MICRO BIO LO GY TO PIC S

Determination of Inoculum
for Microbiological Testing
Scott Sutton

“Microbiology Topics” discusses various topics in microbiology of practi-

cal use in validation and compliance. We intend this column to be a useful
resource for daily work applications.
Reader comments, questions, and suggestions are needed to help us fulfill
MEDICALRF.COM/GETTY IMAGES

our objective for this column. Please send your comments and suggestions
to column coordinator Scott Sutton at scott.sutton@microbiol.org or journal
managing editor Susan Haigney at shaigney@advanstar.com.
OURC MEDICALRF.CO

IMPORTANT
MPOR NT POIN POINTS
Thee fo
followingng ke
key points
oints are discussed:
ussed:
tt
Quality concontrol microbio
microbiology tests re requiree contr
controlled levels
lev of inoc-
i
IMAGE SOURCE,

ula and require

ul qui fresh
fres preparations
pre a ti ns oof ccells for those
lls fo ho iinocula
t
Thee con
t
Th concentration
entra ion ofo cells
cell in a suspension
s ns on can can be estimated
stim by opti-
cal density, but this must be confirmed by plate count
IMA

tt
The
t
Th
The
he optical
oop
ptical density
denssityy readings
rree ding
ngs
gs against
aggaiin
nst cell
cel
ell mass
masss aree specific
sp
pecific
ci to th
the
he
microorganism
cr species
t
The qualification of these readings must be confirmed after major
maintenance to the benchtop spectrophotometer (e.g., after replace-
ment of the bulb).

INTRODUCTION
The ability to quickly estimate the number of viable cells in a microbial
suspension is important in several laboratory applications. It is a critical
component of method suitability tests where the usual expectation is to
demonstrate recovery of low numbers of a set of challenge organisms
in the presence of the product. This is used for sterility tests, microbial
limits tests, and the antimicrobial effectiveness test (AET). The AET
also uses a high inoculum as the challenge system for the test. The most
direct method for determining the microbial count is to plate the micro-
bial suspension, and then infer the number of cells from the number of
colony-forming units (itself, of course, only an approximation). How-
ever, this method requires a significant amount of time (minimum of
18 hours) and is not suitable for these methods, as they require “fresh”
suspensions.

Summer 2011 Volume 15 Number 3 49

MICROBIOLO GY TO PIC S

DETERMINATION OF INOCULUM FOR THE AET “To harvest the…cultures, use sterile saline…Add
The compendial antimicrobial efficacy test (AET) sufficient…to obtain a microbial count of about 1 x
requires inoculation of the product with microor- 108 CFU per mL…[Note: The estimate of inoculum
ganisms to a final concentration of approximately concentration may be performed by turbidimetric
106 CFU/mL. Although this seems to be a minor measurements for the challenge organisms. Refriger-
point, it does serve to illustrate some of the inher- ate the suspension if it is not used within 2 hours].
ent difficulties in microbiological testing and the Determine the number of CFU per mL in each
need for experienced and academically trained suspension …to confirm the initial CFU per mL
microbiologists to head the laboratory. estimate. This value serves to calibrate the size of
The European Pharmacopoeia (1) instruction on the inoculum used in the test.”
preparing the inoculum for the AET states:
“To harvest the…cultures, use a sterile sus- These USP instructions have the advantage of
pending fluid…Add sufficient suspending fluid to being physically possible to perform, an advantage
reduce the microbial count to about 108 micro-or- that cannot be underrated. However, the turbido-
ganisms per milliliter…Remove immediately a suit- metric measure of the cells is also only an approxi-
able sample from each suspension and determine mation of CFU. Thus the instruction to confirm the
the number
n of colony-forming units per milliliter in numbers (after the test is underway) with the plate
each
ch suspension by plate count or membrane filtra- count is an important control on the test.
tion
on (2.6.12). This
Th value serves to determine the
termine th This article explores
his articl plores the turbido
turbidometric
ric approxi-
i
iinoculum andd the baseline
inocu ne to use in the
he test. The
T mation cell numbers,
on for ce umber the important
m controls
nt con
immediately.””
suspensionss shall be usedd immedi
susp on the process, and
he proce nd the potential
nti pitfalls
falls to the
t
process.
p es
There are, of course,
ours two tw problems
prob ems w withh tthese
ese
instructions.. Th
The firs
first is th
that the technician
echn cian iss THEOR
THEORY
instructed to use an inoculum of about 10 8 mi mi- Light scattering techniques to monitor the concon-
croorganisms
crooorg
rganisms
gan smm s per
p r milliliter
pe m
millil
il ite and
a nd then
an th
heen
n instructe
instructed
in
n str uctted centration
entraatiion
en on of
o pure
pu
ure
r cu
cultures
ultu
ures ha
have
ave the eno
enormous
orm
mouus
to ddetermine this
h b by plate
l te count. Colony-forming
C l f advantages
dvaan ge of being
b rapidd and
d nondestructive.
d
units (CFU) and cells (i.e., micro-organisms and However, they do not measure cell numbers nor do
spores) are different measures. This will inevi- they measure CFU. Light scattering is most closely
tably lead to difficulties as the unfortunate lab related to the dry weight of the cells.
worker cannot guarantee the number of cells in Light is passed through the suspension of micro-
the suspension, only the number of CFU found. organisms, and all light that is not absorbed is re-
However, we can accept the scientific inaccuracy, radiated. There is a significant amount of physics
as the numbers will generally work out. The more involved in this, and those interested are referred
serious problem is the instruction to use the plate to optical treatises, particularly those discussing
count CFU for determination of the inoculum for Huygens’ Principle (a good choice is Light Scatter-
the test, and that the suspension shall be used ing by Small Particles by H C Van De Hulst). For
immediately. This, quite frankly, cannot be done. our purposes, it is enough to say that light passing
If you use the suspension immediately, the plate through a suspension of microorganisms is scat-
counts are unavailable; if you use the plate counts tered, and the amount of scatter is an indication of
to set the inoculum, then the suspension is at least the biomass present in the suspension. In visible
a day old. light, this appears “milky” or “cloudy” to the eye
Contrast these instructions with those in the (3). It follows from this that if the concentration of
United States Pharmacopeia (USP) (2) for the same scattering particles becomes high, then multiple
exercise: scattering events become possible.

50 Journal of GXP Compliance

 Scott Sutton

TABLE: McFarland turbidity standards. only for those microorganisms similar to E. coli. In
McFarland Scale CFU (x106/mL) 1% BaCl2/1% H2SO4 (mL) addition, the values are not in the appropriate range
for the AET inoculum and further dilutions may be
0.5 <300 0.05/9.95
required.
1 300 0.1/9.9
2 600 0.2/9.8 Spectrophotometer
3 900 0.3/9.7 The spectrophotometer measures turbidity directly.
4 1200 0.4/9.6 The best case (i.e., most sensitive) would be to have a
narrow slit and a small detector so that only the light
5 1500 0.5/9.5
scattered in the forward direction would be seen by
6 1800 0.6/9.4 the detector. This instrument would give larger ap-
7 2100 0.7/9.3 parent absorption readings than other instruments
8 2400 0.8/9.2 (see Figure).
9 2700 0.9/9.1 As should be obvious, each spectrophotometer
used must be independently calibrated for use in
10 3000 1.0/9.0
estimating microbial concentrations. Not only is the
apparent absorption affected by the width of the in-
METHODS
H strument’s slit, the condition of the filter, and the size
conditio of the
and condition he de or, but aalso each
detector, ach time thehe
McFarl
McFarland Turbidity
urbidi Standards
ndards lamp p is changed
chan the cal
calibration n needs to be repeated ted
McFarland
M cFa
F l standards
tandard can be used to visually
sually approx-
prox-
x fferent bulbs
as different b v
may vary to output.
in total utput.
imate the concentration
ima oncentrati of cells in a suspension.
pen The
T he correlation
The correl abs
of absorption n to
t dryy weight is
McFarland Scale reprerepresents
nt specific
peci c concentrations
con ntr ons ery ggood
very od fo
for d lute ssuspensions
dilute pe io of bacteria (5), and
of CFU/mL and iss designed
des gned to
o be used
sed for
f r estimating
estim
matin his relationship
this re io h seem seemss to h ld rregardless of cell size
hold
concentrations of gram negative bacteria such as E. (although the relationship of absorption to CFU does
oli N
coli. Note
ote that
th
hat tthis
hiis esti
estimate
st mate be
becomes
ec meess uncertain
un
ncerttaiin with
with not)
noot). However,
not). H
Hoowev
w ve , in
in more
more cconcentrated
oncentrate
tr d suspensions,
susspensionnss this
n th s
organisms outside d the h normal
or l usage,
ge as ddifferent
ff rre io (absorption
correlation (b to d h ) no llonger h
dry weight) ld
holds.
species of bacteria differ in size and mass, as do yeast The linear range of absorption to estimated CFU is of
and mold. Use of this method for various organisms limited scope. For this reason, the calibration study
would require calibration and validation. must demonstrate the linear range of the absorbance
McFarland standards are generally labeled 0.5 versus CFU values and the relevant values.
through 10 and filled with suspensions of barium
salts. Latex bead suspensions are also available, Procedure
which extend the shelf life of the material. The As there are a variety of different instruments, there
standards may be made in the lab by preparing a cannot be one single procedure. In general, the spec-
1% solution of anhydrous BaCl2 and a 1% solution trophotometer can be set at a wavelength of 420–550
of H2SO4 by mixing them in the proportions listed nm. This wavelength must be standardized.
in the Table. They should be stored in the dark, in It is important to have the cells in known physi-
a tightly sealed container at 20-25oC, and should be ological state of growth. That is to say, as the cell size
stable for approximately six months (4). varies with phase of growth (i.e., lag, log, stationery),
The advantage of the use of these standards is the approximate relationship between absorbance
that no incubation time or equipment is needed to and CFU will also vary. A recommended practice
estimate bacterial numbers. The disadvantage is might be to pass a single well-isolated colony twice
that there is some subjectivity involved in interpret- on overnight cultures from the refrigerated stock,
ing the turbidity, and that the numbers are valid and harvest the rapidly growing culture from the

Summer 2011 Volume 15 Number 3 51

MICROBIOLO GY TO PIC S

Figure:
Spectrophotometer components.

Scattered Light

Light Detector
Source Filter S lit Microorganism
Suspension

second passage. This also will serve to minimize a CFU/mL after the plate counts are available and use
source of variability for the AET (6). values in the linear range of this graph. This linear
A second
e source of concern might be the cuvette range may require a wide dilution series to identify,
used fo
for the measurement—care must be taken and similar raw dilutions (from the nutrient broth)
to maintain
mai the correct
co orientation
ientati of the cuvette,
cuve mightt not yie
yield samples
mples in the linear range.
he linea nge
and protect it fro
d to p from damage
mage that could affect the
uld affec As there arare several
veral factors
fa that
ha can affect this
passage
p assag of light.
ght. Finally,
Fina it is necessary
neces to blank
bla curvee (e.g., q
qualityy of lam
lamp output,pu size ze of slit,
the spectrophotometer
photomete (i.e., e adjust
adju the absorbance
bso ce condition
c ition of filter, condition
condi off d
detector,
tor microorgan-
micro n
reading to zero)
ro) using
sing a stan
standard,
ard, either
ither water
ater orr the ism
i m characteristic,
h ract ri ic, etc.), this ccalibration should be
et , th
suspending fluid,
fluid and maintain
maint in this
th s practice.
prac ice. confirmed
c nfirm d when hen the conditions
condi ons of the assay change.
In fact, all spectrophotometer estimates should be
Cal bbrrattion
Calibration treated
treat d as just
reatted ju
us ththat
ha aand
hat nd cconfirmed
onf rmmed by platee count nt in
coount
It must b be stressed
d that
h this
h calibration
l b tio should
h ld bbe parallel
rall l with
w h theh test.
done for all organisms. The size of the organism, any
associated pigments, the preparation of the suspen- TROUBLESHOOTING
sion, and other factors all influence the readings. There are several potential areas where the method
This calibration study should also be rechecked after might provide erroneous results.
changing the bulb on the light source and should be
reevaluated throughout the life of the light bulb. Mix-n-Match
The calibration itself is simple to perform. Prepare The first of these potential mistakes involves in-
a concentrated solution of the organism, grown discriminate use of information from the Internet.
under the conditions that will be used for the test. There has been some discussion of this topic on the
Make a series of dilutions to cover the range of Pharmaceutical Microbiology Forum Email List (PM-
absorption measurements of interest; five to eight FList) (7), and it is clear that many workers fail to
dilutions are recommended. Immediately take the understand how instrument-dependent these optical
spectrophotometer readings in sequence, and then density curves are. Each one must be checked for
take a confirmatory reading of the first in series to the specific organism and instrument. Just be-
confirm that no growth has occurred. The dilutions cause a reading of 0.14 gave an acceptable number
are then immediately plated for viable count (serial of CFU/mL in one facility for a specific species of
dilution of the suspensions will be necessary). Graph microorganism does not mean that it will be accept-
the relationship between the absorbance and the able anywhere else or for other microorganisms.

52 Journal of GXP Compliance

 Scott Sutton

Where Did it Come From? Despite the inherent inaccuracy of the method, if the
The QC test is concerned about number of CFU. procedure is adequately controlled and calibrated, the
Optical density (OD) is an indirect measure of cell estimation of microbial numbers by optical density (ei-
mass. Different species, even if all other aspects of ther by McFarland Standards or spectrophotometrical-
the system are identical, will have different optical ly) is sufficiently accurate for use in preparing inocula
density readings for similar CFU/mL concentration. for QC testing. This method offers the advantages of
You cannot use similar OD numbers for different being rapid, low cost, and non-destructive.
species without confirmation. Some cells (e.g., C.
albicans) might have CFU/mL ten-fold lower than REFERENCES
other commonly used microorganisms at the same 1. EP, “5.1.3 Efficacy of Antimicrobial Preservation,” European
OD reading. Pharmacopoeia 6.6 pp, 5129-5130.
2. USP, <51> Antimicrobial Effectiveness Testing, United States
Is it Clumping? Pharmacopeia, 2011.
This method obviously assumes a fairly homoge- 3. Koch, AL. “Growth Measurement,” Methods for General and
neous suspension of cells. This is sometimes not the Molecular Bacteriology Gerhardt, P et al. (ed) American Society
case. For example, B. subtilis is a naturally compe- for Microbiology, Washington, DC. p. 248-277, 1994.
tent (able
a to transfer DNA among cells) microorgan- 4. Smibert, RM and NR Kreig, “Phenotypic Characterization”
ism an
and in the competent state aggregates in media. Section 25.4.9, Methods for General and Molecular Bacteriol-
This
is iis not conducive
du to good O OD readings.
eadings. A.
A gy, Gerhardt,
ogy, har P et al. (ed) American erican Society
Soc forr Microbiology,
M gy,
brasiliensis
bra
asilie spores
ores aalso tend
nd to clump,
clu , or even to get Washington,
Washin ton DC. p. 607-654,
607- 1994.
94.
hung
h ung up in hyphael matss during preparation.aratio Some-
ome- 5. Koch, AL., “Turbidity dity Measurements
Mea nts of Bacterial
cterial Cultures
Cul
times the inclusion
tim nclusion oof small amounts off polysorbate
mall am p rb in
n Some Available
Av Commercial Instruments,”
Comm str ts ” Anal Biochem
Bio
into the suspending
pen ing buffer
b ffer and
nd filtration
fi ratio through
hr ugh 38:252-259,
8 52-25 , 1970. 70.
sterile glass wool
woo is helpful
h lpful too generating
generating a more
m re 6. Gilbert,
Gil t, P. ett al. Inocula
Inocu a for Antimicrobial
A timic Sensitivity Testing: a
homogenous suspension of this organism. Critical Review. J Antimicrob Chemother. 20:147-154,
20:147 154, 1987.
7. www.microbiologyforum.org.
w www.m
.m
miccr iolog
io ogyforum m.oorgg. GXP
XP
OD
DD Drift
if
The lamp in the spectrophotometer will age. It is ARTICLE ACRONYM LISTING
useful (and recommended) that a log book of daily AET Antimicrobial Efficacy Test
OD readings against the measured CFU be kept to CFU Colony Forming Units
assist in determining when the calibration of the OD Optical Density
curve should be repeated or when the lamp should QC Quality Control
be replaced. USP United States Pharmacopeia

CONCLUSIONS ABOUT THE AUTHOR

The use of optical density to estimate CFU in a sus-
pension is possible if basic precautions are taken. It
is important to control the following:
t
The physiological state of the organism
t
The species of the organisms
t
The nature and condition of the equipment.
Scott Sutton, Ph.D., is owner and operator of The Microbiology
Network (www.microbiol.org), providing a network of international
experts for consultation, quality assurance training, and expert
witness services to the regulated industries. Dr. Sutton also operates
the PMFList email discussion group and regularly tweets on micro-
biological topics (@MicrobiologyNet). He may be reached by e-mail
at scott.sutton@microbiol.org.

Summer 2011 Volume 15 Number 3 53