CLIN.
CHEM.
33/1,
(1987)
52-54
Microbiological
Assay
for Vitamin
P. Kelleher,1 Kieran G. Walshe,2 John
Brian
B12 with Use of a Colistin-Sulfate-Resistant
M. Scott,2
and Sean
In this simplified microbiological
assay for serum vitamin B12,
Lactobacillus
Ieichmanhi
(NCIB 8117)
adapted
to tolerate
high concentrations
(500 mg/L) of the polymyxin
antibiotic
colistin sulfate
is used. Results
were similar in parallel
experiments
in which we used both the parent
strain of L.
Ieichmanll (NCIB 8117), and the new adapted strain. Evaluation of assay performance
showed excellent analytical recovery of added cyanocobalamin
(97%,
SD 3%) and good
interassay
and intra-assay
precision (CV <5%).
This modified
assay
system
obviates
the
need
to
sterilize
culture
assay manipulations
may be carried
out openly,
without
aseptic
precautions.
Moreover, this adapted organism would be suitable for use in
an automated
microbiological
assay system.
medium
Consequently,
and glassware.
Microbiological
assays
of vitamin
B12 in serum
have been
extensively
used as an index of cobalamin
status
in man and
give consistent
results,
with good clinical
correlates
(1-4).
The stringent
aseptic
precautions
necessary
in such assays
(5, 6) make them both long and tedious.
The development
of
radioisotope
dilution
assays
for vitamin
B12 in the 1960s (7,
8) led to the widespread
use of several
different
commercially available
kit systems.
However,
radiodilution
assays
based
on the use of intrinsic
factor
as a competitive
cobalamm binder
have
given
clinically
misleading
results
(9, 10)
and continue
to do so despite
considerable
modifications
of
methodology
(6).
The
development
of
an
for folates
in which
a
of Lactobacillus
casei was
that assay (11). The objective
of the
present
study
was to develop
a similar
type
of assay
for
vitamin
B12. Such an assay would be both clinically
reliable
and easily
performed
in the general
diagnostic
laboratory.
chloramphenicol-resistant
used greatly
simplified
Materials
and
assay
strain
Methods
Materials
Colistin
sulfate
(BP
grade)
was obtained
as a sterile
powder from Pharmax
Limited,
Bexley,
Kent, U.K. Vitamin
B12 medium
(usp grade) was obtained
from
Difco Laboratories, Detroit,
MI 48201.
Lactobacillus
leichmanii
(NCIB
8117) was obtained
from Torry
Research
Station,
Aberdeen,
Scotland.
All other chemicals
used
were of reagent
grade.
Glass-distilled
water
was used throughout.
Assays
were
completed
in 13 x 100mm
glass culture
tubes, and we made
additions
of extract
with
a “Stepper”
automatic
pipette
(Socorex,
1020, Renens,
Switzerland).
Tubidimetric
readings were made
in a Beckman
Model
35 spectrophotometer
(Beckman,
High
Wycombe,
Bucks
HP12 4SL, U.K.).
1
Department
of Haematology,
James’s
Hospital,
James’s
2Department
of Clinical
Hospital, Dublin 8, Ireland.
3To whom correspondence
Received
April 14, 1986;
52
CLINICAL
CHEMISTRY,
Central
Street,
Pathology
Dublin 8, Ireland.
Trinity
College,
Medicine,
should
accepted
Laboratory,
St. James’s
be addressed.
September
22, 1986.
Vol. 33, No. 1, 1987
St.
Organism
D. O’Broln’
Procedures
Development
of the new strain.
With
use of broad-spectrum antibiotics,
a sensitivity
proffle of L. leichmanii
revealed
that the organism
has a natural
resistance
to the
polymyxin
antibiotic,
colistin
sulfate.
This
inherent
resistance was enhanced
as follows:
vitamin
B12 medium
was
reconstituted
(56.7 g/L), dispensed
in 20-mL volumes
into
glass
screw-capped
containers,
and autoclaved
at 115 #{176}C
for
15 mm. When
the medium
was cool, we added 250 ng of
vitamin
B12 and stored
the broths
at -20 #{176}C.
The organism
was serially
subcultured
in broths
containing
colistin
sulfate in a range of concentrations
from 0 to 500 mg/L by the
following
protocol.
The colistin
sulfate
concentration
was
increased
stepwise
by 2.5 mg/L each time in the range
0-25
mg/L,
by 5 mg/L
each time in the range 25-50
mg/L, by 10
mg/L
each
time
in the range
time
in the
range
50-100
mg/L,
by 25 mg/L
mg/L,
and by 50 mg/L each
mg/L.
Broths
were inoculated
by
growth
was observed
in about
48
100-200
in the range
200-500
each time,
and good
each
time
loop
h in
the range 0-200 mg/L, and in about 24 h when the organism
was adjusted
to a tolerance
above 200 mg of antibiotic
per
liter.
The organism
was maintained
in a medium
having
a
colistin
sulfate
concentration
of 500 mg/L. Subsequently
its
response
to added vitamin
B12 was evaluated
at different
concentrations
of antibiotic
under
standard
assay
conditions.
Evaluation
of the new strain. We compared
the response
of
the colistin-sulfate-resistant
organism
with that of the parent strain
(NC1B
8117)
in parallel
assays,
utilizing
both
cyanocobalaniin
standards
and serum
extracts.
Colistin
sulfate
was used to maintain
sterility
in one assay,
which
was completed
without
aseptic
precautions,
whereas
the
other
assay
was completed
aseptically
in a laminar-flow
cabinet.
Blood
was taken
into plain glass tubes and serum
was separated
by centrifugation
(1000
x g, 10 miii).
Serum
“extracts”
were prepared
by a 10-fold dilution
of 1 mL o
serum
with
extraction
buffer
(NaOH,
8.3
mmolIL,
CH3COOH,
20.7
mmol/L,
NaCN,
0.45
mmol/L,
pH 4.5).
After
autoclaving
(115 #{176}C,
10 mm) extracts
were “rimmed,’
centrifuged
(1000 x g, 10 miii), and dispensed
(2 x 1.0
and 2 x 0.5 mL) into glass culture
tubes
for each assay.
range of cyanocobalamin
concentrations
of 6.25, 12.5,25,50,
and 100 pg per tube, together
with appropriate
blanks,
w
also dispensed.
Extraction
buffer
was added,
where app
priate,
to give a final volume
of 1 mL in all instances.
Vitamin
B12 assay
medium
was reconstituted
(56.7 g/L) an
dispensed
in 3-mL
volumes
into
all tubes
to give a
volume of 4 mL. Medium
dispensed
for the colistin-sulfa
resistant
organism
contained
100 mg of this antibiotic
pe
liter.
Both organisms
were subcultured
by loop into broths,
one
into basic vitamin
B12 culture
broth as described
and the
other into a similar
broth incorporating
500 mg of colist
sulfate
per liter. After incubation
at 37 #{176}C
for 16 h, inocul
were prepared
by centrifugation
of 5 mL of each (1000 X g,
1000
10 mm)
and resuspension
in 10 mL of sterile water. Inoculum strength
was standardized
turbidimetrically.
After
50fold re-dilution
of inoculum
with sterile
water,
the appropriate assay tubes
were inoculated
dropwise
and incubated
at
37 #{176}C
for 42 h.
A comparison
of two assays. We compared
the results
for
serum
vitamin
B12 by both the colistin-sulfate-based
assay
as described
and a conventional
method
in routine
use in
our laboratory.
The conventional
method was based on that
of Spray (2), modified
to include
a final autoclaving
(115 #{176}C,
6 mm)
of the tubes containing
medium
and vitamin
B12
extracts
before inoculation.
We determined
the consistency
and reproducibility
of
values
for control
serum
and the analytical
recovery
of
added cyanocobalamin
with this assay
procedure.
801
60O
ie
I-
N
400
C,
..
.#{149}
=
...
..
200
#{149}
r=097
Results
Figure
cobalamin
1 shows response
curves
in media
containing
of the organism
various
200
to cyano-
concentrations
800
600
1000
B12 (ng/L)
of
colistin
sulfate.
A colistin
sulfate concentration
of 100 mg/L
was chosen
for use in the routine
assays
because
the
concentration
drop from 500 mgfL to 100 mg/L gives a good
growth
response
while maintaining
sterility.
We evaluated,
in parallel
assays,
the comparative
responses
of both the colistin-sulfate-resistant
organism
and
the parent
strain,
using
aliquots
of the same
extracts,
medium,
and standards
for both (Figure
2). The coefficient
of
correlation
(r) was 0.97 (n = 84), which
is highly
significant
(p <0.001).
A subsequent
comparison
of 90 serum
vitamin
B12 values
estimated
by both
the present
assay
and the
conventional
technique
gave similar
results
(Figure
3): r
was 0.98 (p <0.001).
The performance
data for interassay
and intra-assay
variations
are
shown
in Table
1. Precision
between
and
within
assays
was good, CVs being <5% in both. Analyticalrecovery
experiments
were performed
by adding
known
amounts
of cyanocobalamin
standard
to 10 samples
at each
400
L.LEICHMANII
-
ASEPTIC
TECHNIQUE
of serum
vitamin B12 values obtained in a
of assays utilizing both a colistin-sulfate-resistant
organism and the parent strain (NCIB 8117) (n = 84)
Use of the resistant organism obviates an autodaving step in technique. To
obtain a true compaflson, we added extracts aseptically in the assay involving the
non-resistant parent strain
Fig. 2. Regression
direct comparison
analysis
iooo
800
600
-
In
N
.
400
-J
#{149}#{149},
I.
0
C,
.
#{149}
1.
20O
V= 106x-7.26
-j
U
w
-I.
I
200
I
400
600
800
1000
B12 (ny/I)
L.LEICHMANII
-CONVENTIONAL
AUTOCLAVING
TECHNIQUE
Fig. 3. Comparison
of serum vitamin 812 concentrations
as measured
with the present procedure and the conventional
technique (n = 90)
The “conventional” technique requires an autodaving step to maintain sterility
of four different
shows
concentrations,
and assaying
them.
Table
2
the results.
Discussion
Megaloblastic
CYANOCOBALAMIN
Fig. 1. Response
of the colistin-sulfate-resistant
balamin at different concentrations
of antibiotic
Colistin sulfate concentration (mgIL): O-O (zero); IJ-D
(500)
(pg/TUBE)
organism to cyanoco(25-200);
-U
(400);
anemia
in man
as a result
of deficiency
in
either
folate
or vitamin
B12 presents
with identical
hemstological
findings,
but it is essential
to distinguish
between
these deficiencies
because
low cobalamun
status
may cause
irreversible
neurological
damage.
In addition,
inappropriate
treatment
of vitamin
B12 deficiency
with folate can mask
or
exacerbate
the nerve
damage
associated
with
such a deficiency. Cobalamin
concentrations
in serum,
when measured
by microbiological
assay,
correlate
well with clinical
findCLINICAL
CHEMISTRY,
Vol. 33, No. 1, 1987
53
Table
1. ReproducibilIty
Estimated
of Control
Serum
by the New Method
Mean
Infra -assay
4.67
8.64
12.13
355
495
3.38
3.27
3.42
23.64
Iriterassay
3.80
3.06
9.01
4.02
3.86
4.84
124
224
370
14.30
504
24.39
Table
2. AnalytIcal
Recovery
of Added
from Serum
contamination
been
observed.
microbiological
assay
is ideally
suited
for
use in the general
diagnostic
laboratory.
It offers simplicity
of assay technique
and the clinical
accuracy
associated
with
more conventional
microbiological
assays.
The colistin-sulfate-resistant
organism
would
also be suitable
for use
automated
microbiological
assay
systems.
Cyanocobalamln
Recovery,
100
10
=
ings
95
192
95.00
407
767
101 .75
95.88
%
96.00
throughout.
(3,4).
Nowadays,
however,
most
diagnostic
laboratories
estimate
kits
are
serum
vitamin
commercially
technical
difficulties
B12 by radiodilution
assay.
Assay
available
and easy to use, whereas
the
associated
with
microbiological
assays
have confined
their use to large hospitals
and teaching establishments
(5,6). The observation
that radiodilution
assays
in which impure
intrinsic
factor is used as a binder
will lead to a misdiagnosis
of pernicious
anemia
in patients
(9, 10) has led to significant
modifications
of assay
systems
(12, 13). Despite
these
modifications,
misdiagnoses
still
occur
(6, 14). This has led to an erosion
of confidence
in the
reliance
that can be placed on serum vitamin
B12 estima-
tions and has
testing
(6).
prompted
unnecessary
additional
diagnostic
of the colistin-sulfate-resistant
L. leichmanii
compared
well with that of the parent
strain
(NCIB
8117) in parallel
assays.
The assays
were similar-the
same
serum
extracts
and standards
are used-but
colistin
sulfate
at a concentration
of 100 mg per liter was included
in media
dispensed
for the resistant
organism.
Correlation
of 84
results
for serum vitamin
B12 was good (Figure
2), as was
analytical
recovery
of added cyanocobalamin
(97%,
SD 3%)
and precision
(Tables
1 and 2).
The present
microbiological
assay
for serum vitamin
B12
offers considerable
advantages
over traditional
microbiological assay
techniques.
L. leichmanii
(NCIB
8117)
has an
inherent
resistance
to the antibiotic
and may
be easily
The
adapted
54
performance
to tolerate
high
CLINICAL CHEMISTRY,
in
References
ng/L
200
400
800
Medium
This modified
Measured
Added
are
not needed.
by autoclaving
and
dispensing,
rather
than
bacterial
10, throughout.
=
n
sterilized
CV, %
138
264
need
not be preis thus inoculated
in bulk
before
by addition
of inoculum
to
individual
tubes.
Such bulk “seeding”
of medium
is not only
more convenient,
it also produces
less tube-to-tube
variation. Tubes
may now be “capped”
by covering
whole racks
with
“Paraflim,”
obviating
the use of individual
caps that
must be recycled.
Over 8000 routine
estimations
of vitamin
B12 in serum
have been completed
by use of this simplified
microbiological
assay;
in no instance
has any extraneous
precautions
as
SD
ng/L
n
Values
concentrations
(Figure
Vol. 33, No. 1, 1987
1). Aseptic
1. Rosenthal
BR, Sarett
HP. The determination
of vitamin
activity
in human
serum.
J Biol Chem 1952;199:433-42.
2. Spray GB. An improved
method
for the rapid estimation
vitamin
B12 in serum. Clin Sci 1955;14:661-7.
B12
3. Temperley
serum
B12
estimations
1965;475:317-25.
serum vitamin
B12
IJ, Collery D. The significance
of
in clinical
practice. Irish J Med Sci
4. Mollin DL, Anderson
BB, Burman
JF. The
level: its assay and significance.
Clin Haematol
vitamin
of
1976;5:521-46.
I. In: The megaloblastic
anaemias.
2nd ed. London:
Blackwell Scientific
Publications,
1979:130-1.
6. Lindenbaum
J. Status
of laboratory
testing in the diagnosis of
megaloblastic
anaemia.
Blood 1983;61:624-7.
5. Chanarin
7. Lau K, Gottleib
C, Wasserman
serum
vitamin
B12 level using
charcoal.
Ibid., 1965;26:202-14.
CR, Herbert
radioisotope
V. Measurement
of
dilution
and coated
8. Frenkel
EP, Keller S, McCall MS. Radioisotopic
vitamin
B12 with the use of DEAE cellulose.
assay of serum
J Lab Clin Med
1966;68:510-22.
9. Cooper
pernicious
cobalamin
BA, Whitehead
VM. Evidence
that some patients
with
anaemia
are not recognised
by radiodilution
assays
for
in serum.
N Engl J Med 1978;299:816-18.
10. Kolhouse
mm analogs
JF, Kondo H, Allen NC, Podell E, Allen RH. Cobalaare present
in human
plasma
and can mask cobalainin
deficiency
because
current
radioisotope
dilution
assays
are not
specific for true cobalamin.
ibid., 1978;299:785-92.
11. Davis RE, Nicol JD, Kelly A. An automated
method
measurement
of folate activity. J Clin Pathol 1970;23:47-63.
12. Kubasic NP, Ricotta
M, Sine HE. Commercially-supplied
ers for plasma
cobalamin
(vitamin
B12) analysis-”purified”
sic factor, “cobinamide”
blocked
R-protein
intrinsic
factor-R-protein
binder-compared
assay. Clin Chem 1980;26:598-600.
13. Chen 1W, Silberstein
EB, Maxon
Clinical
significance
of serum vitamin
munoassay
using pure intrinsic
factor.
binder,
with
and
for the
bindintrinnon-purified
microbiological
HR, Sperling M, Barnes
E.
B12 measured
by radioimJ Nucl Med 198 1;22:447-51.
14. Gijzen AHJ, de Kook HW, Meulendijk
PN, et al. The need for a
sufficient
number
of low level sera in comparisons
of different
serum
vitamin
B12 assays. Clin Chim Acta 1983;127:581-95.