TIBS 25 – MARCH 2000 99 0968 – 0004/00/$ – See front matter © 2000, Elsevier Science Ltd. All rights reserved. PII: S0968-0004(99)01535-2 SEVERAL METHODS HAVE been devel- oped to address questions concerning the in vivo regulation of the interactions between cis elements and trans-acting factors in the context of chromosome structure and nuclear organization. These include in vivo footprinting, chem- ical and light-induced crosslinking and immunocytochemistry. The ability to provide direct evidence that given regu- latory proteins are associated ‘in time and space’ with specific genomic regions is a key determinant of the merits of the various techniques. In particular, those methods that use in vivo fixation com- bined with immunoprecipitation appear to have an advantage over other ap- proaches, such as in vivo footprinting, which provide indirect information about ‘occupancy’ of a given site by a nu- clear protein in chromatin. In addition, regulatory proteins that structurally re- semble histones are not detected by the latter method. The formaldehyde fixation and chro- matin immunoprecipitation approach (X-CHIP) offers the ability to detect any protein at its in vivo binding site di- rectly. In particular, proteins that are not bound directly to DNA or that depend on other proteins for binding activity in vivo can be analysed with this method. Macromolecular chromosomal struc- tures in living material, like tissue cul- ture cells or embryos, are fixed very efficiently, and the chromatin is used as a substrate for immunoprecipitation. Antibodies directed against the protein of interest allow immunoselection of all genomic binding sites. At the end, the crosslinking can be fully reversed and the DNA analysed. The chemistry of formaldehyde crosslinking: the advantage of de-crosslinking Formaldehyde is a tight (2 Å) crosslink- ing agent that efficiently produces both protein–nucleic acid and protein–protein crosslinks in vivo. Formaldehyde is a very reactive dipolar compound in which the carbon atom acts as a nucleophilic cen- tre. Amino and imino groups of amino acids (lysines, arginines and histidines) and of DNA (primarily adenines and cy- tosines) readily react with formaldehyde leading to the formation of a Schiff base. This intermediate can further react with a second amino group and condense to give the final DNA–protein complex 1,2 . These reactions take place in vivo within minutes after addition of formaldehyde to living cells or embryos (Fig. 1). A key ad- vantage of the use of formaldehyde as a crosslinking agent is that the crosslinks are fully reversible. This is achieved pri- marily by protonation of imino groups at low pH in aqueous solution. For the identification and the charac- terization of the in vivo DNA targets of a given protein, after immunoprecipi- tation of crosslinked chromatin, the DNA is purified and analysed by conventional methods like Southern-blot hybridi- zation and PCR analyses. Such analysis requires the removal of all proteins from the immunopurified chromatin fraction. To achieve this, the DNA is released from crosslinked material by extensive diges- tion with proteinase K and mild heat treatment, and then purified by standard methods 3 . Specific crosslinking reversal condi- tions are also known for protein analysis in chromatin. It is often of interest to determine whether a given protein is present in the crosslinked chromatin. In this case, low pH and specific, highly de- naturing conditions (e.g. 8 M hydroxy- urea) are used 4 . A simpler alternative in- volves boiling crosslinked chromatin in conventional SDS-PAGE protein gel load- ing buffer for up to 1 hour 5 . Because the amount of the protein of interest is often below the limit of detection by simple Western blotting, total proteins can be labelled in vivo with 35 S, which improves the sensitivity of detection of specific chromatin components after immuno- precipitation 5 . It is technically difficult, however, to determine the efficiency of immunopre- cipitation precisely based on the per- centage of precipitated protein of inter- est. Generally, it is hard to predict how much of this antigen will actually be chromatin-bound, how many sites will be engaged and in which portion of the genome. In addition, certain sites might be more available to the antibody than others. Hence, at the protein level, X- CHIP is a non-quantitative technique. Extent of crosslinking: finding the right balance Efficient fixation of a protein to chro- matin in vivo is crucial for the X-CHIP technique. The extent of crosslinking is probably the most important parame- ter. Two major problems concerning the subsequent immunoprecipitation step should be taken into account: first, an ex- cess of crosslinking can result in the loss of material or reduced antigen availabil- ity in chromatin, or both, and second, the relative sensitivity of the antigen epitopes to formaldehyde. Crosslinking times range between 10 minutes and several hours. Nucleosomal proteins are normally analysed following a crosslinking time 10 minutes 6–8 . Longer exposure to formaldehyde leads to loss of immunoprecipitated material, although the reason for this is not understood. It TECHNIQUES Mapping chromosomal proteins in vivo by formaldehyde- crosslinked-chromatin immunoprecipitation Valerio Orlando Gene regulation is a complex process. Numerous factors appear to be re- quired for the accurate temporal and spatial regulation of each gene. Often these factors are assembled into multiprotein complexes, contributing to specific gene regulation events. Understanding how all these factors are organized in the chromosome and how their function is regulated in vivo is a challenging task. One of the most useful techniques for studying this level of gene regulation is the in vivo fixation by formaldehyde crosslinking of proteins to proteins and proteins to DNA, followed by immunoprecipi- tation of the fixed material. V. Orlando is at the DIBIT HSR Biomedical Scientific Park, Via Olgettina 58, 20132 Milano, Italy. Email: v.orlando@hsr.it