Klipp E., Herwig R., Kowald A., Wierling C., Lehrach H. Systems Biology in Practice: Concepts, Implementation and Application

Подождите немного. Документ загружается.

hybrids by complementary base pairing. Instead, it is based on the specific interac-

tion between antibody and antigen.

4.1.4.1 Southern Blotting

The technique of Southern blotting was invented in 1975 and is used to analyze

complex DNA mixtures (Southern 1975). Normally, the target DNA is digested by re-

striction enzymes and separated by gel electrophoresis. If the DNA is of genomic ori-

gin, the number of resulting fragments will be so large that no individual bands are

visible. If we are interested in a certain gene and know its sequence, small DNA frag-

ments can be synthesized, which are complementary to the gene. However, before

the actual hybridization step, the digested DNA has to be transferred from the gel

onto the surface of a nitrocellulose or nylon membrane so that the DNA molecules

are accessible for the hybridization. This is achieved with the help of a blotting appa-

ratus, as shown in Fig. 4.5. Originally the transfer was achieved by placing the nitro-

cellulose filter between the gel and a stack of blotting paper. Capillary forces lead to a

flow of water and DNA fragments from the gel onto the blotting paper. In this way,

the DNA gets trapped by a nitrocellulose or nylon filter. Nowadays, blotting ma-

chines that transfer the DNA by applying a voltage across the gel and membrane are

used. Once the DNA is blotted, the membrane is placed into a plastic bag and incu-

bated for several hours with a solution containing the labeled DNA probe. In the

case of a radioactive label, the membrane is placed against an X-ray film. The radio-

active DNA fragments expose the film and form black bands that indicate the loca-

tion of the target DNA.

With this technique, not only can the presence of the gene of interest be tested,

but also modifications of the gene structure (in the case of a mutation) can be stu-

died. By performing several Southern blots with DNA probes that correspond to dif-

ferent regions of the gene, modifications such as deletions and insertions can be de-

tected. Point mutations, however, cannot be identified with Southern blotting.

120

4 Experimental Techniques in a Nutshell

Fig. 4.5 Elementary steps required for South-

ern blotting. Following gel electrophoresis, the

DNA fragments are treated with an alkaline solu-

tion to make them single-stranded. The nitrocel-

lulose or nylon membrane is sandwiched be-

tween the gel and a stack of blotting paper and

the DNA is transferred onto the membrane

through capillary forces. Finally, the membrane

is incubated with the labeled DNA probe (here,

radioactive labeling) and the bands are then vi-

sualized by X-ray film exposure. (Courtesy of Dr.

P. Weingarten, Protagen AG.)

4.1.4.2 Northern Blotting

Northern blotting is very similar to Southern blotting. The only difference is that

mRNA, not DNA, is used for blotting. Although the experimental technique is very

similar, Northern blotting can be used to answer different questions than Southern

blotting. Even though mRNA is only an intermediate product on the way from the

gene to the protein, it is normally a reasonable assumption that the amount of

mRNA is correlated to the amount of the corresponding protein in the cell. Northern

hybridization is therefore used not only to verify the existence of a specific mRNA

but also to estimate the amount of the corresponding protein via the amount of

mRNA. Since the expression profile of genes varies among tissues, Northern blot-

ting gives different results for different organs, in contrast to Southern blotting,

which is based on genomic DNA.

Surprisingly, the technique did not get its name from the researcher who invented

it; rather, it was named as a humorous allusion to the double meaning of the name

of E. M. Southern, who gave his name to the Southern blot.

4.1.4.3 Western Blotting

So far we have seen techniques for blotting different types of nucleic acids. A similar

type of technique exists for proteins, the Western blot. As with the Northern blot, its

name indicates only that it is a special type of blotting technique. Depending on the

problem at hand, 1D or 2D protein gels can be used for blotting. It is more difficult

to obtain specific probes for proteins than for nucleic acids. Apart from special cases,

antibodies that are directed against the desired protein are used. If no commercial

antibodies are available for the protein of interest, they have to be produced, e.g., by

immunizing rabbits.

Once the protein is transferred to the nitrocellulose membrane, it is incubated

with the primary antibody. The primary antibody recognizes the protein and forms

an antibody-protein complex with the protein of interest. In a further step, the mem-

brane is incubated with the so-called secondary antibody, which is an antibody

against the primary antibody. If the primary antibody was obtained by immunizing a

rabbit, the secondary antibody could be a goat-anti-rabbit antibody. This is an anti-

body from a goat that recognizes all rabbit antibodies. The secondary antibody is che-

mically linked to an enzyme, such as horseradish peroxidase, that catalyzes a chemi-

luminescence reaction, and exposure of an X-ray film finally produces bands, indi-

cating the location of the protein-antibody complex. The intensity of the band is pro-

portional to the amount of protein. Unlike Northern blotting, Western blotting gives

a direct estimate of the protein content. The secondary antibody serves as signal am-

plification step. This is the reason that the enzyme is not linked directly to the pri-

mary antibody.

4.1.4.4 In situ Hybridization

The described blotting and hybridization techniques are applied to mixtures of nu-

cleic acids or proteins that have been extracted from cells or tissues. During this pro-

cess, all information about the spatial location is lost. In situ hybridization solves this

problem by applying DNA probes directly to cells or tissue slices.

121

4.1 Elementary Techniques

One common application is the location of specific genes on chromosomes. For

this purpose metaphase chromosomes, which have been exposed to a high pH to se-

parate the double strands, are incubated with labeled DNA probes. This makes it

possible to directly see where and how many copies of the gene are located on the

chromosome. If the label is a fluorescent dye, the technique is called FISH (fluores-

cent in situ hybridization). Not only chromosomes but also slices of whole tissues

and organisms can be hybridized to DNA probes. This can be used to study the spa-

tial and temporal expression pattern of genes by using probes specific to certain

mRNAs. This method is often used to study gene expression patterns during embry-

ogenesis. Finally, immunostaining uses antibodies to localize proteins within cells.

Knowledge about the subcellular localization often helps one to better understand

the functioning or lack of functioning of the studied proteins.

4.1.5

Further Protein Separation Techniques

4.1.5.1 Centrifugation

One of the oldest techniques for the separation of cell components is centrifugation.

This technique fractionates molecules (and larger objects) according to a combina-

tion of size and shape. However, in general it is true that the larger the object, the

faster it moves to the bottom. A typical low-speed centrifugation (around 1000-fold

gravitational acceleration, g) collects cell fragments and nuclei in the pellet, at med-

ium speeds (50,000 g) cell organelles and ribosomes are collected, and at ultrahigh

speeds (up to 500,000 g) even typical enzymes end up in the pellet. Of course, such a

separation is not quantitative, but repeating the centrifugation after re-dissolving the

pellet improves the purity. The sedimentation rate for macromolecules is measured

in Svedberg units, S, after Theodor Svedberg, who invented ultracentrifugation in

1925. S depends on the mass (m) and density of the particle (r

par

), as well as on the

density (r

sol

) and friction ( f ) of the medium. These properties control the velocity

(v) of the particle (o

= angular velocity, r = distance from center of rotation).

S 



m 1  r

sol

par



(4-2)

The ribosomal subunits, for instance, got their name from their sedimentation

coefficient (40S subunit and 60S subunit). Because the friction is controlled not only

by the size of the particle but also by its shape, S values are not additive. The com-

plete ribosome (40S plus 60S) sediments at 80S and not at 100 S.

From Eq. (2) it also follows that the sedimentation rate is zero if the densities of

the particle and the surrounding medium are identical. This is the basis for the equi-

librium centrifugation method, in which the medium forms a stable density gradi-

ent (caused by the gravitational forces) such that the density of the studied particles

lies within the density range of the gradient. If the centrifugation is run long en-

ough, the particles move to the position where the densities of the medium and the

particle are identical and form stable bands there. Thus, equilibrium centrifugation

122

4 Experimental Techniques in a Nutshell

separates the molecules by density, independent of their size. The necessary density

gradients are typically formed with saccharose or cesium chloride (CsCl).

4.1.5.2 Column Chromatography

Other classical separation techniques include the different forms of column chroma-

tography (Fig. 4.6). A column (often made of glass, a few centimeters wide and a few

dozen centimeters tall) is filled with a solid carrier material and the protein mixture

is placed on top of it. Then a buffer is slowly washed through the column and takes

the protein mixture along with it. However, different proteins are held back to a dif-

ferent degree by the column material and arrive at different times at the bottom of

the column. The eluate can be fractionated and tested for the presence of the desired

protein. The ratio of desired protein to total protein is a measure of the purity, which

can additionally be checked on normal SDS gels. Different column materials are

available and the success of a separation often depends critically on the choice of ma-

terial.

The material for ion exchange chromatography (Fig. 4.6a) contains negatively

(carboxymethyl or phosphor cellulose) or positively (diethylaminoethyl cellulose)

charged beads that can be used to separate hydrophilic proteins according to charge.

The binding between the proteins and the beads is also controlled by the salt concen-

tration and pH of the elution buffer. Some proteins possess hydrophobic surfaces that

can be used to separate proteins by hydrophobic interaction chromatography (Fig.

4.6b). For this purpose short aliphatic side chains are attached to the surface of the

column material. Gel filtration chromatography (Fig. 4.6c) is also often used to sepa-

rate proteins according to size. The beads contain a range of pores and channels that

123

4.1 Elementary Techniques

Fig. 4.6 In column chromatography a protein

mixture is placed on top of the column material

and then eluted with buffer. Different types of

material are available to separate the proteins

according to (a) charge, (b) hydrophobicity,

cule. Often, different chromatographic steps

have to be used successively to purify the desired

protein to homogeneity (lanes 1–3 of the gel).

(Courtesy of Dr. P. Weingarten, Protagen AG.)

allow small molecules to enter, which increases their retention times. This allows

not only their separation by size but also estimation of the absolute size of a protein

or protein complex. A more recent development is affinity chromatography (Fig.

4.6d), which makes use of highly specific interactions between a protein and the col-

umn material. This can be achieved, for instance, by chemically linking antibodies

to the column material. For affinity chromatography the column is loaded with the

protein mixture in the first step. The proteins of interest bind, while the other pro-

teins pass through the column. In the second step, the elution process is started by

using a high-salt or high-pH buffer that frees the bound protein from the column.

A major improvement regarding speed and separating power was achieved

through the development of high performance liquid chromatography (HPLC). The

columns are much smaller and the carrier material is packed more densely and

homogenously. To achieve reasonable buffer flow rates, very high pressures (up to

several hundred atmospheres) are needed. Therefore, special pumps are used and

the column housing is made of steel.

The enrichment factor of a single chromatographic step is normally between 10-

and 20-fold. However, since many proteins represent only a tiny fraction of the total

protein content of a cell, different chromatographic columns often have to be used

consecutively. A notable exception is affinity chromatography, which can achieve en-

richments up to 10

in a single step. In combination with modern recombinant

DNA techniques, affinity chromatography has many applications. Recombinant pro-

teins can be designed to contain special short sequences of amino acids that do not

compromise the functionality of the protein but can serve as molecular tags. A short

stretch of histidine residues is called a His-tag and is specifically recognized by a

nickel surface or special His antibodies.

4.2

Advanced Techniques

4.2.1

PCR

The polymerase chain reaction (PCR) was invented in the mid-1980s and allows the

billion-fold amplification of specific DNA fragments (typically up to 10 kbp) directly

from genomic DNA (Saiki et al. 1985). A pair consisting of short oligonucleotides

(15–25 bp), the primers, is synthesized chemically such that they are complementary

to an area upstream and downstream of the DNA of interest. DNA is made single-

stranded by heating, and during the cooling phase primers are added to the mixture,

which then hybridize to the single-stranded DNA (Fig. 4.7). In the next step a DNA

polymerase extends the primers, doubling the copy number of the desired DNA frag-

ment. This concludes one PCR cycle. Each additional cycle (denaturation, annealing,

and amplification) doubles the existing amount of DNA that is located between the

primer pair. Thus, 30 cycles correspond to a 2

= ~10

-fold amplification step (30–40

cycles are typically used). A special heat-stable DNA polymerase is used that remains

124

4 Experimental Techniques in a Nutshell

active during the heating step and thus obviates the need to add fresh enzyme after

each cycle. Today, the different steps of the PCR reaction don’t have to be performed

manually. Small, automated PCR machines, also called thermal cyclers, can perform

dozens of PCR reactions in parallel, and the typical 30 cycles are finished in roughly

half an hour.

In the last few years, sequence information for many complete genomes has be-

come available, which allows one to use PCR to clone genes directly from genomic

DNA without the use of DNA libraries. PCR has revolutionized modern molecular

genetics and has many applications. For instance, by combining reverse transcrip-

tase (which makes a DNA copy from RNA) with PCR, it is also possible to clone

mRNA with this technique. The extreme sensitivity of PCR also makes it the method

of choice for forensic studies. Highly variable tandem repeats are amplified and used

to determine whether genetic material that comes from hair follicle cells, saliva, or

blood stains belongs to a certain suspect. This is possible because different indivi-

duals have tandem repeats of different length, which results in amplified DNA frag-

ments of different length. By looking at a large number of different tandem repeats,

the chances of a false positive result can be made arbitrarily small. This principle is

also the basis of paternity tests.

Another type of application is the use of multiple, or degenerate, primers. This

can be useful when a consensus sequence is known and the aim is to amplify genes

that contain this conserved sequence. Degenerate primers are also used if a stretch

125

4.2 Advanced Techniques

Fig. 4.7 The polymerase chain reaction. (a)

After double-stranded DNA is heated to obtain

single strands, (b) short DNA primers (P1 and

P2) are hybridized to a region that is upstream

or downstream, respectively, of the DNA of inter-

est. (c) A DNA polymerase is then used to ex-

tend the primers. (d) Then the cycle is repeated

by melting the DNA and annealing new primers.

For clarity the primers that were used in the first

cycle are not drawn in bold in the second cycle.

of the amino acid sequence of a new protein is known (e.g., through protein sequen-

cing) and the corresponding gene has to be found. Because of the degeneracy of the

genetic code, several nucleotide sequences can code for a given amino acid sequence.

The use of degenerate PCR can solve this problem.

An important, more recent, development is the technique of real-time PCR, which

is especially suited to quantify the amount of template that was initially present.

Classical PCR is normally unable to give quantitative results because of saturation

problems during the later cycles. Real-time PCR circumvents these problems by

using fluorescent dyes that either intercalate in double-stranded DNA or are bound

to sequence-specific oligonucleotides (TaqMan probe). The increase of fluorescence

with time is used, in real time, as an indicator of product generation during each

PCR cycle.

4.2.2

DNA and Protein Chips

4.2.2.1 DNA Chips

DNA chips, also called DNA microarrays, are a recently developed method for the

high-throughput analysis of gene expression (DeRisi et al. 1997). Instead of looking

at the expression of a single gene, microarrays allow one to monitor the expression

of several thousand genes in a single experiment, resulting in a global picture of the

cellular activity. This idea is also the basis of the systems biological modeling ap-

proach, and chip data can therefore be a very important source of data.

A microarray experiment starts with the construction of the chip (Fig. 4.8) from a

DNA library (see Section 4.1.2). The inserts of individual clones are amplified by

PCR (a single primer pair that is specific for the vector used to construct the library

can be used) and spotted in a regular pattern on a glass slide or nylon membrane.

These steps are normally automated and performed by robots. Then, total mRNA is

extracted from two samples that we would like to compare (e.g., yeast cells before

and after osmotic shock). Using reverse transcriptase, the mRNA is transcribed into

126

4 Experimental Techniques in a Nutshell

Fig. 4.8 DNA chips are used to study the ex-

pression level of thousands of genes in parallel.

Individual genes from a DNA library are ampli-

fied by PCR and spotted onto glass slides. cDNA

is prepared from the cells or tissues that are to

be compared and is labeled with red and green

fluorescent dyes. The color reveals, after hybridi-

zation, the relative mRNA amounts in the two

sources. Finally, the expression data are clus-

tered.

cDNA and labeled with a fluorescent dye. It is important that the dyes emit light at

different wavelengths; red and green dyes are commonly used. The cDNAs are now

incubated with the chip where they hybridize to the spot that contains the comple-

mentary DNA fragment. After washing, the ratio of the fluorescence intensities for

red and green are measured and displayed as false color pictures (Fig. 4.8, middle).

Red or green spots indicate a large excess of mRNA from one or the other sample,

while yellow spots show that the amount of this specific mRNA was roughly equal

in both samples. Very low amounts of both mRNA samples result in dark spots.

These ratios can, of course, also be quantified numerically and used for further cal-

culations, such as the generation of a clustergram. For this analysis, a complete link-

age cluster of the genes that were spotted on the chip is generated and the mRNA ra-

tio (represented as color) is displayed in this order (Fig. 4.8, right). This helps to test

whether groups of related genes (maybe all involved in fatty acid synthesis) also

show a similar expression pattern.

Oligonucleotide chips, a variant of DNA chips, are based on an alternative experi-

mental design. Instead of spotting cDNAs, short oligonucleotides (25–50mer) are

used. Approximately a dozen different and specific oligonucleotides are used per

gene. In this case only one probe of mRNA is hybridized per chip, and the ratio of

fluorescence intensity of different chips is used to estimate the relative abundance of

each mRNA. Chips from the companies Affymetrix or Agilent are most commonly

used for this approach.

During the past several years, this technique has been used to study such diverse

problems as the effects of caloric restriction and aging in mice (Lee et al. 1999), in-

fluence of environmental changes on yeast (Causton et al. 2001), and the conse-

quences of serum withdrawal on human fibroblasts (Eisen et al. 1998).

4.2.2.2 Protein Chips

Despite the large success of DNA chips, it is clear that the function of the genes is

realized through the proteins and not by the mRNAs. Therefore, efforts are under

way to construct chips that consist of spotted proteins instead of DNA. In this case

the starting point is an expression library for obtaining large quantities of the recom-

binant proteins. The proteins are spotted and fixed on a glass slide and can then be

incubated with interaction partners. These could be (1) other proteins to study pro-

tein complexes, (2) antibodies to quantify the spotted proteins or to identify the re-

cognized antigens, (3) DNA to find DNA-binding proteins, or (4) drugs to identify

compounds of pharmaceutical interest (Cahill and Nordhoff 2003).

However, the generation of protein chips poses more problems than DNA chips

because proteins are not as uniform as DNA. One challenge is to express sufficient

amounts of recombinant proteins in a high-throughput approach. Another problem

is that the optimal conditions (temperature, ionic strength, etc.) for interaction with

the reaction partner are not identical for different proteins. But the field is making

rapid progress, and protein chips will be used more often in the near future.

The main advantage of DNA and protein chips is the huge amount of data that

can be gathered in a single experiment. However, there are also points that need

careful consideration. The quality of an expression profile analysis based on array

127

4.2 Advanced Techniques

data is highly dependent on the number of repeated sample measurements and

on the array preparation, hybridization, and signal quantification procedure (Wier-

ling et al. 2002). The many samples on the chip also pose a problem regarding

multiple testing. If care is not taken, a large number of false positives is to be ex-

pected.

4.2.3

Yeast Two-hybrid System

The yeast two-hybrid (Y2H) system is a technique for the high-throughput detec-

tion of protein-protein interactions. It rests on the fact that some transcription fac-

tors (TF), such as the yeast Gal4 gene, have a modular design. The DNA-binding

domain is separated from the activating domain. To test whether two proteins (we

will call them bait and prey) interact, their genes are fused to the DNA-binding or

DNA-activating domain, respectively, of the TF (Fig. 4.9). The bait binds to the

DNA via its TF fragment. If bait and prey do interact, the two domains of the tran-

scription factor come close enough together to stimulate the expression of a repor-

ter gene. If bait and prey do not interact, the reporter gene is silent. This technique

can be used to find all interacting partners of the bait protein. For this purpose

yeast cells are transformed with an expression library containing prey proteins

fused to the activating part of the TF. Although the detection occurs in yeast, the

bait and prey proteins can come from any organism. This single-bait–multiple-prey

system can even be extended to a multiple-bait–multiple-prey approach, which

makes it possible to obtain the complete protein interactome of yeast (Uetz et al.

2000). However, as with most high-throughput techniques, the two-hybrid system

is prone to false positives, as indicated by the fact that a second study using the

same technique derived a quite dissimilar yeast interactome (Ito et al. 2001). To cor-

roborate the interactions obtained with Y2H, affinity chromatography or co-immu-

noprecipitation can be used.

128

4 Experimental Techniques in a Nutshell

Fig. 4.9 The yeast two-hybrid system identifies protein-protein in-

teractions. The genes of the bait and prey proteins are fused to parts

of a yeast transcription factor. If bait and prey interact, the different

parts of the transcription factor come close enough together to acti-

vate the expression of a reporter gene.

4.2.4

Mass Spectrometry

The identification of individual proteins of a cell is an essential part for studying bio-

logical processes. The first step towards identification often involves a separation of

the protein broth. 2D gels (see Section 4.1.3) or the different forms of column chro-

matography (see Section 4.1.5.2) are frequently the method of choice. The separated

proteins can then be identified by cleaving them enzymatically or chemically into

specific peptides and determining the exact size of these fragments. The resulting

size distribution, the fingerprint of the protein, is compared to the theoretical finger-

prints that are calculated for all proteins in the sequence databanks. This is possible

because it is known at which amino acid sequence proteases like trypsin cut a pro-

tein.

Mass spectrometry (MS) has been used for the past 50 years to measure the

masses of small molecules with high accuracy. However, its application to large bio-

molecules has been limited by the fragility and low volatility of these materials. But

these problems have been solved with the development of the matrix-assisted laser

desorption/ionization time-of-flight (MALDI-TOF) technique. The polypeptide is

mixed with a solvent and an excess of the matrix material and is placed inside the

mass spectrometer, where the solvent evaporates. The probe is then targeted with a

laser beam that transfers most of its energy to the matrix material, which heats up

and is vaporized. During this process the intact polymer is charged and carried into

the vapor phase. An electrical field accelerates the molecules, which then traverse

down an evacuated tube, (smaller ions arrive in a shorter time at the detector than

massive ions), and the time of flight is recorded. The masses of the different ions

can then be calculated, taking into account that the TOF is proportional to the square

root of the mass divided by the charge of the ion (TOF 



m=z

The accuracy of MALDI-TOF is extremely high. The error is less than 1 part in

. That means the error in determining the mass of a 100-kDa protein is less than

1 Dalton! Because of this very high accuracy, MS can also be used for sequencing of

peptide fragments. In this case two mass spectrometers have to be used in succes-

sion. The first one separates the peptides (according to their time of flight) and feeds

them individually into the second MS, where they are further fragmented at their

peptide bonds. The output of the second spectrometer is therefore mass peaks that

are separated by the mass of a single amino acid, which can be used to construct the

sequence of the peptide. With the use of degenerate primers (see Section 4.2.1), the

gene for the protein can then be cloned. This is useful if the fingerprint of the pro-

tein is not found in the database and the protein cannot be identified this way.

Finally, MS-based peptide sequencing also yields information about posttransla-

tional modifications. If an amino acid is chemically modified (i. e., phosphorylated

or glycosylated), this results in a characteristic change of the mass and thus it can be

identified.

129

4.2 Advanced Techniques