Gene Therapy (2004) 11, & 2004 Nature Publishing Group All rights reserved /04 $

Similar documents
Supporting Information

Supporting Online Material for

BIPN 140 Problem Set 6

BIPN 140 Problem Set 6

3.E.2 Continued. This is the essential knowledge statement from the curriculum framework. Detect---process--- response

Chapter 3 subtitles Action potentials

Neuronal Differentiation: CNS Synapse Formation

The Brain & Homeostasis. The Brain & Technology. CAT, PET, and MRI Scans

Supporting Online Material for

Chapter 5 subtitles GABAergic synaptic transmission

Presynaptic NMDA receptor control of spontaneous and evoked activity By: Sally Si Ying Li Supervisor: Jesper Sjöström

Lecture 22: A little Neurobiology

Synaptic Transmission: Ionic and Metabotropic

Section: Chapter 5: Multiple Choice. 1. The structure of synapses is best viewed with a(n):

Neurons of the Bed Nucleus of the Stria Terminalis (BNST)

nachr α 4 β 2 CHO Cell Line

1) Drop off in the Bi 150 box outside Baxter 331 or to the head TA (jcolas).

ab CytoPainter Golgi/ER Staining Kit

Portions from Chapter 6 CHAPTER 7. The Nervous System: Neurons and Synapses. Chapter 7 Outline. and Supporting Cells

Supplementary Information

Chapter 3 Neurotransmitter release

NEURONS COMMUNICATE WITH OTHER CELLS AT SYNAPSES 34.3

Concept 48.1 Neuron organization and structure reflect function in information transfer

Synaptic Communication. Steven McLoon Department of Neuroscience University of Minnesota

MOLECULAR AND CELLULAR NEUROSCIENCE

Synapse Formation. Steven McLoon Department of Neuroscience University of Minnesota

Alterations in Synaptic Strength Preceding Axon Withdrawal

Ube3a is required for experience-dependent maturation of the neocortex

Nature Neuroscience: doi: /nn.2275

Branches of the Nervous System

Human TRPC6 Ion Channel Cell Line

3) Most of the organelles in a neuron are located in the A) dendritic region. B) axon hillock. C) axon. D) cell body. E) axon terminals.

Supplementary Figure 1. SybII and Ceb are sorted to distinct vesicle populations in astrocytes. Nature Neuroscience: doi: /nn.

Plasticity of Cerebral Cortex in Development

Part 11: Mechanisms of Learning

The mechanisms that govern synaptic vesicle (SV) retrieval

Bioscience in the 21st century

1) Drop off in the Bi 150 box outside Baxter 331 or to the head TA (jcolas).

9/28/2016. Neuron. Multipolar Neuron. Astrocytes Exchange Materials With Neurons. Glia or Glial Cells ( supporting cells of the nervous system)

Actin-Dependent Activation of Presynaptic Silent Synapses Contributes to Long-Term Synaptic Plasticity in Developing Hippocampal Neurons

The mammalian cochlea possesses two classes of afferent neurons and two classes of efferent neurons.

Anatomy of a Neuron. Copyright 2000 by BSCS and Videodiscovery, Inc. Permission granted for classroom use. Master 2.1

Is action potential threshold lowest in the axon?

Activity Dependent Changes At the Developing Neuromuscular Junction

What is Anatomy and Physiology?

Action potential. Definition: an all-or-none change in voltage that propagates itself down the axon

Instructions. Fuse-It-Color. Overview. Specifications

Ivy/Neurogliaform Interneurons Coordinate Activity in the Neurogenic Niche

What effect would an AChE inhibitor have at the neuromuscular junction?

The Nervous System. Nervous System Functions 1. gather sensory input 2. integration- process and interpret sensory input 3. cause motor output

Supplementary Figure 1. ACE robotic platform. A. Overview of the rig setup showing major hardware components of ACE (Automatic single Cell

Chapter 6 subtitles postsynaptic integration

Tyrodes solution in a custom-built imaging chamber as described previously. Images were acquired

MCB MIDTERM EXAM #1 MONDAY MARCH 3, 2008 ANSWER KEY

Neurons Chapter 7 2/19/2016. Learning Objectives. Cells of the Nervous System. Cells of the Nervous System. Cells of the Nervous System

Communication within a Neuron

7.06 Spring of PROBLEM SET #6

Neurons! John A. White Dept. of Bioengineering

Chapter 45: Synapses Transmission of Nerve Impulses Between Neurons. Chad Smurthwaite & Jordan Shellmire

Optimization of the Fuse-It-mRNA Protocol for L929 Cells in the µ-plate 24 Well

Medcelična signalizacija -

SUPPLEMENTARY FIGURE LEGENDS

Chapter 2. The Cellular and Molecular Basis of Cognition Cognitive Neuroscience: The Biology of the Mind, 2 nd Ed.,

AP Biology Unit 6. The Nervous System

SUPPLEMENTARY INFORMATION

biological psychology, p. 40 The study of the nervous system, especially the brain. neuroscience, p. 40

Human Brain and Senses

Outline. Neuron Structure. Week 4 - Nervous System. The Nervous System: Neurons and Synapses

Supplementary Figure 1. Basic properties of compound EPSPs at

Communication Between

Ch. 45 Continues (Have You Read Ch. 45 yet?) u Central Nervous System Synapses - Synaptic functions of neurons - Information transmission via nerve

Structure of a Neuron:

1. What are the two basic types of cells in the nervous system? Neurons and Glial Cells

Synaptic Transmission

SUPPLEMENTARY INFORMATION

Cellular Neurobiology / BIPN 140

The action potential travels down both branches because each branch is a typical axon with voltage dependent Na + and K+ channels.

QUIZ/TEST REVIEW NOTES SECTION 7 NEUROPHYSIOLOGY [THE SYNAPSE AND PHARMACOLOGY]

Instructions. Fuse-It-mRNA easy. Shipping and Storage. Overview. Kit Contents. Specifications. Note: Important Guidelines

Axon initial segment position changes CA1 pyramidal neuron excitability

CHO α 1 β 2 γ 2 GABAA Cell Line

FLASH CARDS. Kalat s Book Chapter 2 Alphabetical

How Nicotinic Signaling Shapes Neural Networks

How Synapses Integrate Information and Change

SUPPLEMENTARY INFORMATION. Supplementary Figure 1

LESSON 3.2 WORKBOOK How do our neurons communicate with each other?

photometry on the extruded cytoplasm.

Axon Nerve impulse. Axoplasm Receptor. Axomembrane Stimuli. Schwann cell Effector. Myelin Cell body

Mouse primary keratinocytes preparation

Nervous System. Master controlling and communicating system of the body. Secrete chemicals called neurotransmitters

File name: Supplementary Information Description: Supplementary Figures, Supplementary Table and Supplementary References

Microinsemination (Intracytoplasmic Sperm Injection) Microinsemination schedule. 1. Preparation of mediums

Neurons. Pyramidal neurons in mouse cerebral cortex expressing green fluorescent protein. The red staining indicates GABAergic interneurons.

Synapses and Neurotransmitters

Chapter 4 Neuronal Physiology

Short-term synaptic plasticity is important for synaptic communication

Summarized by B.-W. Ku, E. S. Lee, and B.-T. Zhang Biointelligence Laboratory, Seoul National University.

Study Guide Answer Key Nervous System

Synaptic plasticityhippocampus. Neur 8790 Topics in Neuroscience: Neuroplasticity. Outline. Synaptic plasticity hypothesis

CRISPRaTest Functional dcas9-activator Assay Kit v1 Last update: 2018/07/04 Cellecta, Inc.

Transcription:

RESEARCH ARTICLE High Ca 2+ -phosphate transfection efficiency enables single neuron gene analysis M Jiang, L Deng and G Chen Department of Biology, The Pennsylvania State University, University Park, PA, USA (2004) 11, 1303 1311 & 2004 Nature Publishing Group All rights reserved 0969-7128/04 $30.00 www.nature.com/gt Introducing exogenous genes into cells is one of the most important molecular techniques to study gene functions. Comparing to other type of cells, neurons are more difficult to transfect with cdnas because they are very sensitive to microenvironmental changes. Among various gene transfer techniques, the Ca 2+ -phosphate transfection method is one of the most popular tools in neuroscience research because of its low cell toxicity and easiness to use. However, it is well known that the Ca 2+ -phosphate transfection efficiency in neurons is very low, typically in the range of 1 5%, which has limited its applications in gene functional analyses. Here we report a novel Ca 2+ -phosphate transfection protocol that dramatically increased the transfection efficiency by 10-fold, up to 60%, while maintaining low cell toxicity. The critical factors are the formation of homogenous snow-like precipitate with particle size about 1 3 mm and the subsequent removal of the precipitate. Using this new transfection protocol, we were able to routinely transfect single autaptic neurons in hippocampal microisland cultures and combine it with electrophysiology and fluorescent imaging methods to study gene functions. This high efficiency, low toxicity, and simple to use gene transfer method will have a broad application in gene research at the single cell level. (2004) 11, 1303 1311. doi:10.1038/ sj.gt.3302305; Published online 1 July 2004 Keywords: GFP; hippocampal neuron; gene transfer; calcium phosphate transfection; microisland culture Introduction DNA transfection is a powerful tool for studying gene functions. There are various methods to introduce exogenous genes into target cells, including Ca 2+ - phosphate precipitation, electroporation, liposome fusion, gene gun, and viral infection. 1,2 Among these different methods, Ca 2+ -phosphate precipitation remains one of the most popular and cost-effective techniques when transfecting neurons. 3 6 Neurons are very sensitive cells and easy to die if microenvironment changes. Although liposome fusion and viral infection yield higher transfection rate, they also typically lead to higher neuronal toxicity than the Ca 2+ -phosphate transfection method. 1,2 In addition, viral infection requires more preparatory work to package viruses, which makes it less attractive to many labs. Ca 2+ -phosphate transfection offers an advantage of low toxicity and being easy to use. 7 However, the transfection rate of the Ca 2+ -phosphate method is usually low, about 1 5% in average. 1 3 There have been many attempts in various labs to improve the Ca 2+ -phosphate transfection efficiency in neurons. 8,9 However, the highest efficiency reported so far is 13.5%, 8 far less than other methods such as viral infection, which often reaches more than 40% of transduction efficiency. 10,11 To increase the transfection rate while maintaining low toxicity is the ultimate goal for every gene transfer Correspondence: Dr G Chen, Department of Biology, 208 Mueller Lab., The Pennsylvania State University, University Park, PA 16802, USA Received 16 December 2003; accepted 12 April 2004; published online 1 July 2004 method. The low efficiency of Ca 2+ -phosphate transfection has significantly limited its applications, such as in biochemical and electrophysiological analyses of gene functions. Traditionally, Ca 2+ -phosphate transfection has been carried out at high-density neuronal cultures so that a sufficient number of transfected neurons can be obtained even if the overall transfection efficiency is low. 3 However, high-density cultures are not suitable for certain gene functional studies. For example, microisland cultures containing only a single autaptic neuron or a few neurons have been widely used to study molecular and cellular mechanisms of synaptic transmission and synaptic plasticity. 12 18 With the usually low efficiency of 1 5%, Ca 2+ -phosphate transfection is not suitable for transfecting single autaptic neurons to study gene functions. In order to transfect single autaptic neurons in microisland cultures, the transfection efficiency has to be greatly improved. Our lab is engaged in studying synapse formation and synaptic protein functions, primarily using microisland cultures of rat hippocampal neurons. We have experimented with various conditions using the Ca 2+ -phosphate transfection method in order to obtain maximal transfection efficiency. We demonstrate that up to 60% transfection efficiency can be obtained using a substantially improved Ca 2+ -phosphate transfection method. The key elements in achieving this high rate are to form precipitate particles of optimal size, and to dissolve the Ca 2+ -phosphate precipitate after transfection. The effectiveness of this improved Ca 2+ -phosphate transfection method has been demonstrated in our recent study of endophilin calcium channel interactions. 19

1304 Results Formation and subsequent dissolution of the DNA/Ca 2+ -phosphate precipitate Although Ca 2+ -phosphate transfection has been widely used for a long time, the exact mechanism is still not well understood. It is believed that DNA/Ca 2+ -phosphate precipitate enters cells through endocytosis and some DNA molecules further accumulate in the nucleus (Anne Marie Craig). Therefore, the formation of appropriate size of the precipitate is one the critical factors for high transfection efficiency. Previous Ca 2+ -phosphate transfection protocols commonly used in many labs often recommend that the DNA/Ca 2+ solution be thoroughly mixed with phosphate buffer solution by continuous shaking on a vortexer. However, we found that continuous vortexing often resulted in large and unevenly distributed precipitation (Figure 1a). Instead, gentle and intermittent vortexing helped to form fine precipitation, which evenly covers the whole field (Figure 1b). This homogenous layer of precipitation is critical for high transfection efficiency, possibly because it covers the whole cell surface and the precipitate particles have higher probability being endocytosed into cells (see Discussion). The second important factor for transfection efficiency is the incubation time of cells in the presence of DNA/Ca 2+ -phosphate precipitation. Unlike cell lines such as HEK 293 cells, long-term exposure of neurons to the fine precipitation will lead to a high level of cell toxicity. Therefore, it is important to remove the precipitation after certain incubation time. According to previous protocols, including the one recommended by Clontech, the precipitation is usually washed with fresh transfection medium for several a c b d Figure 1 Formation of optimal DNA/Ca 2+ -phosphate precipitate and subsequent dissolution to stop transfection. (a) Continuous vortexing when mixing DNA with Ca 2+ and phosphate buffer results in large clusters of precipitate. The precipitate was examined after 1 h incubation. (b, c) Formation of optimal DNA/Ca 2+ -phosphate precipitate achieved by gently vortexing the DNA/Ca 2+ solution with phosphate buffer. For detail, see the Materials and methods section. Image taken after 1 h incubation. (d) Dissolving precipitate with slightly acidic transfection medium preequilibrated in a 10% CO 2 incubator in order to reduce its toxicity to neurons. Scale bar, 50 mm. times. This wash step was assumed to be sufficient to stop the transfection. However, we found that this simple washing could not adequately remove the precipitate, and some neurons died after transfection. To reduce the cell toxicity, we developed a method to effectively remove the precipitate by exposing the precipitate to a slightly more acidic transfection medium preequilibrated in a 10% CO 2 incubator. Figure 1c illustrates cultured neurons covered with fine and evenly distributed precipitate particles after 1 h incubation in the presence of DNA/Ca 2+ -phosphate precipitate in a 5% CO 2 incubator. Figure 1d demonstrates the successful removal of the precipitate after changing into the acidic transfection medium (preequilibrated in a 10% CO 2 incubator) and incubated for 15 min in the 5% CO 2 incubator. High transfection efficiency in hippocampal microisland cultures Since the DNA/Ca 2+ -phosphate precipitate can be effectively removed at any time point, it allowed us to experimentally determine the optimal incubation time for neuronal transfections. We found that 1 3 h incubation time is sufficient for high transfection efficiency while maintaining low cell toxicity. Different plasmids may require different incubation time for best transfections. Figure 2 demonstrates three representative microislands in which a significant proportion of neurons were transfected with EFGP in medium-density cultures. In Figure 2a, 5 out of 6 neurons were transfected. In Figure 2b, 9 out of 12 neurons were transfected. In Figure 2c, the whole microisland contained a total of 22 neurons, among which 17 neurons were transfected, corresponding to B80% transfection efficiency. We counted a total of 211 neurons in that coverslip (represented by Figure 2c) and 127 transfected neurons were found. The transfection rate was 60.2% for the entire coverslip, far more than the usual 1 5% transfection efficiency reported in the literature. 3,8,9 Our routine transfection efficiency usually falls in the range of 25 40% (n410). Typically, the earliest GFP fluorescence signal was observed in cell bodies within 4 h after transfection (dissolving precipitate). The EGFP-transfected hippocampal neurons usually can survive for 7 10 days. It is important to evaluate whether such a high transfection efficiency is accompanied by high cell death rate. To this end, we performed a series of cell death assay using a cell death kit (LIVE/DEAD s viability/ cytotoxicity kit, Molecular Probes, Eugene, OR, USA). We found that the average cell death rate induced by transfection was 5.570.6% (n ¼ 7). In addition, the cell death rate induced by transfection did not appear to be strongly correlated to the age of cultures (4 20 days in vitro (DIV)) or plating density (n ¼ 3). Interestingly, very few glial cells were transfected comparing to the number of neurons transfected in our cultures, consistent with previous findings. 9 This is possibly because the monolayer of astrocytes were treated with Ara-c (cytosine-barabinofuranoside, 4 mm) to stop their overproliferation before plating neurons on top of them. It is interesting to note that much more glial cells will be transfected before Ara-C treatment. Therefore, it seems that in addition to stopping glial cell proliferation, Ara-c may also affect the DNA transfection efficiency in glias. To test the glial cell

a1 b1 c1 1305 a2 b2 c2 Figure 2 High transfection efficiency achieved with our improved protocol in medium-density microisland cultures of hippocampal neurons. (a c) Three independent transfections showing EGFP-transfected neurons. (a1 c1) phase-contrast micrographs. (a2 c2) fluorescent images of GFP-transfected cells. Neurons are transfected with EGFP plasmids 10 15 days after plating. Images were taken 1 2 days after transfection in Figure 2 through Figure 6. Scale bar, 50 mm. death rate, we transfected pure glial culture with EGFP and found the average cell death rate induced by transfection is below 2%. Single neuron transfection in low-density microisland culture Microisland cultures containing only a single autaptic neuron have been widely used as a model system for studying molecular and cellular mechanisms of synaptic transmission and plasticity. 13,14,16,19,20 However, it is extremely difficult to routinely transfect single autaptic neurons because of the low efficiency associated with previous Ca 2+ -phosphate transfection method. Since our improved protocol has dramatically increased the transfection efficiency, transfection of single autaptic neurons has become routine in our lab. One typical example of a single autaptic neuron transfected with GFP-tagged endophilin is shown in Figure 3. Figure 3a illustrates the phase contrast image and Figure 3b shows the GFP image. To examine whether this transfected neuron was still functional, we loaded the cell with FM 4-64, a styryl dye that has been widely used for studying synaptic vesicle recycling. We found that FM 4-64 was actively taken into synaptic vesicles when applied together with a40mm K + solution, which depolarizes cells (Figure 3d). Additional examples of single autaptic neurons transfected with GFP are demonstrated in Figure 4c and d and Figure 5. Moreover, the inset in Figure 5d1 illustrates that whole-cell recording of a GFP-transfected single autaptic neuron revealed a mild paired-pulse depression of the excitatory postsynaptic currents (blocked by 10 mm CNQX, not shown). These functional analyses with FM imaging and electrophysiological recording on transfected single autaptic neurons demonstrate that this improved Ca 2+ -phosphate transfection method is an ideal technique to study gene functions in single neurons. Functional application of high transfection efficiency in low-density microisland culture Besides transfection of single autaptic neurons, our lowdensity microisland cultures also offer many microislands containing 2 3 neurons, of which 1 2 neurons may be transfected. Figure 4a1 illustrates dual wholecell recording on a pair of adjacent neurons, of which one is transfected with EGFP (top, green) and another nontransfected. Electrical stimulation of the transfected neuron (presynaptic) evoked inhibitory postsynaptic currents, which were blocked by 20 mm bicuculline (not shown), indicating the transfected neuron was GABAergic (Figure 4a2). It also revealed a significant paired-pulse depression of the inhibitory postsynaptic currents. Figure 4b illustrates a pair of neurons both transfected with GFP. Different from traditional dual whole-cell recordings in high-density cultures with low transfection efficiency, our low-density cultures with high transfection efficiency offer a great advantage of reliable synaptic connections between the transfected and non-transfected (or both transfected) pair of neurons. This has greatly improved the experimental efficiency and reduced the variations of synaptic connections among different pairs of neurons. High efficiency transfection is applicable to different ages of culture The maturity of neurons is one important factor during transfection. Typically, neurons cultured for 3 10 days were selected for transfection because older neurons are more difficult to be transfected or easy to die. To test whether our improved protocol is useful for transfecting different ages of neurons, we selected a wide range of cultures, from 2 to 82 DIV, for transfection. Figure 5a c shows three examples of 2 DIV single hippocampal neurons expressing EGFP in low-density microisland cultures. Figure 5d and e show two single autaptic neurons transfected with EGFP at 21 DIV. Figure 5d1 inset shows that after 24 h transfection, whole-cell

1306 a b c d Figure 3 Successful transfection of single autaptic neurons in low-density hippocampal microisland cultures. (a) Phase-contrast image of an autaptic neuron in a single microisland. (b) Fluorescent image of the same neuron transfected with GFP-endophilin. (c) High magnification phase-contrast image of selected field in (a). (d) FM 4-64 labeling of presynaptic boutons demonstrating activity-dependent synaptic vesicle recycling in the transfected neuron. Transfection was performed 15 days after plating. Scale bar, 50 mm. recording of the transfected neuron revealed substantial excitatory postsynaptic currents, which were blocked by 10 mm CNQX (not shown), indicating the neuron being glutamatergic and showing mild paired-pulse depression. The substantial synaptic currents also demonstrate that neurons transfected at 21 DIV are healthy and capable of synaptic transmission, thus suitable for gene function analyses. Even more strikingly, some of our microisland cultures can survive a long time, up to 3 months, and still can be successfully transfected. Figure 5f illustrates an old neuron of 82 DIV transfected with EGFP. Why some cultures can survive much longer than others (usually survive for 1 month) is now under investigation. These experiments demonstrate that our improved Ca 2+ -phosphate transfection protocol is useful for transfecting virtually any age of hippocampal cultures, and hence greatly increases its applicability in a variety of functional studies. We also applied our transfection method to both lowdensity hippocampal neuronal culture plated on astrocytes and pure neuronal culture (Banker-type culture). For the low-density hippocampal neuronal culture on astrocytes, our method yielded an average transfection efficacy of 25.2% (n ¼ 2). For the Banker-type culture, the average transfection efficacy is 21.6% (n ¼ 2). High transfection efficiency achieved with different constructs To further demonstrate that our protocol is suitable for expressing different exogenous genes rather than EGFP alone, we transfected hippocampal neurons in mediumdensity cultures with EGFP-Rab3A (Figure 6a), EGFP- VAMP (Figure 6b) and EGFP-CREB (Figure 6c). Note a significantly different expressing pattern among these three proteins. Rab3A and VAMP are presynaptic proteins that play important roles in regulating the docking and fusion of synaptic vesicles. 21,22 Therefore, GFP-fused Rab3A and VAMP had a puncta-like distribution pattern along axons, although they also existed in soma and dendrites probably due to overexpressing of the proteins. On the other hand, CREB (camp response element binding protein) is a transcription factor and CREB-dependent gene expression is critical for a variety of cellular functions in the central nervous system. 23 As shown in Figure 6c, GFP-CREB fusion protein localized specifically to the nucleus, as expected, demonstrating that it was correctly targeted following expression. In addition, Figure 6a illustrates that seven out of seven neurons in this field were successfully transfected with GFP-Rab3A, suggesting that our improved protocol is applicable for expressing a variety of exogenous genes with high transfection efficiency. Discussion This study demonstrated a greatly improved Ca 2+ - phosphate transfection method that dramatically increases the transfection efficiency up to 60%. Our modifications have overcome the major disadvantage of previous Ca 2+ -phosphate transfection method in terms of low transfection efficiency (1 5%). With such a high transfection efficiency and low cell toxicity, we expect

a1 b1 1307 a2 b2 c1 d1 c2 d2 Figure 4 Transfection of hippocampal neurons in low-density microisland cultures facilitates electrophysiological analysis. (a1) Overlay of fluorescent and phase-contrast images illustrating a pair of transfected (green) and nontransfected neurons. (a2) Double whole-cell patch-clamp recording of the pair of neurons shown in (a1). Upper trace shows presynaptic current and lower trace the corresponding postsynaptic current. (b) A pair of neurons both transfected with EGFP. (c, d) Two more examples illustrating single autaptic neurons in microislands transfected with EGFP. Neurons were transfected 10 15 days after plating. Scale bar, 50 mm. that this improved Ca 2+ -phosphate transfection technique will have wider application in studies of gene function(s). In order to achieve the highest transfection efficiency, we found the most critical factor is to form optimal size of DNA/Ca 2+ -phosphate coprecipitate particles. Although it is still not clear why cdna can enter cells through this co-precipitation method, it is generally believed that the precipitate particles are endocytosed into cells. If so, it is sensible that particles too large in size may not be effectively endocytosed. On the other hand, particles too small in size may not be a sufficient stimulus to activate an endocytic response. The optimal size of precipitate particles may be in the range of 1 3 mm. However, we would like to caution that this observed 1 3 mm particles might simply be the apparent size. The real effective size particles that are endocytosed into cells can be smaller. In addition to an optimal size, the precipitate should also be formed homogenously to cover the whole field. Speculatively, there

1308 a1 b1 c1 a2 b2 c2 d1 e1 f1 d2 e2 f2 Figure 5 Transfection of hippocampal neurons at different ages. (a c) Transfection of young neurons with EGFP 2 days after plating and images taken 1 day after transfection. Upper panel, phase-contrast images. Lower panel, GFP images. (d, e) Transfection of mature neurons with EGFP 21 days after plating. Whole-cell patch-clamp recording of the transfected autaptic neuron in d1 reveals evoked excitatory postsynaptic currents (inset in d1). (f) Transfection of very old neurons with EGFP 82 days after plating. Scale bar, 50 mm. could be some hot spot for endocytosis to take place on cell surface. 24 If precipitate particles happen to fall on or nearby the hot-spot area, the cell may likely be transfected. A homogenous snow-like cover of the precipitate will ensure the maximal likelihood of DNAcontaining precipitate particles encountering putative endocytic spots. A second critical factor of this improved protocol is to dissolve the coprecipitate after certain incubation time. While a snow-like cover of precipitate all over the cell surface is desirable for high transfection efficiency, it was also found to cause cell death after a long-term incubation. One possible reason is that continuous endocytosis of the DNA/Ca 2+ -phosphate coprecipitate particles may result into excessive accumulation of Ca 2+ -phosphate inside the cell and trigger cell death. Therefore, to reduce cell toxicity, it is vital to dissolve the coprecipitate to stop its accumulation inside the cells. After trying various methods to change the ph of the transfection medium aiming at dissolving the precipitate, we found the best way to achieve this goal with minimal cell toxicity is to preincubate the transfection medium in a 10% CO 2 incubator for at least 2 4 h to make it slightly acidic than the normal transfection medium equilibrated in a 5% CO 2 incubator. For those who culture their cells and carry out transfections in an incubator with CO 2 level higher or lower than 5%, they should consider dissolving the precipitate by preequilibrating the transfection medium in an incubator with correspondingly higher or lower than 10% CO 2 level. It is important to point out that when using acidic medium to dissolve the precipitate, the cells should be placed in the 5% CO 2 incubator rather than the 10% CO 2 incubator. Neurons do not survive very well in the 10% CO 2 incubator if they are normally maintained in a 5% CO 2 incubator. Therefore, the acidic transfection medium preequilibrated in a 10% CO 2 incubator will reequilibrate in the 5% CO 2 incubator while it gradually dissolves the precipitate. By the end of 15 20 min incubation, the ph value will be close to the original one, which is certainly less harmful than placing neurons in a 10% CO 2 incubator.

a1 b1 c1 1309 a2 b2 c2 Figure 6 Correct targeting of a variety of GFP-tagged proteins in hippocampal neurons after transfection. (a) Hippocampal neurons transfected with GFPtagged presynaptic protein Rab3A. (b) Hippocampal neurons transfected with GFP-tagged presynaptic protein VAMP (synaptobrevin). Note the punta-like distribution pattern of the fluorescent signal along axons. Scale bar, 50 mm. (c) Hippocampal neurons transfected with GFP-tagged transcription factor CREB. (c1) Phase-contrast image, (c2) overlay image showing that the GFP signal was only found in the nucleus. The inset in c2 shows the GFP image of the same field. Scale bar, 50 mm. It is also worth of emphasizing that it is important to save the original culture medium before starting the transfection procedure. The reason is obvious that after the stress of transfection, cells will probably recover much faster if returned to their original culture medium. We found this step to be pivotal to maintain low cell toxicity. If neurons are cultured on coverslips, we prefer to transfer the coverslips to a new container (Petri dish or 24-well plate, etc.) to complete the whole transfection procedure and then return the coverslips back to their original environment. This transfection will keep the cell toxicity at a minimal level. Traditionally, ph is a critical factor in forming appropriate Ca 2+ -phosphate precipitation. We have been using Clontech CalPhost kit to do the transfection and found it very reliable in forming the right size of precipitation if gently vortexed. This kit has saved our time by avoiding the need to prepare and test a series of phosphate buffered saline solutions in order to determine which one works best. We have demonstrated that with high transfection efficiency, we are able to routinely transfect single autaptic neurons in microisland cultures. Recently, Cline and colleagues developed a novel single-cell electroporation method to transfer genes to single neurons. 25 Compared to our method, their technique has an advantage of selecting a target neuron to do gene transfer. However, the disadvantage of the single-cell electroporation method is more laborious and expensive. Our improved Ca 2+ -phosphate transfection method offers a simple, inexpensive, but very efficient way to transfect a number of single autaptic neurons. Typically, we can easily get 3 8 transfected single autaptic neurons per 12 mm round coverslip. Therefore, this improved Ca 2+ -phosphate transfection method offers a great opportunity for gene analysis in single neurons. Materials and methods Cell culture Hippocampal microisland cultures were prepared from newborn Sprague Dawley rats (1 2 day postnatal) as described. 13,16,18 In brief, the hippocampal CA1 CA3 region was dissected out and incubated for 30 min in 0.05% trypsin-edta solution (ph 7.2). After enzyme treatment, tissue blocks were gently triturated with a fire-polished Pasteur pipette. Dissociated cells were plated onto a monolayer of astrocytes grown on microislands, which were constructed by spraying 0.4 mm poly-d-lysine and 0.25 mm collagen onto dried thin layer of agarose on 12 mm round glass coverslips. Some lowdensity cultures were prepared by plating neurons on a monolayer of astrocytes without microislands. For Banker-type of culture, neurons were directly plated on poly- D-lysine/collagen-coated glass coverslips and turned upside-down to face a monolayer of astrocytes. 26 The coverslips reside in 24-well plates. The culture medium contained 500 ml of MEM (GIBCO), 5% FBS (HyClone), 10 ml of B-27 supplement (GIBCO), 100 mg of NaHCO 3,

1310 20 mm D-glucose, 0.5 mm L-glutamine, and 25 U/ml penicillin streptomycin (GIBCO). Cells were maintained in a 5% CO 2 culture incubator at 371C. Transfection Hippocampal microisland cultures were transfected with EGFP (Clontech) or GFP-tagged cdna plasmids using a substantially modified Ca 2+ -phosphate transfection method from the Clontech CalPhost transfection protocol. The detailed procedures are as the following: Step 1. Transfer selected coverslips containing microisland cultures from their original well to a new 24-well plate with 0.5 ml prewarmed transfection medium, which is similar to the culture medium except that it is serum free. Return both the original and the new plate to the incubator. Step 2. Prepare the DNA/Ca 2+ -phosphate precipitate using Clontech CalPhost Mammalian Transfection Kit (BD Bioscience, Palo Alto, CA, USA), according to the instruction. Typically, we use about 1 mg cdna, 3.1 ml 2M CaCl 2, and 25 ml 2 HBS solutions (after adjusted with H 2 O, the total volume is 50 ml) per 12 mm round coverslip for transfection. One important modification of the procedure is that when mixing solution A (cdna Ca 2+ H 2 O) with solution B (2 HBS), add about 1/8 volume of solution A at a time into solution B by quickly pipetting several times and gently vortexing for only 2 3 s (VWR MV1 mini vortexer, speed at 600 rpm). The mixed solution will be kept at room temperature for 20 min without any further vortexing to form fine particles of precipitate. Step 3. Add the DNA/Ca 2+ -phosphate suspension solution dropwise to each coverslip (50 ml/coverslip). Then, incubate cells in the presence of the precipitate for 1to3hin5%CO 2 culture incubator at 371C. At the end of incubation, if examined under a microscope, the precipitate should be homogenous and look like a cover of snow all over the field (see Figure 1b and c). Step 4. After incubation, the precipitate should be dissolved to reduce cell toxicity, which is achieved by incubating the precipitate with transfection medium preequilibrated in a 10% CO 2 incubator for 15 20 min. It is important to note that the transfection plate, after adding the 10% CO 2 -equilibrated medium, should be returned to 5% CO 2 incubator rather than the 10% CO 2 incubator. We found a higher level of toxicity if the neurons were placed into 10% CO 2 incubator. Under microscopic examination at the end of 15 20 min incubation, the precipitate should largely disappear. Step 5. Transfer the transfected coverslips back to their original wells containing the original culture medium. This will keep the cell toxicity to a minimal level. EGFP and GFP-CREB plasmids were from Clontech. GFP-endophilin was a gift from Dr Jifang Zhang (University of Pennsylvania). GFP-VAMP was a gift from Dr Richard Scheller (Genetech). GFP-Rab3A was a gift from Dr Ronald Holz (University of Michigan). Electrophysiology Whole-cell recordings were performed by using Multiclamp 700A patch-clamp amplifiers (Axon Instruments, Foster City, CA, USA). Patch pipettes were pulled from borosilicate glass (Warner, Connecticut) and fire-polished with a microforge (Narishige, Japan). Bath solution contained 128 mm NaCl, 30 mm D-glucose, 25 mm HEPES, 5 mm KCl, 2 mm CaCl 2, and 1 mm MgCl 2 (ph 7.3, adjusted with NaOH). Pipette solution contained 147 mm KCl, 5 mm Na 2 -phosphocreatine, 2 mm EGTA, 10 mm HEPES, 4 mm MgATP, and 0.5 mm Na 2 GTP (ph 7.3, adjusted with KOH). Data were acquired with pclamp 9 software, sampled at 10 khz, and filtered at 1 khz (Axon Instruments). FM Imaging Imaging experiments were performed on a Zeiss inverted microscope (Axiovert 200) equipped with an ORCA-ER CCD camera (Hamamatsu, Japan) and DG-5 fast wavelength switcher (Sutter Inc., CA, USA). Fluorescence images were acquired with Simple PCI imaging software (Compix Inc., Pittsburg), and processed with Photoshop. Coverslips were stained in bath solution containing 10 mm FM 4-64 and 40 mm KCl solution for 2 min. After collecting FM staining images, neurons were subjected to destaining in 40 mm KCl solution for 6 min. Images obtained after destaining were used for background subtraction. Acknowledgements We thank Dr Cheryl Keller for critical reading of the manuscript. This work was supported by a start-up fund from Penn State University and the Penn State Life Science Consortium Innovative Biotechnology Research Fund (GC). References 1 Craig AM. Transfecting Cultured Neurons. MIT Press: Cambridge, MA, 1998. 2 Washbourne P, McAllister AK. Techniques for gene transfer into neurons. Curr Opin Neurobiol 2002; 12: 566 573. 3 Xia Z, Dudek H, Miranti CK, Greenberg ME. Calcium influx via the NMDA receptor induces immediate early gene transcription by a MAP kinase/erk-dependent mechanism. J Neurosci 1996; 16: 5425 5436. 4 Micheva KD, Buchanan J, Holz RW, Smith SJ. Retrograde regulation of synaptic vesicle endocytosis and recycling. Nat Neurosci 2003; 6: 925 932. 5 Chen WG et al. Upstream stimulatory factors are mediators of Ca 2+ -responsive transcription in neurons. J Neurosci 2003; 23: 2572 2581. 6 Holz RW et al. A pleckstrin homology domain specific for phosphatidylinositol 4, 5-bisphosphate (PtdIns-4, 5-P2) and fused to green fluorescent protein identifies plasma membrane PtdIns-4, 5-P2 as being important in exocytosis. J Biol Chem 2000; 275: 17878 17885. 7 Graham FL, van der Eb AJ. A new technique for the assay of infectivity of human adenovirus 5 DNA. Virology 1973; 52: 456 467. 8 Kohrmann M et al. Fast, convenient, and effective method to transiently transfect primary hippocampal neurons. J Neurosci Res 1999; 58: 831 835. 9 Watanabe SY et al. Calcium phosphate-mediated transfection of primary cultured brain neurons using GFP expression as a marker: application for single neuron electrophysiology. Neurosci Res 1999; 33: 71 78. 10 Shi SH et al. Rapid spine delivery and redistribution of AMPA receptors after synaptic NMDA receptor activation. Science 1999; 284: 1811 1816.

11 Janson CG et al. Viral-based gene transfer to the mammalian CNS for functional genomic studies. Trends Neurosci 2001; 24: 706 712. 12 Potter DD, Landis SC, Furshpan EJ. Adrenergic cholinergic dual function in cultured sympathetic neurons of the rat. Ciba Found Symp 1981; 83: 123 138. 13 Bekkers JM, Stevens CF. Excitatory and inhibitory autaptic currents in isolated hippocampal neurons maintained in cellculture. Proc Natl Acad Sci USA 1991; 88: 7834 7838. 14 Tong G, Jahr CE. Multivesicular release from excitatory synapses of cultured hippocampal neurons. Neuron 1994; 12: 51 59. 15 Rao A, Cha EM, Craig AM. Mismatched appositions of presynaptic and postsynaptic components in isolated hippocampal neurons. J Neurosci 2000; 20: 8344 8353. 16 Chen G, van den Pol AN. Multiple NPY receptors coexist in preand postsynaptic sites: inhibition of GABA release in isolated self-innervating SCN neurons. J Neurosci 1996; 16: 7711 7724. 17 Deng LB, Chen G. Cyclothiazide potently inhibits gammaaminobutyric acid type A receptors in addition to enhancing glutamate responses. PNAS 2003; 100: 13025 13029. 18 Chen G, Harata N, Tsien RW. Paired-pulse depression of unitary quantal amplitude at single hippocampal synapses. PNAS 2004; 101: 1063 1068. 19 Chen Y et al. Formation of an endophilin-ca 2+ channel complex is critical for clathrin-mediated synaptic vesicle endocytosis. Cell 2003; 115: 37 48. 20 Furshpan EJ, Landis SC, Matsumoto SG, Potter DD. Synaptic functions in rat sympathetic neurons in microcultures. I. Secretion of norepinephrine and acetylcholine. J Neurosci 1986; 6: 1061 1079. 21 Jahn R, Lang T, Sudhof TC. Membrane fusion. Cell 2003; 112: 519 533. 22 Sudhof TC, Scheller RH. Mechanism and regulation of neurotransmitter release. In: Cowan WM, Sudhof TC, Stevens CF (eds). Synapses. The Johns Hopkins University Press: Baltimore, MD, 2000, pp 177. 23 West AE, Griffith EC, Greenberg ME. Regulation of transcription factors by neuronal activity. Nat Rev Neurosci 2002; 3: 921 931. 24 Blanpied TA, Scott DB, Ehlers MD. Dynamics and regulation of clathrin coats at specialized endocytic zones of dendrites and spines. Neuron 2002; 36: 435 449. 25 Haas K et al. Single-cell electroporation for gene transfer in vivo. Neuron 2001; 29: 583 591. 26 Goslin K, Asmussen H, Banker G. Rat hippocampal neurons in low-density culture. In: Banker G, Goslin K (eds). Culturing Nerve Cells. MIT Press: Cambridge, MA, 1998. 1311