BSCI 1510L Literature and Stats Guide: Microbiology 101

Some of these items will be discussed in more detail in subsequent sections.

1. The non-handle portion of instruments that you use to transfer inoculants should never touch any non-sterile object.  This is to prevent bacteria and fungi from the environment from getting into the sterile media and also to prevent the bacteria that you are culturing from getting onto lab equipment and students.  Do not remove disposable sterile spreaders or loops from their packaging until immediately before you will use them.  Open their package at the handle end, touch only the handle, and as you remove the implement do not let it touch the place where you grasped the packaging to open it.  Never lay a spreader or loop on the bench top outside of its packaging.  If you wish to attach a serological pipette to its pump ahead of use, open only a small portion of the non-tip end of the plastic sleeve and leave the rest of the pipette undisturbed in the sleeve while you attach the pipette to the pump.  You can then place the assembly on the benchtop and maintain sterility until you pull the pipette out of the sleeve.  Immediately after use, place contaminated implements in the appropriate waste container (i.e. the autoclave bag). 

2. Media containers (screw capped tubes containing broth and plates containing agar-based media) should be left closed except during the brief interval that you are inoculating them or removing a sample. 

3. Agar plates should be stored lid side down to prevent condensation from dripping onto the media.  The exception to this is immediately after adding liquid inoculants or media additives (such as X-gal and IPTG).  You should allow the liquid to be absorbed completely before inverting the plates. 

4. A bacterial colony on a plate results from a single bacterium that was present when the plate was inoculated.  Thus all of the many bacteria in the colony should be identical to the one from which they originated.  Starting a pure broth culture from a plate requires touching only a single colony with the inoculating loop.  The colony should be widely separated from others to prevent accidentally touching two colonies (and therefore having the possibility of producing a mixture of two types of bacteria in the broth culture).  The colony should also be perfectly round.  An oblong colony may have formed from the merger of two colonies that were near each other when they started to grow.

5. When using the colony counters, be sure that you remove the lid and place the ground wire in contact with the edge of the agar.  Since you will destroy the sterility of the plate when you touch the probe tip to the colonies, do not count any plates from which you plan to later culture colonies.  When you have finished counting the plates, wipe the tip of the probe with a tissue.

6. When you no longer need cultured plates, place them in the special autoclave bags.  Do not place them in the regular trash.  Do not put regular trash in the autoclave bags.  Liquid bacterial waste should be poured into the designated containers and the empty tubes should be placed in an autoclave bag.  Container with small amounts of liquid bacterial waste (e.g. microfuge tubes) can be placed directly in the autoclave bags without emptying the liquid.

A common task in the field of microbiology is to determine the titer of a suspension of bacteria.  Functionally, cell titer is similar to concentration and generally the same kinds of calculations that can be done with concentrations can be done with titers.  However, concentration is appropriate for solutions, whereas cell cultures are suspensions.  Concentration is expressed in grams or moles per unit volume, while titer is expressed in cells per unit volume (usually per cm³). 

A fundamental difficulty in measuring bacterial titer is that it requires "counting" something that cannot be seen with the naked eye and is visible only under high magnification through a microscope.  For this reason, indirect methods are typically used.  Two of the most common methods are spectrophotometry and plating.

Although spectrophotometer readings are a quick and easy way to measure bacterial suspensions, they provide only a relative measure of titer.  Although there are some general rules of thumb relating absorbance units to actual numbers of bacteria per cm³, the exact relationship could vary depending on the type of bacteria, type of culture medium, etc.  To determine a more exact relationship requires an actual count of the number of bacteria present in a certain volume of medium.  This presents some logistical challenges given that the bacteria are effectively invisible and that there may be over a million of them present in each cm³ of suspension!  Fortunately, a standardized method has been developed for counting bacteria.  This method is known as serial dilution and plating. 

Before making serial dilutions, the bacteria suspension must be thoroughly mixed.  This is important because unlike solutions, which remain uniformly mixed indefinitely, suspended materials eventually settle down to the bottom of their containers.  If a dense bacterial suspension is left to stand undisturbed overnight, many bacteria can be seen as a white pellet in the bottom of the tube.  Even over shorter time periods the titer of the suspension at the bottom of the tube will be higher than the titer at the top.  The best way to re-suspend bacteria is to use a vortexer.  In our lab the vortexers (a.k.a. Vortex Genies) are sitting on the top of each bench. 

To reduce the number of bacteria present in a suspension to a reasonable number, the suspension must be diluted a great deal.  As a rule of thumb, you can expect that an overnight culture of Escherichia coli will have around 1 x 10⁸ cells per cm³.  To reduce this number to a few hundred cells per cm³ would require measuring a very tiny amount of stock suspension into a very large volume of culture medium.  It is better to do the dilution in several steps with each dilution using a small fraction of the previously diluted suspension.  Keep in mind that each dilution will introduce systematic error into the final measurement.  This error should be kept to a minimum by careful measurement and thorough mixing after each dilution. 

Each dilution can be described by comparing the volume of suspension before the dilution with the volume present after the dilution.  The dilution factor is the ratio of the initial volume over the final total volume; it is always a fraction less than one.  For example, if 10 µl of suspension is added to 990 µl of buffer (for a total final volume of 1000 µl or 1 ml), the dilution factor would be 10 µl/1000 µl or 0.01 .  A dilution factor can be used to calculate the titer of the final suspension by multiplying the titer of the initial suspension by the dilution factor.  (Dilution factor can also be used to calculate final concentrations of solutions from initial concentrations.)  The reciprocal of the dilution factor (i.e. the final volume over the initial volume) is also used to describe a dilution.  For example, the previous example can also be described as a 100-fold dilution. 

The total effect of serial dilutions can be calculated from the product of the dilution factors of each of the individual dilutions.  For example, two 100-fold dilutions followed by a dilution of 5 µl into 20 µl (final volume of 25 µl; dilution factor 0.2) would produce a combined dilution factor of 0.01 x 0.01 x 0.2 = 0.00002 which is a 50 000-fold dilution.  Because of the very large and very small numbers involved, it is usually advisable to use scientific notation (i.e. dilution factor=2x10^-5 and 5x10⁴-fold dilution). 

Fig. 6 Bacterial colonies on a nutrient agar plate (colony counter grid visible through agar)

After the number of the bacteria present in the diluted suspension has been reduced to a reasonable level, they must be counted.  Because of the small size of the bacteria, this would be impossible to do directly - like searching for tiny needles in a very large haystack!   Fortunately, there is a simple way to determine how many bacteria were present in a volume of suspension.  A known volume of suspension is spread over the surface of a Petrie plate containing nutrient agar.  At the location on the agar plate where each bacterium comes to rest, a colony of bacteria will begin to grow radially from the location of the inoculating bacterium.  After incubating the plate overnight at 37˚C, the colonies will be of sufficient size that they can easily be seen by the naked eye (Fig. 6).  By counting the colonies on the plate, the number of bacteria present in the suspension on the previous day can be inferred.  These bacteria represent only a tiny fraction of those present in a cm³ of the original undiluted solution, but by considering the amount of solution plated and the combined dilution factor of all of the dilutions used to create the plated solution, the titer of the original solution can be inferred. 

One common misconception is that the conditions under which the plate is maintained will affect the titer calculation.  The amount of time that the plate is incubated at 37˚C will affect the size of the colonies (i.e. longer incubation will make the colonies larger and vice versa), but won't affect the number of colonies.  Likewise, after the overnight incubation the plates can be stored in a refrigerator or cold room (i.e. about 5˚C) indefinitely until they are counted.  Their growth rate at that temperature is so slow that the size of the colonies changes little over any reasonable amount of time - again there is no effect of time on the number of colonies.  

There is no particular amount of diluted suspension that must be spread over the surface of the agar plate.  However, a volume of less than about 25 µl is difficult to spread uniformly over a typical plate.  Volumes of greater than about 500 µl are not hard to spread, but tend to pool toward one side of the plate or another if the plate isn't level and that amount of liquid takes a long time to be absorbed by the agar.  100 µl is a very convenient amount to spread over a plate with a diameter of about 8 cm.  Not only is it easy to spread uniformly and is easily absorbed, but it also simplifies the calculation of the titer.  It is standard to specify the titer as cells per cm³ (i.e. per ml = 1000 µl).  If 100 µl is plated rather than one cm³, then the titer inferred using the counts and total dilution factor is on the basis of per 100 µl rather than per 1000 µl.  To convert the inferred titer to the standard units, one simply multiplies the count by 10. 

Although in theory any number of colonies on a plate can be counted, in practice it is best to try to count colonies on a plate having between about 20 and 200 colonies.  If a plate has fewer than 20 colonies, the task of counting becomes very easy and there is little likelihood of miscounting.  However, the fewer the number of colonies measured, the more likely random non-representative sampling is to affect the accuracy of the measurement.  As the number of colonies approaches 10, the resulting calculated titer effectively has only one significant digit of precision, which translates into a very poor estimate.  On the other hand, if a plate having a diameter of about 8 cm has more than 200 colonies, accurate counting becomes more difficult because it becomes difficult to keep track of which colonies have already been counted.  Additionally, the higher the density of colonies on a plate, the more likely it is that two bacteria will fall so close to each other that their two colonies merge and appear to be a single colony, resulting in an undercount. 

Fortunately, a tool called a colony counter makes it easier to count dense colonies on a plate.  A colony counter is a relatively simple electrical device having a wire placed in the agar near the edge of the plate, and a metal probe (connected to the counter by another wire) used to complete the circuit each time it is poked into the agar at the location of a colony.  Each time the circuit is completed, the counter increments the displayed count by one.  The lighted stage of the counter is divided into a 1 cm grid that can be used to help keep track of which sections of the plate have been counted (Fig. 6).  Using a colony counter, it is practical to count colonies numbering in the hundreds. 

Although it is preferable to count plates having fewer than about 200 colonies, it is possible to estimate colony numbers ranging up into the thousands under certain conditions.  This can be done by counting a known fraction of the plate (e.g. half or quarter) or known area of the plate (one or more centimeter squares on the grid) and then scaling the count up to the area of the whole plate.  For example, if a quarter of the plate was counted, the count would be multiplied by four.  If 3 cm² of a plate having a total area of 51 cm² was counted, then the count would be multiplied by 51/3.  In order to make this kind of estimation accurately, the colonies must be spread uniformly over the plate so that it is safe to assume that the part of the plate that was counted was representative of the plate as a whole.  If any part of the plate has a streak of solid bacteria, then the entire plate becomes impossible to use in a titer determination since it is not possible to know how many colonies were overlapping in the streak. 

Here are some rules of thumb for counting colonies:

• If there are a few colonies (about 100 or less) on the entire plate, count every colony. 
• If there are an intermediate number of colonies (100 to 400), use the grid on the counter to divide the plate into quarters.  Count the number of colonies in ¼ of the plate, then multiply by 4 to estimate the total number of colonies on the plate.
• If there are a large number of colonies (greater than about 400), count the number of colonies in a certain number of representative 1 cm squares (use the grid on the counter as a guide; Fig. 6), then use the area of the plate (pi r²) to estimate the total number of colonies on the plate.  Rulers are available in the lab drawers. 
• If the colonies form a continuous sheet (known as a "lawn"; Fig. 7), it is not possible to obtain a count.

Fig. 7 Bacterial lawn on a nutrient agar plate

At this point you may be wondering how it is possible to know how much to dilute the stock suspension in order to arrive at a plate having between 20 and 200 colonies.  The answer is that it is not possible to know!  For that reason, microbiologists usually inoculate a series of plates that differ in dilution factor by as much as an order of magnitude each.  It is assumed that some or many of the plates will be useless because they either contain few or no colonies, or contain too many overlapping colonies to count.  That is fine as long as at least one plate falls into the countable range.  Since the colony densities on all of the plates are related by known dilution factors, any of the countable plates can be used for a titer estimate.  If the microbiologist has an expectation of the approximate titer of the stock solution based on prior experience, the number of plates in the dilution series could be as few as two or three.  But it is better to make too many plate dilutions than too few, because producing no countable plates means repeating the experiment and waiting at least a day for the colonies on the new plates to develop.

What volumes should be used to make the serial dilutions?  There is no exact answer to this question but you should use volumes that are:

• large enough to be accurately measured,
• using the most precise measuring instrument available, and
• in a way that doesn't waste a lot of materials. 

Using a P20 micropipetter, it is possible to accurately measure volumes as small as 5.0 µl if you are careful.  Although the P20 has marks that allow it to be set to the nearest 0.02 µl, due to the physical limitations of you and the instrument, you are probably really measuring to an accuracy of about 0.2 to 0.5 µl.  Thus a 5.0 µl measurement is probably accurate to about 10% and any smaller measurements would be even worse.  (We will commonly measure out volumes as small as 1 µl later in the semester, but those cases we aren't doing quantitative work and for those materials it is OK for the volumes to be approximate.)

What if you want to measure 19 µl?  Any of the micropipetters could be set for that value, but a P20 would be precise to a few tenths of a microliter, a P200 would be precise to a few microliters, and a P1000 would be precise to around ten microliters.  Thus the P20 would be the best choice.  What about 25 µl?  That is higher than a P20 can measure, so you will either have to use a P200 or add together two volumes (e.g. 5 and 20 µl) using a P20.  For convenience, use the P200.  What about 1 ml?  You could measure that with a 10 ml serological pipette, but you would be lucky to achieve a precision of 0.1 ml (i.e. 100 µl) with that instrument.  On the other hand, a P1000 set on 1000 µl (i.e. 1 ml) would be precise to about 10 µl.  So use the P1000.  The bottom line is that you should use the smallest instrument that has the capacity to go up to the volume that you need to measure. 

Now let's say that you need to make a 10 000-fold dilution and intend to plate the resulting suspension.  You could measure out 10.0 ml of buffer using a serological pipette (technically 9.999 ml, but you can't measure that precisely with a serological pipette!), then add 1 µl (i.e. 0.001 ml) of the stock suspension using a P20 micropipetter.  That would be a really bad idea for two reasons.  There would be a relatively high percent error in measuring out the 1 µl (probably 20-50% error).  You would also be making about a hundred times more solution than the 100 µl that you need to spread on the plate.  On the other hand, you could achieve the same thing if you diluted 10.0 µl of stock suspension in 990 µl of buffer in a 1.5 ml microfuge tube, then repeated the dilution using 10.0 µl of the resulting diluted solution into another 990 µl of buffer in a second tube.  In this case, your percent error would probably be only a few percent if you measured carefully and vortexed each dilution to mix it thoroughly.  In addition, you would only be using about 2 ml of buffer (still way more than you need for one plate, but better than the 10 ml in the first example).

Gerhardt, Philipp. 1994. Methods for general and molecular bacteriology. Washington, DC: American Society for Microbiology.