Immunocytochemistry for inositol lipids

Immunostaining for lipids represents a unique challenge compared to more conventional staining for protein antigens. Aldehyde-based fixatives cross-link primary amines found all in proteins - but absent in most lipids including the inositol lipids. Coupled with the need to permeabilize the cell - that is, disrupt the membrane to allow entry of cytochemical probes - makes any protocol used to detect lipids somewhat destructive to the membranes.

Our first clues came from then Lucocq groups observation that keeping specimens ice-cold could prevent the extraction of membrane lipids on cryosections [5]. We performed an extensive (but not exhaustive) exploration of conditions that would preserve the lipid component of membranes, with the conclusion that different conditions preserve different compartments to different extents [3]. Over the years, we have developed and refined the procedures [4,3,2,1], and an updated protocol is described below, with italics providing brief explanations of how conditions were selected.

1. Bojjireddy N, Botyanszki J, Hammond GRV, Creech D, Peterson R, Kemp DC, et al. Pharmacological and genetic targeting of pPI4KA reveals its important role in maintaining plasma membrane PtdIns4p and PtdIns(4,5)p2 levels. J Biol Chem. American Society for Biochemistry and Molecular Biology; 2014 Jan 10;289(9):6120–32.
2. Hammond GRV, Fischer MJ, Anderson KE, Holdich J, Koteci A, Balla T, et al. PI4P and PI(4,5)P2 are essential but independent lipid determinants of membrane identity. Science. American Association for the Advancement of Science; 2012 Aug 10;337(6095):727–30.
3. Hammond GRV, Schiavo G, Irvine RF. Immunocytochemical techniques reveal multiple, distinct cellular pools of PtdIns4P and PtdIns(4,5)P(2). Biochem J. 2009 Aug 15;422(1):23–35.
4. Hammond GRV, Dove SK, Nicol A, Pinxteren JA, Zicha D, Schiavo G. Elimination of plasma membrane phosphatidylinositol (4,5)-bisphosphate is required for exocytosis from mast cells. J Cell Sci. The Company of Biologists Ltd; 2006 May 15;119(Pt 10):2084–94.
5. Watt SA, Kular G, Fleming IN, Downes CP, Lucocq J. Subcellular localization of phosphatidylinositol 4,5-bisphosphate using the pleckstrin homology domain of phospholipase C delta1. Biochem J. 2002 May 1;363(Pt 3):657–66.

To stain the plasma membrane:

• Remove media and fix cells for 10 min, room temperature in PBS with either 4% formaldehyde + 0.2% glutaraldehyde or 0.2% formaldehyde + 0.2% glutaraldehyde. The inclusion of glutaraldehyde improves cross-linking of primary amine containing lipids such as PtdEtn or PtdSer, so stabilizing the membrane. We have found it can be advantageous to reduce the formaldehde concentration to prevent osmotic swelling of the cells after adding the fixative. Use EM-grade fixatives. We use 16% MeOH-free formaldehyde stock solutions.
• Wash the cells three times with freshly made 10 mg/ml sodium borohydride in PBS. This step reduces glutaraldehyde-induced autofluorescence by 1-2 orders of magnitude. It is essential when lower concentrations for formaldehyde are used. This step also has the added bonus of restoring antigenicity of many protein targets.
• Rinse twice with PBS. This removed the residual sodium borohydride which can inactivate fluorophores.
• Place the vessel (slide, coverslips etc) on a cold, flat surface (typically a metal block) in an ice bath and allow to chill for > 2 min. Ensure all solutions from here on are chilled to be ice-cold
• Optional: Incubate 30 min on ice in antibody solution containing  an empirically determined concentration of GST-tagged lipid binding domain
• Incubate 30-60 min on ice in antibody solution (0.5% saponin, fresh from frozen 5% aliquots stored at –20˚C in PIPES-BS; 10% normal goat serum; PIPES-BS: 137 mM NaCl, 2.7 mM KCl, 20 mM PIPES to pH 6.8 with NaOH) containing empirically determined dilutions of lipid antibodies, such as anti-PtdIns(4,5)P2 clone 2C11 (Echelon biosciences Z-P045, 1:100), anti-PtdIns4P (Echelon biosciences Z-P005, 1:200) or anti-GST if previous optional step was employed (Millipore AB3282, 1:100). The antibodies are generally specific enough not to require a pre-blocking step; inclusion of 10% goat serum appears sufficient to avoid non-specific binding in our hands. We use goat serum because all of our secondaries are relied in goat; ensure you are careful to match blocking sera with your antibody sera! Saponin is used as the permeabilizing agent, and is the cheap stuff from Quillaja bark. 0.5% is an empirically determined optimum amount - it could vary for your cells. We have no idea why the relatively “dirty” agent works so well when purer, more homogeneous detergents like digitonin (an active ingredient of saponins) fail - we screened several detergents in developing this assay, and this worked best! PIPES is buffer is used, but Tris or HEPES at pH 7.4 do work as well. Avoid PBS - the phosphate (~12 mM) acts as a competitor for many of the lipid probes; 2C11 loses > 90% of its signal when used in this buffer.
• Briefly rinse in ice-cold PIPES-BS
• Incubate 30-45 min on ice in antibody solution containing appropriate fluorescent secondary antibodies. Keep times to a minimum to achieve adequate signal. We have found that PtdIns4P staining degrades the longer the incubation period.
• Rinse four times in ice-cold PIPES-BS
• Fix 10 min in ice-cold 4% formaldehyde then 5 min at room temp. Though we never strictly tested whether this step is truly necessary, the idea is to fix the antibodies in place to there is no disruption once the slides/coverslips are warmed up.
• Rinse three times in PBS and once in ultra-pure water (to remove salt)
• Mount in ProLong Gold ± DAPI. Slides can be imaged immediately, or for best results allow mountain to cure for 24 hours at room temp. before sealing with nail varnish.

To Stain endomembranes (all at room temperature):

• Remove media and fix cells for 10 min, room temperature in PBS with either 4% formaldehyde. The inclusion of glutaraldehyde reduces staining of internal membranes and is best avoided.
• Wash the cells three times with 50 mM ammonium chloride or glycine PBS. The primary amine quenches any unreacted aldehyde groups in the sample, eliminating a source of non-specific binding and autofluorescence.
• Permeabilize 5 min, room temp. in PIPES-BS containing 20 µM digitonin. The concentration was empirically determined. A brief permeabilization step preserves lipid staining much better than including it throughout the protocol.
• Rinse three times in PIPES-BS
• Optional: Incubate 30 min in antibody solution containing  an empirically determined concentration of GST-tagged lipid binding domain
• Incubate 30-60 min on ice in antibody solution (10% normal goat serum; PIPES-BS: 137 mM NaCl, 2.7 mM KCl, 20 mM PIPES to pH 6.8 with NaOH) containing empirically determined dilutions of lipid antibodies, such as anti-PtdIns4P (Echelon biosciences Z-P005, 1:200) or anti-GST if previous optional step was employed (Millipore AB3282, 1:100). The antibodies are generally specific enough not to require a pre-blocking step; inclusion of 10% goat serum appears sufficient to avoid non-specific binding in our hands. We use goat serum because all of our secondaries are relied in goat; ensure you are careful to match blocking sera with your antibody sera! PIPES is buffer is used, but Tris or HEPES at pH 7.4 do work as well. Avoid PBS - the phosphate (~12 mM) acts as a competitor for many of the lipid probes.
• Briefly rinse in PIPES-BS
• Incubate 30-45 min in antibody solution containing appropriate fluorescent secondary antibodies.
• Rinse four times in ice-cold PIPES-BS
• Fix 10 min in 4% formaldehyde.
• Rinse three times in PBS and once in ultra-pure water (to remove salt)
• Mount in ProLong Gold ± DAPI. Slides can be imaged immediately, or for best results allow mountain to cure for 24 hours at room temp. before sealing with nail varnish.