C H A P T E R F O U R

Preparing a Purification
Summary Table
Richard R. Burgess

Contents
1. Introduction 29
2. The Importance of Footnotes 32
3. The Value of an SDS–Polyacrylamide Gel Analysis on
Main Protein Fractions 32
4. Some Common Mistakes and Problems 32

Abstract
Once a protein purification scheme has been developed, the purification, char-
acterization, and use/structure of a target protein are usually published. It is
highly desirable to present the major steps in the purification and the
corresponding features of the protein at each step summarized in the form of
a purification summary table. In considering whether to repeat a published
protein purification, a reader needs this information to evaluate the purification,
and to decide if it is worth following or if it needs major modifications. In this
chapter, I discuss the main characteristics of a useful purification summary
table and point out common mistakes and problems I see in many such tables.

1. Introduction
As an executive editor and editor-in-chief of the journal, Protein
Expression and Purification, I have reviewed on the order of 100 protein
purification papers a year for over 18 years. It is remarkable how many
manuscripts I receive where there is either no purification summary or one
that is severely lacking in necessary information and accuracy. The essentials
of a reasonable purification table are illustrated by the example below.
Suppose one set out to purify an enzyme from the bacterium E. coli
starting with 10 g of wet weight cell pellet from a 4-l culture (10 g of wet
weight cells typically would contain about 2 g of dry weight and about

McArdle Laboratory for Cancer Research, University of Wisconsin-Madison, Madison, Wisconsin, USA

Methods in Enzymology, Volume 463 # 2009 Elsevier Inc.

ISSN 0076-6879, DOI: 10.1016/S0076-6879(09)63004-4 All rights reserved.

29
30 Richard R. Burgess

1200 mg of total protein). The cells are lysed by sonication to give a crude
lysate and the debris is removed by centrifugation to give a crude extract.
A 45–50% saturated ammonium sulfate cut was prepared. The 50%
saturated ammonium sulfate pellet was dissolved in buffer and diluted to
low salt and applied to a DEAE anion exchange column. The column was
washed at low salt and then eluted with a linear salt gradient from 0.1 to
0.6 M NaCl, the peak of activity eluting at about 0.25 M NaCl. The peak
was pooled and applied to a Sephacryl S-300 gel filtration column and
eluted with a buffer at constant salt (isocratically). The fractions of peak
activity were pooled and shown by SDS–PAGE with Coomassie blue
staining to be a single band. The specific activity of the final material is
the same as that of a known pure reference sample. The main fractions were
all assayed for enzyme activity and protein determinations were carried out.
The resulting data are given in Table 4.1.
A purification summary table should allow a reader to evaluate easily the
procedure and readily detect particularly effective and ineffective purifica-
tion steps. It should be easy to see if large losses occurred at a particular step.
A suitable table will contain the following columns:
1. Major steps in the purification. These typically include steps like:
Crude lysate (the result of cell or tissue disruption). This step is often
omitted since assays may be difficult but it is useful and even essential
when much of the expressed recombinant target protein is in an
insoluble inclusion body.
Crude extract (the lysate after any insoluble material has been removed
by centrifugation)

Table 4.1 A typical purification summary table

Total Total Specific

protein activity activity Yield Purity
Step (mg)b (units)c (units/mg) (%) (%)
Crude lysatea 1200 120 0.10 100 0.8
Crude extract 1000 110 0.11 92 0.9
Ammonium 180 75 0.42 62 3.4
sulfate
45–55% cutd
DEAE column 24 60 2.5 50 20
(pooled peak)
Sephacryl column 3.6 46 12.5 38 100
(pooled peak)
a
From 10 g of wet weight E. coli cell pellet (from 4 l of bacterial culture).
b
Protein concentration determined by Bradford assay using BSA as a standard protein.
c
Enzyme activity measured as described in the methods section.
d
Crude extract material that is soluble at 45% but precipitates at 55% saturated ammonium sulfate.
Preparing a Purification Summary Table 31

Ammonium sulfate cut

Pooled peak from an ion exchange column
Pooled peak from a gel filtration column
Pooled peak from an affinity column
Concentrated and dialyzed final product
Solubilized inclusion bodies. This step and the following two are often
used in the case where an expressed recombinant protein is produced
as insoluble inclusion bodies; see Chapter 17.
Washed inclusion bodies
Refolded, centrifuged, and concentrated material
2. Amount of total protein (mg). This is usually determined by a standard
protein assay. Most commonly these days a Bradford dye-binding assay
or a bicinchoninic acid (BCA) assay is used (see Chapter 8). It is
important to indicate in the methods section what protein is used as
the protein standard (typically BSA). Once the protein is purified it can
also sometimes be quantified by measuring its absorbance at 280 nm and
the use of an appropriate molar extinction coefficient.
3. Amount of target protein or total activity (mg or units). If there is a suitable
enzyme assay for the target protein, it should be carried out on material
from each major step. If the protein is not an enzyme or there is no
quantitative assay, and if the protein is visible on a Coomassie blue
stained SDS–PAGE, then often the stained gel is scanned and the
amount of protein in the target protein band is determined. In other
words, purity is determined and multiplied by the total protein to give an
estimate of the total target protein.
4. Specific activity (units/mg). If enzyme activity assays are possible, then the
total activity (units) is divided by the total protein (mg) to give specific
activity as units/mg.
5. Overall yield (%). The yield at a step in the procedure is the amount of
target (either total target protein or total activity) at that step divided by
the amount of target in the first step (defined as 100%).
6. Purity of target protein (%). Purity is often determined by scanning a
stained SDS–PAGE and measuring the amount of the stain associated
with the target band as a fraction of the stain associated with all the bands
on the gel. If one has a reliable assay, then if the final material is pure, its
specific activity can be used to define purity. For example, if an earlier
step has a specific activity 10% of the final pure material, then the purity
at that step would be 10%.
7. Relative or fold purification. This is not essential since it can be calculated
from the other values above, but it is often useful. This is merely setting
the initial purity at a value of one and then giving the purity at each step
relative to that of the first step. For example, in Table 4.1, the final
step represents an overall fold purification of 125.
32 Richard R. Burgess

2. The Importance of Footnotes

Every protein and purification is different and footnotes are needed
to help the reader understand what has been done. There should be a
footnote that indicates the amount of raw material used in the preparation
being summarized. For recombinant protein expressed in bacteria, for
example, one should always give the number of grams of wet weight
cell pellet used in the preparation. It is also useful to know the volume of
the bacterial culture used, but that in itself is not enough since, depending
on the growth media and conditions, the yield of wet weight cell pellet
can range from 1 to 80 g/L. Another useful footnote indicates how the
protein amount was determined (e.g., sometimes a Bradford assay is used
on the early steps, but absorbance and extinction coefficient is used on the
final product).

3. The Value of an SDS–Polyacrylamide Gel

Analysis on Main Protein Fractions
I find that an SDS–polyacrylamide gel image is a very valuable com-
plement to the purification summary table. If the same samples that repre-
sent the various steps in the purification table are also shown in a gel photo,
then it is particularly easy to see the progress of the various fractionation
steps toward production of a purified final target protein. The most useful
gels are ones on which an equal proportion of the material at each step is
loaded on the gel lanes.

4. Some Common Mistakes and Problems

1. Use too many significant figures. This is one of my pet peeves. I suspect that
the concept of significant figures is no longer taught, because I find that a
good 75% of the purification papers I review give protein amounts like
235.052 mg and yields like 46.72%. Just because a calculator or com-
puter can divide two numbers and give one the result to eight figures
does not mean that value is what one should enter into the table. Just
remember that most protein quantification methods or enzyme assays are
not accurate to better than 5–10%. When one writes 23.47 mg, one
implies that it is not 23.46 or 23.48, but 23.47. In other words, one is
implying that it is accurate to one part in over 2000 when it is not even
Preparing a Purification Summary Table 33

clear that it is accurate to one part in 20 (is it 22, 23, or 24?). No numbers
should be given to more than three significant figures and in general
most percentages can be given to two significant figures. Also, remember
that any number resulting from the division of two numbers each
accurate to
10% will only be accurate to
20%.
2. Calculate values erroneously. Remarkably many tables contain simple
arithmetic errors. All numbers should be checked and rechecked before
submission of a manuscript.
3. Use step yields instead of overall yields. A step yield is the yield from a single
step in the purification procedure, that is, the amount of target protein or
activity after that step compared to that in the previous step of the
procedure. A series of four fractionation steps might all give 60% step
yields, but the overall yield is (0.6)4 ¼ 0.6  0.6  0.6  0.6 ¼ 0.13 or
13%. Overall yield is more useful. A procedure that gives an overall yield
of a few percent may be due to a very lengthy and difficult purification of
a rare or unstable protein, but more likely it indicates that the procedure
has not been optimized very well.
4. Calculate yield as yield of total protein. Often I see a table in which the yield
given is yield of total protein, rather than target protein. This is relatively
useless information. Yield is always recovery of total target protein or
activity.
5. Write up and try to publish the first purification that gives any product. There is
a tendency for readers, especially inexperienced readers, to assume that a
published purification is the result of many cycles of improvement and
optimization, and represents the best way to purify the protein. This is
very often not true. Many times, it is just a series of steps, often chosen
arbitrarily that happens to result in some product. One should look very
carefully at a purification to assess if it is a procedure that is worth trying
to follow if one wants to purify some of the target protein. This is why a
proper purification summary table is so valuable. If huge losses occur at
a particular fractionation step, if the overall yield is very low, if the final
purity is not high, or if similar fractionation steps are used several times,
then perhaps the procedure should merely be used as a beginning point
in designing a better, more effective purification.
6. What to do when purifying a protein where a fusion partner is cleaved off during
the procedure? Very often a recombinant protein is expressed as a fusion
with another protein or tag that aids in its folding or purification
(see Chapter 16). Since most often the desired final product is the target
protein without the fusion partner or tag, especially for structural studies,
the fusion partner must be cleaved off by one of several specific proteases.
Let us say that the target is 20 kDa and the fusion partner is 40 kDa, so the
fusion protein expressed is 60 kDa. At the step where the fusion protein
is cleaved, the yield of target protein seems to decrease by 67% even if
all of it is recovered. How is this indicated in the summary table?
34 Richard R. Burgess

I suggest that the column on target protein amount contain two numbers;
the mg of fusion protein and the calculated mg of the final target protein in
parenthesis. That way at the step where the cleavage has occurred, one can
continue giving just the mg of the cleaved product and the theoretical
amount of cleaved product in the first step can be used to calculate the
overall yield.