 but, compared
with theendoplasmicreticulum,the density spread is
considerably less. Enzymes for certain of the mitochondrial subcomponents have been assayed. Thus
succinic dehydrogenase (inner membrane), malate
and glutamate dehydrogenase (matrix) and monoamine oxidase (outer membrane) marker enzymes
(Ernster & Kuylenstierna, 1969) all had identical
distributionswith a sharply deiined modal density of
1-16. This is significantly lower than the modal
density of 1.19 reported for rat (Beaufay et al., 1964;
Leighton et al., 1968; Peters & Shio, 1976) and
human liver (Seymour et al., 1974). However,
a similar figure of 1.16-1.17 was found for the
equilibrium density of rat enterocytes when fractionated under similar conditions (Peters & Shio,
1976). The outer and inner membrane marker
enzymes have similar distributions, indicating that
there is no significant disruption of the mitochondria during the fractionation. Any dissociated
outer mitochondria1 membrane would be expected
to have a significantly lower equilibrium density
(Parsons, Williams & Chance, 1966; Peters-Joris,
Vandevoorde & Baudhuin, 1975), and if the inner
membrane is damaged and rendered permeable to
sucrose the equilibrium density would be approximately 1.22 (Wattiaux, 1974).
Lysosomes have been shown previously to occur
in human jejunum (Peters et al., 1975) according to
the accepted biochemical criteria (de Duve & Wattiaux, 1966). Evidence is now presented for the
heterogeneity of these lysosomes. The principal
lysosomal component has an equilibrium density of
1-20-1.21 and is completely disrupted by digitonin
treatment. A second population, identi6ed by its
unique content of cathepsin C and acid diesterase,
has a lower equilibrium and is solubilized by digitonin treatment. BGlucuronidase and acid phosphatase, although partially localized to the high-density
population of Iysosomes, probably contributemainly
to a third low-densitypopulation of lysosomes which
appear to be relatively resistant to disruption by
digitonin. On the present evidence it is difficult to
distinguish definitively this third population of
lysosomes from elements of the endoplasmic reticulum containing these enzymes. The functional
significanceof these distinctpopulations of lysosomes
remains to be determined.
Peroxisomes
The demonstration of latent and sedimentable
catalase in the tissue extracts and the co-sedimentation of catalase and D-amino acid oxidase on the
density gradients satisfiesthe biochemical criteria for
peroxisomes (Baudhuin, Beaufay, Rahman-Li, Sellinger, Wattiaux, Jacques & de Duve, 1964) and thus
this is the first report of their occurrence in human
jejunum, although they have been conclusively
demonstrated in rat (Peters & Shio, 1976) and guineapig (Connock, Kirk & Sturdee, 1974) enterocytes.
Their significantly lower density than rat liver
peroxisomes (Baudhuin ef al., 1964; Leighton et al.,
Brush borders
Brush borders have the highest equilibrium density (1.21) of the organelles studied, being characterized by disaccharidases and other hydrolases
(alkaline phosphatase, p-chloromercuribenzoateresistant bgalactosidase, leucyl-Bnaphthylamidase,
y-glutamyl transpeptidase and Zn’ +-resistant aglucosidase), many of which have been localized to
this organelle both histochemically and by analysis
of highly purified brush-border membranes isolated
by preparative fractionation procedures (seeSchmitz
et al., 1973). This study confirms these findings, but
demonstrates that certain of the enzymes have
significant extra-brush-border components in other
organelles.The brush borders increase in equilibrium
density after digitonin treatment, although the
increase is less than that for the basal-lateral membrane. This presumably reflects the different lipid
composition of the brush-border membranes (Doug-
Subcellular fructwtuatwn of human jejunum
las et al., 1972; Kwai, Fujita & Nakao, 1974; Lewis,
Gray, Coleman & Michell, 1975b). Similar differencesin response to digitonin have been noted for
basal-lateral and brush-border membranes from
guinea-pig enterocytes (Lewis et al., 1975a).
The subcellularfractionationand separationof the
organelles in human jejunal biopsies proved easier
than in isolated rat enterocyte8, where all the organelles, apart from the basal-lateral membrane, had
very similar equilibrium densities and complex
gradients were neccsp~ly to achieve separations
(Peters & Shio, 1976). Having established the distribution of the various organelles in the gradients,
the localization of hitherto unassigned enzymes and
other componentscan be determined. The technique
will be used to investigate the subcellular pathology
of small-intestinal diseases including the coeliac
syndrome (Peters, Heath, Jones & Peacham, 1975).
Although this new technique has been applied in the
6rst instance to human jejunum it is applicable to
the study of the physiology or pathology of any organ
amenable to biopsy, and thus has potentially a very
wide application in the study of cell pathology.
Acknowlerllpnents
I am particularly grateful to Ms Janet Heath,
Mr P. Hope and Mr P. White for their expert
technical assistance, to Ms Margaret Chandler for
performing the computations and Ms Jean de Luca
for typing the manuscript. The constructivecriticism
of colleagues, particularly Mr R. Batt, and the
enthusiastic support of Professor C. C. Booth, are
gratefully acknowledged. This work is supported by
the Medical Research Council and The Wellcome
Trust.
Re€@€SIW
AHMBD,Z. & R w ,T.L. (1958) The activation and inhibition
of S’-nucleotidase. Biochemical Journal, 69,386-387.
AauJI-CWTaSac, A., WIW, M., TH1”-SesrpovX. D.,
BBAUFAY.
H. & BERTHBT,
J. (1974) Analytical study of
microsomes and isolated subcellular membranes from rat
liver. Biochemical. physical and morphological modification of microsod components induced by digitonin.
EDTA and pyrophosphate. Journal of Cell Bio&gy, 62,
717-745.
ANOBLINI,
C. & ENOEL,A.G. (1973) Subcellulardistribution
of acid and neutral a-glycoridasw in normal, acid maltasc
deficient and myophosphorylase deficient human skeletal
muscle. Archives of Biochemistty and Biophysics, 156,
350-355.
ASP,N-G. & DAHLQVIST,A. (1972) Human small intestine
/?-galactosidases. Speci5c assay of three different enzymes.
Analytical Biochemistry, 47, 527-538.
573
BACQNO,F.M..RITA, O.A. & ZURFITI, M.F. (1971) Studies
on the structurabound scdimentability of some rat liver
lysosome hydrolases. Biochemical Journal, 122,363-371.
BAUDHUIN,P., BBAUPAY,
H., RAHMAN-LI,Y.. SBLLINOBR,
0.2..WATTIAVX.
R., JACQUES,P. & DE DWB, C. (1964)
Tissue fractionation studies. 17. Intracellular distribution
of monoamine oxidases, aspartate aminotransferasc,
alanine aminotransferase, mamino acid oxidase and
catalase in rat liver tissue. Biochemical Journal, 92,179-184.
BMUFAY,
H. (1966) La centrifugationen gradient de densit&.
These d‘Agdgation de I’Enseignement Superieur Universitc Catholique dc Louvain, pp. 132. Ctutarick SA.
Louvain, Belgium.
BMUFAY.H. (1972) The non-lysosomal localisation of acid
(pnitro)phenyl-phosphatascactivity in various tissues. In:
Lysosotnea. A Laboratory H a h o o k , pp. 33-35. Ed.
Dingle, J.T. North Holland, Amsterdam.
BBAUFAY,
H., JACQUES, P.. BAUDHUIN,
P., SELLINGBR,
O.Z.,
BERTHLIT, J. & DB DUVE, C. (1964) Tissue fractionation
studies. 18. Resolution of mitochondrial fractions from
rat liver into three distinct populations of cytoplasmic
particla by means of density equilibration in various
gradients. Biochemical Journal, 92,184-205.
B L O O F
~.J., & PETERS,
T.J. (1974) Analyticalsubcellular
fractionation of guinea pig myocardium with special
rcfaencc to the localisation of the adenosine triphosphatasea. Biochemical Society Transactions, 2,1098-1 101.
BODANSKY,0. (1946) Mechanism of inhibition of phosphatasc by glycine. Journal of Biological Chemistry, 165,
605613.
BROWN,B.I. & BROWN,D.H. (1965) The subcellular distribution of enzymes in type II glycogonesis and the
ocmmnce of M oligo-a-1.4-glucan glucohylrolaw in
human tissws. Biochimica et Biophysica Acra. 110.124-133.
CAMPBELL,D.M. (1962) Determination of 5’-nucleotidase in
blood serum. Biochemical Journal, 82.34~.
CONNOCIC,
M.J.. KIRK,P.P. & STURDBB,
A.P. (1974) A zonal
rotor method for the preparation of microperoxisomes
from epithelial cells of guinea pig small intestine.Journal of
Cell Biology. 61, 123-133.
DAHLQW. A. (1964) Method of assay of intestinal disaccharidases. Analytical Biochemistry, 7, 18-25.
DB Dwe, C. & BAUDHUIN,
P.H. (1966) Peroxisomes (microbodies and related particles). Physiological Reoiews. 46,
323-357.
DUVE,C. & BERTHLIT,
J. (1954) The usc of differential
centrifugation in the study of tissue enzymes. International
Review of Cytology, 3,225-275.
DB DWB, C., BBRTHLIT.
J. & BMUFAY.H. (1959) Gradient
centrifugation of cell particles: theory and applications.
ProgressinBiophysicsandBiophysical Chemistry,9,325-369.
DS Dvve, C., F
”
,
B.C., GIANBTTO,
R., W A ~ A W X
R.,
& APPLBMANS.
F. (1955) Tissue fractionation studies.
6. Intracellular distribution patterns of enzymes in rat liver
tissue. Biochemical Journal, 60,604-617.
DB DWB,C. & WATTIAVX.
R. (1966) Functions of lysosomes.
Annual Review of Physiology, 23,435-492.
DOWOW,A.P.. KERLBY,R. & ~ S ~ ~ L ~ A K.J.
C H ~(1972)
R,
Preparation and characterization of the lateral and basal
plasma membranes of the rat intestinal epithelial cell.
Biochemical Journal, 128,1329-1338.
E m ,G. & GOLDBBRO,
D.M. (1972) Optimal conditions for
the kinetic assay of serum glutamate dehydrogcnase
activity at 37OC. Clinical Chemistry, 18,523-528.
ERNSTER,
L. & KUYLBNSTIBRNA,
B. (1969) Structure, composition and function of mitochondrial membranes.
FEBS Symposium. 17, 5-3 1.
FISHMAN,
W.H., GOLDMAN,
S.S. & DBLRLLIS.R. (1967)
Dual localisation of &glucuronidase in endoplasmic
reticulum and in lysosomes. Nature(London), 213,457460.
DB
574
T.J. Peters
GAMKLOU,
R. & SCHERSTEN,
T. (1972) Activity of a-amylase
and a-I-Cglucosidase in subfractions of human liver homogenates. Scandinaoian Journal of Clinical and Laboratory
Investigation, 30, 209-21 3.
GUILBAULT,G.G., BRIONAC,P. & ZIMMER,M. (1968)
Homovanillic acid as a fluorometric substrate for oxidative
enzymes. Analytical applications of the peroxidase,
glucose oxidase and xanthine oxidase systems. Analytical
Chemistry, 40, 190-196.
KAWAI,K., FUIITA,M. & NAKAO,M. (1974) Lipid components of two different regions of an intestinal epithelial
cell membrane of mouse. Biochimica et Biophysica Acta,
369,222-233.
L m m , N., THIN~S-SBMPOUX,
D. & HERS, H.G. (1963)
Tissue fractionation studies. 6. Intracellular distribution
and properties of a-glucosidases in rat liver. Biochemical
Journal, 86, 16-21.
LEIOHTON,F., POOLE,B., BEAUFAY,
H., BAUDHUIN,
P.,
COFFEY,J.W., FOWLER,S. & DE DWE, C. (1968) The
large-scale separation of peroxisomes, mitochondria and
lysosomes from livers of rats injected with Triton WR 1339.
Journal of Cell Biology, 31,482-513.
LEWIS,B.A., ELKIN, A., MICHELL,R.H. & COLEMAN.
R.
(1975a) Baso-lateral plasma membrane of intestinal epithelial cells. Identification by lactoperoxidase-catysed
iodination and isolation after density perturbation with
digitonin. Biochemical J O W M ~ 152,71-&4.
,
.EWIS, B.A., GRAY,G.M., COLEMAN,
R. & MICHELL,R.H.
(1975b) Differences in the enzymic, polypeptide, glycopeptide, glycolipid and phospholipid composition of plasma
membranes from the two surfaces of intestinal epithelial
cells. Biochemical Society Transactions, 3,752-753.
DWRY, O.H., ROBERTS,N.R. & KAPPHAHN,J.I. (1957) The
fluorometric measurements of pyridine nucleotides.
Journal of Biological Chemistry, 224, 1047-1064.
MABSTRACCI,
D., SCHMITZ,
J., PREISER,H. & CRANE,R.K.
(1973) Proteins and glycoproteins of the human intestinal
brush border membrane. Biochimica et Biophysica Acta,
323,113-124.
MWER, M. & DAHLQVIST,
A. (1966) A one-step ultramicro
method for the assay of intestinal disaccharidases. Analytical Biochemistry, 14,376-392.
MILUNOTON,P.F., CRITCHLEY,
D.R. & TOVELL,P.W.A.
(1966) The role of calcium in the isolation of brush border
from epithelial cells of rat small intestine. JownaI of Cell
Science, 1, 415424.
NEIL, M.W. & HORNER,M.W. (1964) Studies on acid
hydrolases in adult and foetal tissues. 2. Acid phenyl
phosphomonoesterases of adult mouse liver. Biochemical
JOWMI,93,220-224.
PAIQEN,
K. (1961) The effect of mutation on the intracellular
location of /3-glucuronidase. Experimental Cell Research,
25,286-301.
PARSONS,D.F., WILLIAMS,G.R. & CHANCE,B. (1966)
Characteristics of isolated and purified preparations of the
outer and inner membranes of mitochondria. Annals of the
New York Academy of Sciencs. 131, 646-666.
PEETERS-JORIS,C., VANDEVOORDE,
A-M. & BAUDHUIN,
P.
(1975) Subcellular localisation of superoxide dismutase in
rat liver. Biochemical Journal, 150.31-39.
PETERS,T.J., BAIT, R.M., HEATH, J.R. & TILLERAY,
J.
(1976) The micro-assay of intestinal disaccharidases.
Biochemical Medicine, 15, 145-148.
PETERS,
T.J. & DE D w e , C. (1974) Lysosomes of the arterial
wall. 11. Subcellular fractionation of aortic cells from
rabbits with experimental atheroma. Experimental and
Molecular Pathology, 20, 228-256.
PETERS,T.J.,HEATH,J.R., JONES,P.E. & PEACHAM,
A.D.
(1975) Subcellular fractionation studies on jejunal biopsies
from control subjects and from patients with coeliac
disease. Gut, 16, 826.
PETERS,T.J., HEATH,J.R., WANSBROUGH-JONES,
M.H. &
DOE, W.F. (1975) Enzyme activities and properties of
lysosomes and brush borders in jejunal biopsies from control subjects and patients with coeliac disease. Clinical
Science and Molecular Medicine, 48,259-267.
PETERS,T.J., M~LLER,
M. & DB D w , C. (1972) Lysosomes
of the arterial wall. I. Isolation and subcellular fractionation of cells from normal rabbit aorta. JournaI of Experimental Medicine. 136, 1117-1 139.
PETERS,T.J. & SHIO,H. (1976) Analytical subcellular fractionation studies on rat liver and on isolated jejunal
enterocytes with special reference to the separation of
lysosomes, peroxisomes and mitochondria. Clinical Science
and Molecular Medicine, 50, 355-366.
PENNINGTON,
R.J. (1961) Biochemistry of dystrophic muscle.
Mitochondria1succinate-tetrazolium reductase and adenosine triphosphatase. Biochemical Jowna/, 80, 649-654.
PREISBR,H., MENNARD,D., CRANE,R.K. & CERDA,J.J.
(1974) Deletion of enzyme protein from the brush border
membrane in sucrase-isolactase de6ciency. Biochimica et
Biophysica Acta, 363,279-282.
SCHMITZ,
J., PREISER,H., MABSTRACCI,
D., CRANE,R.K.,
TROESCH,
V. & HADORN.B. (1974) Subcellular localisation
of enterokinase in human small intestine. Biochimica er
Biophysica Acta, 343,435-439.
SCHMITZ,
J., PREISER,H., MAESTRACCI,
D., GHOSH,B.K.,
CERDA,
J. &CRANE,R.K. (1973) Purification of the human
intestinal brush border membrane. Biochimica et Biophysics Acta, 323, 98-1 12.
SEYMOUR,
C.A., NEALE,G. & PETERS,
T.J. (1974) Analytical
subcellular fractionation studies on human liver biopsies.
Biochemical Society Transactions, 2, 1101-1 104.
SOLYOM,
A. & TRAMS,E.G. (1972) Enzyme markers in
characterisation of isolated plasma membranes. Ehzyme,
13,329-372.
SUZUKI,I. & KUSHIDA,H. (1973) Studies on mammalian
glyccsidases. V. Effects of metal ions upon acidic and
neutral a-glucosidases, 8-galactosidases and a-mannosidases in liver extracts of rabbits. JownaI of Biochemistry,
74,627429.
SZASZ, G. (1969) A kinetic photometric method for serum
gamma-glutamyl transpeptidase. Clinical Chemistry, 15,
124-136.
THINBP-SEMPOUX,
D., A ~ ~ ~ - c o s r e S A.,
n c , BEAUPAY,
H. &
BERTHET,J. (1969) The association of cholesterol, 5'nucleotidase and alkaline phosphodiesterase 1 with a distinct group of microsomd particles. Journal of cell
Biology, 43, 189-192.
TILLERAY,
J. & PETERS,
T.J. (1976) Analytical subfractionation of microsomes from the liver of control and GUMstrain rats. Biochemical Society Transactions, 4, 248-250.
TULKENS,
P., BEAUPRAY,
H. & Tnove~,A. (1974) Analytical
fractionation of homogenatcs from cultured rat embryo
fibroblasts. Journal of Cell Biology, 63, 383401.
VANHA-PBRT~ULA.
T., HOPSU, U.K., SONNINEN,V. &
GLENNER,
G.G. (1965) Cathepsin C activity as related to
some histochemical substrate. Histochemie, 5, 170-181.
WA~AUX
R., (1974) Behaviour of rat liver mitochondria
during centrifugation in a sucrose gradient. Molecular and
Cellular Biochemistry, 4.21-29.
WELSH, J.D., PREISER,H.. WOODLEY,
J.F. & CRANE,R.K.
(1972) An enriched microvillus membrane preparation
from frozen specimens of human small intestine. Gastroenterology, 62,572-582.
WURTMAN,
R.J. & AXELROD,
J. (1963) A sensitive and specific
assay for the estimation of monoamine oxidase. Biochemical Pharmacology, 12,1439-1441.