Yan He and Peter W. Baas
Department of Neurobiology and Anatomy Drexel University College of Medicine Philadelphia, Pennsylvania 19129
I. Introduction II. Methods
D. Preparation of Dissociated Cultures
E. Maintenance of Cultures
F. Preparation of Explant Cultures
G. Reducing Nonneuronal Contamination III. Cultures
A. Short-Term Chick Dorsal Root Ganglia Culture
B. Short-Term Rat Sympathetic Culture
Cultures of vertebrate peripheral neurons have been used to address a variety of issues related to the cell biology of the neuron. They are particularly amenable to experimental manipulations, such as microinjection, and can be cultured under a variety of different conditions designed to meet the needs of the particular experiment. This chapter focuses on cultures of rat sympathetic neurons from the superior cervical ganglia and on cultures of chick sensory neurons from the lumbosacral dorsal root ganglia. Information is provided on methods for dissection, preparation of culture dishes and substrates, composition of media, relevant growth factors, reduction of nonneuronal contamination, and maintenance of the cultures.
METHODS IN CELL BIOLOGY, VOL. 71 Copyright 2003, Elsevier Science (USA). All rights reserved. 0091-679X/03 $35.00
Over the past several years various different kinds of neurons from the peripheral and central nervous systems of various different kinds of animals have been cultured successfully. These cultures have been used to investigate a number of different issues related to the cell biology of the neuron. Several factors contribute to the selection of a particular type of neuron for a particular study. Although some studies focus specifically on an issue related to one type of neuron, most studies are open to the use of different kinds of neurons. For example, our laboratory studies mechanisms of cytoskeletal organization and axonal transport, and these mechanisms are likely to be preserved across different types of neurons. In these cases, the main factor in choosing which neuron to use is how amenable a particular type of neuron is to the particular experimental manipulations that are required to carry out the study. In our laboratory, we most frequently culture rat hippocampal neurons, rat sympathetic neurons from the superior cervical ganglion, and chick sensory neurons from lumbosacral dorsal root ganglia. Hippocampal neurons, from the central nervous system, have the advantage of undergoing very stereotypical and well-defined stages of axonal and dendritic development. However, they are more tedious to culture, somewhat finicky, and generally die if microinjected. In contrast, the two types of peripheral neurons are easier to culture, more rugged, survive injection, and can be grown under a much wider variety of culture conditions tailored to specific experimental needs. The reason why we use chick for sensory neuronal cultures and rat for sympathetic cultures is simply because of the backlog of knowledge from previous work. It is certainly possible to culture rat sensory neurons and chick sympathetic neurons, but fewer studies have been performed on these cultures, at least in our field.
The cultures of sensory neurons have the advantage of simplicity, they generate axons but not dendrites, and display relatively simple morphologies. However, sympathetic cultures have the advantage of complexity, they generate dendrites as well as axons, and display a much wider range of axonal and dendritic behaviors depending on the culture conditions. Therefore, if the purpose of the study is simply to measure the effects of an experiment on axonal outgrowth or length, sensory neurons would be the better choice. However, if the purpose of the study is to investigate finer features of axogenesis, branch formation, dendrites, or other more complex issues, then sympathetic neurons would be the better choice. As far as we can tell, both types of neurons are about the same in terms of their viability to experimental manipulations, such as microinjection, axotomy, transfection, and application of anticytoskeletal drugs. The dorsal root ganglia (DRG) of chick embryos are much easier to dissect than the superior cervical ganglia of rat pups, but after dissection is completed, subsequent steps in the preparation and care of the cultures are quite similar.
There are superb chapters in other books that outline in great detail various options for generating these cultures (Bray, 1991; Higgins et al., 1991; Mahanthappa and Patterson, 1998; Smith, 1998; Johnson, 2001). Our main purpose here is not to be exhaustive with regard to options, but rather to share the methods that we have come to rely on in our laboratory, as well as insights that we have gained over the years.
= II. Methods A. Substrate
One of the most important issues for neuronal cell culture is substrate. Neurons are poorly adhesive cells compared to many kinds of cells that can be cultured, and yet adhesion to the substrate is absolutely essential for their viability and their capacity to extend neurites. In general, neurons do not adhere to plain glass or even to plastic culture dishes (except when methylcellulose is added to the medium; see later). There are several possibilities for treating these surfaces that enhance their capacity to adhere to cells. As with the rest of this chapter, we describe the most commonly used approaches in our laboratory. In theory, with appropriate substrate treatments, cultures can be grown either on the surface of tissue culture dishes or on glass coverslips. Glass coverslips offer a more optimal surface for light microscopy, but are hard to manipulate for experiments if they are simply placed inside petri dishes. To circumvent this problem, we routinely grow cultures on glass coverslips that are adhered to the bottom of a 35-mm petri dish into which a hole has been drilled. This permits the cells to be visualized easily using an inverted microscope, microinjected, and marked for relocation. Using terminology originally coined by the Banker laboratory, we refer to these hybrid dishes as "special dishes.'' When culturing directly on plastic, it is important to buy dishes that are treated for cell culture, but for the manufacture of special dishes, it is fine to save money by using untreated petri dishes. Sometimes it is useful to prepare special dishes using glass coverslips that have been prepho-toetched for easy relocation of particular cells of interest. Details for cleaning glass coverslips and preparing special dishes are provided in Table 1.
The most common method for promoting the adhesion of neurons is to treat the plastic or glass with a series of positive charges, which are attracted to the negative charges on the surface of the cell membrane. We use poly-D-lysine for this. Several different varieties of polylysine are available; details on the one we use commonly are provided in Table 2. It is necessary to use a relatively low concentration of poly-D-lysine and to rinse very extensively after the glass or plastic is treated. Any excess poly-D-lysine that does not attach to the plastic or glass can be extremely toxic to the cells, as can too much poly-D-lysine that has attached. Generally speaking, we use either 0.1 or 1.0 mg/ml, depending on whether any toxicity arises with the higher concentration. The presence of serum in the medium usually permits the use of the higher concentration, while it is often necessary to use the lower concentration if there is no serum in the medium. Poly-D-lysine alone can be a reasonable substrate for chick sensory neurons (as well as cultures of brain neurons) but generally does not support axonal outgrowth particularly well from
Preparing Special Dishes
A. Place circular glass coverslips (Carolina Biological Supply Company) or etched grid coverslips (Bellco) into ceramic coverslip holders. Immerse the holders under HNO3 in a glass container. Let coverslips soak in HNO3 overnight in the dark.
B. Rinse coverslips once with double-distilled water and wash three more times, 1 hour each time.
C. Dry coverslips in oven.
Note: Keep HNO3 in dark at all times because it is light sensitive. If HNO3 becomes brown rather than colorless, stop using it.
A. Introduce a circular smooth-edged hole into the bottom of a 35-mm petri dish. This can be done using a vertical drill press with a forstner bit or with a lathe. The diameter of the hole can vary depending on the needs of the experiment; we generally use a diameter of 13 mm.
B. Prepare a mixture of three parts of paraffin and one part of vaseline. Melt in hot water bath.
C. Using a paint brush, paint a ring of wax around the hole in each dish. Take care not to paint too close to the hole and do not use too much wax. Too much wax can leak onto the region of the coverslip where the cells are to be plated.
D. Invert the dish and place the clean coverslip on top of the hole.
E. Put under inverted hot plate, let wax melt, and coverslip fall down evenly. Pull out and let cool before any wax slops into the well.
F. Sterilize the dish and lid by spraying 70% ethanol evenly on the surface and radiating them under UV light in a tissue culture hood for at least 30 min. After the ethanol evaporates completely, the dishes are ready for plating neurons or coating with different substrates.
Note: If the cultures are ultimately to be used for electron microscopy, it is desirable to use a more heat-resistant adhesive such as epoxy to attach the coverslip to the plastic dish.
sympathetic neurons. The reason for this is unclear, but it may be that the poly-D-lysine is actually "too sticky'' for sympathetic neurons.
A more favorable and more physiological substrate is provided by introducing laminin after the poly-D-lysine. The application of laminin, a biologically active matrix molecule, promotes better axonal outgrowth from sensory neurons and extraordinarily rapid and robust axonal outgrowth from the sympathetic neurons (Rivas and Goldberg, 1992; Tang and Goldberg, 2000). Laminin can either be applied over the poly-D-lysine prior to plating the cells or be added to the medium together with or after the cells have been plated. The precise morphology and rate of axonal outgrowth will vary somewhat depending on when the laminin is added. The laminin adheres to the positive charges of the poly-D-lysine and provides a robust substrate for the neurons. Even more robust results can be achieved using a partially defined mixture of substrate-related growth factors called Matrigel rather than laminin (Yu et al., 2002). Additional details are provided in Table 2.
Years ago we commonly used a substrate of rat tail collagen to promote the attachment of sympathetic neurons. We rarely do this anymore because the substrate is rather thick, is rather three-dimensional (which is not good for microscopic analyses), and because axonal outgrowth is very slow on collagen.
A. Prepare borate buffer (40 ml)
Ingredient Amount Final concentration
Borax (Sigma) 190 mg 4.75 mg/ml
Boric acid (Sigma) 124 mg 3.1 mg/ml
Double-distilled H2O 40 ml 100% (v/v)
Good for up to 1 month at room temperature.
B. To prepare 1 mg/ml stock solution, dissolve 10 mg of poly-D-lysine (Sigma) in 10 ml borate buffer.
C. Filter sterilize using a 0.2-^m filter. Aliquot desired amount and freeze at —20 C.
D. Add 0.2 ml of poly-D-lysine to the well. Two different methods can be used, one with a higher concentration and the other with a lower concentration of poly-D-lysine. The higher concentration generally provides better adhesion, but sometimes can produce toxicity, particularly if serum-free medium is used. If the latter proves to be the case, then the lower concentration should be used.
Method 1: Coat the glass well with 1 mg/ml poly-D-lysine for 3 h at room temperature. Rinse the dish six times with sterile distilled water, 5 min each time. At the end, add 2 ml of water to the dish and keep the dish in a 37 C incubator overnight. On the next day, rinse the dish with water one more time. Take off water and let dish dry completely. Put the lid back on the dish and keep at room temperature. They are good for up to 1 month.
Method 2: Coat the glass well with 0.1 mg/ml poly-D-lysine (dilute 1 mg/ml stock solution 10 times with borate buffer) overnight at 4 C. On the next day, rinse the dish six times with sterile distilled water, 5 min each time. After the dish dries completely, put the lid back on the dish. They can be kept at room temperature for up to 1 month.
A. Prepare poly-D-lysine-coated dishes as described previously.
B. On the day of culturing, dilute 10 ^l laminin (Invitrogen) in 1 ml Leibovitz's L-15 medium. Add 200 ^l diluted laminin into each glass-bottomed well that has been precoated with poly-D-lysine.
C. Incubate the dish in a 37 C, CO2-free incubator for 3 to 4 h.
D. Take off laminin immediately before plating cells into the well.
A. Purchase collagen derived from rat tail (BD Biosciences). The product will generally be roughly 4 mg/ml in a dilute acetic acid solution.
B. Collagen can be applied directly to the surface of a plastic culture dish. For glass, precoat with poly-D-lysine as described earlier.
C. The amount of collagen to produce a desirable substrate is determined empirically. Generally, 2-3 drops of the undiluted product is sufficient to coat the glass-bottomed well of a special dish. An insufficient amount of collagen will dry unevenly in patches.
However, the collagen substrate has one advantage that is particularly helpful for certain types of experiments. Everyone who works with cultured neurons struggles with the fact that the neurons often lose their attachment and simply "float away'' if they are extracted too strongly or exposed to anything that might mechanically or otherwise disturb them. When grown on rat tail collagen, sympathetic neurons are extremely well attached, such that they can be extracted with very high concentrations of detergents such as Triton X-100, and even with high salt concentrations. This permits a clean separation of soluble and insoluble components of the cytoplasm for biochemical studies and also produces "squeaky clean'' cytos-keletal polymers for immunostain analyses. We have found this particularly useful for immunoelectron microscopic analyses of tubulin isoforms within microtubules (Baas and Black, 1990). A collagen substrate is relatively easy to work with on plastic (see Table 2), but is somewhat harder to deal with on glass. The problem is that while the neurons are extremely well attached to the collagen, the collagen itself can sometimes roll off the glass or plastic in a sheet, which is more common on glass than on plastic. The glass needs to be pretreated with poly-D-lysine, but even this does not rectify the problem in many cases.
For making substrates of rat tail collagen, we used to prepare our own collagen by plucking collagenous tendons from a rat tail and dissolving them in acetic acid. This is relatively simple and there are published methods for doing so (Johnson and Argiro, 1983). However, it is now more time and cost-efficient to simply purchase sterile collagen solutions for cell culture. A few drops of the product are placed on the glass or plastic and are then smeared evenly using a Pasteur pippette into which a 90° bend has been introduced by placing it in a flame. (For glass, it is first necessary to treat with poly-D-lysine.) The collagen is then allowed to air dry. If insufficient collagen is applied, it will tend to dry in patches. The optimal quantity is best determined empirically. As noted earlier, we rarely use collagen substrates for dissociated cultures because they make for cultures with poor optical qualities for light microscopy and because the axons are very slow to grow. Our protocol for collagen coating is provided in Box 2.
We generally use two different types of media, both of which are adequate for either sensory or sympathetic cultures. One of these is a modified L-15-based medium, which has the advantage of maintaining pH in air, but has the disadvantage of supporting cultures only for about 1 or perhaps 2 days. The other medium is an N2-based medium, which is excellent for long-term cultures, but requires a 5% carbon dioxide environment to maintain pH. Long-term cultures can be grown in the N2-based medium and then transferred to the L-15-based medium for short-term experiments on the microscope stage. We have found that peripheral neuronal cultures respond much better to L-15-based media than to media buffered with HEPES, although other laboratories have had good luck with HEPES (Gallo et al., 1997). Formulations for the L-15-based and the N2-based media are provided in Table 3. The L-15-based medium requires serum to be present, whereas the N2-based medium contains supplements that permit it to be used either with or without serum. We have found that serum, although not required, appears to make the cells somewhat more resilient to experimental manipulations such as microinjection. Other laboratories have achieved excellent results with adult rat serum (Wang et al., 2000), but we have not tried this. Both media contain
Was this article helpful?