Comparative evaluation of fresh, fixed, and cryopreserved solid tumor cells for reliable flow cytometry of DNA and tumor associated antigen.

Five different protocols for the short-term preservation of cells used for multiparameter flow cytometric assay of tumour associated antigens (TAA) and DNA were assessed in cell suspensions prepared by mechanical disaggregation of 15 gynecological tumors. The protocols at 4 degrees C were 1) storage in buffer, 2) storage in 50% methanol, and 3) storage in buffer after formalin fixation. Tissues were also cryopreserved as cell suspensions and tissue blocks. When the TAA expression and DNA histograms of the preserved cells were compared with those in fresh cell suspensions, cryopreservation was found to be the best method: TAA expression was well preserved and there was a good correlation between TAA expression and the quality of the DNA histograms, respectively, in fresh and cryopreserved cells (RS: 0.82-0.91, P less than 0.001 for all TAAs). The cell suspensions preserved at 4 degrees C all showed a significant increase in background fluorescence (P less than 0.05) and a reduction in the TAA specific fluorescence (P less than 0.011). Methanol fixation was better than buffered formalin for the proteins studied, though both gave significantly worse results than cryopreservation. The quality of these cell suspensions and the correlation with TAA measurements in fresh cell suspensions deteriorated progressively with time, particularly if they were stored more than a week.     

Effects of preservation on wing morphometry of the little brown bat (Myotis lucifugus)

The effects of formalin fixation and subsequent alcohol preservation on various morphometric variables and their derivatives (lifting surface area, wingspan, mass, aspect ratio, wing loading and minimum power speed) and on the results of procedures that estimate lifting surface area of the little brown bat ( Myotis lucifugus ) are identified and quantified. Statisitical analysis demonstrates that the values of all of the examined morphometric variables depends upon the specimen type from which they are obtained (live animal; freshly killed specimen; immediately following formalin fixation; or after 36 weeks in alcohol). Over the short term, the choice of preservation fluid is ot important with respect to determination of the six variables studied. The fixation positin of the wing is an important factor in the determination of all variables except mass. Although originally suggested for study skins and not fluid-preserved specimens, 'intermediate' and 'extended' wing positions are demonstrably better than the conventional 'compressed' position. The estimation procedures of both Pirlot (1977) and Blood & McFarlane (1988) significantly underestimate analogous lifting surface areas determinedby tracing live bats. Smith & Starrett's (1979) procedure was found to yield accurate estimates occasionally: on live animals and preserved specimens with wings fixed in the extended position. Aldridge's (1988) method also yields accurate estimates of lifting surface area, but is limited to those museum specimens where the live or freshly-killed mass is known. Such conclusions permit recommendation of procedures that minimize changes arising through the fixation and preservation process in fluid-preserved museum specimens when compared to the live animal.     

Is formalin fixation and ethanol preservation able to influence in geometric morphometric analysis? Fishes as a case study

Geometric morphometric analysis has increased in the recent years, turning into a powerful tool to explore shape and size variation. Several biological studies use specimens that have been through some kind of preservation, mainly formalin preservation, commonly used in biological collections. This study analyzed the effect of preservation in shape on two fish species: Eucinostomus argenteus and Pomadasys corvinaerformis. Twenty-nine individuals of E. argenteus and twenty-five of P. corvinaeformis were collected, photographed twice, preserved in 10 % formalin for 1 week, and then transferred to 70 % ethanol for 83 days. We evaluated three levels of error: (1) error of landmark digitalization, (2) error of taking the picture and storage in JPEG format, and (3) the formalin and ethanol fixation error using Procrustes ANOVA, Discriminant Analysis, and Principal Component Analysis. Significant difference between treatments was observed on both species with Procrustes ANOVA and Discriminant Analysis. In addition, Principal Component Analysis showed a separation between groups of treatment on both species. These results represent the first evidence of preservation effects in studies of geometric morphometrics and show that according to the statistical test utilized, the fixation could affect the shape variations in different ways and could lead the researcher to false results or wrong conclusions. Other methods to explore the shape variation of organisms previously fixed should be tested in order to assess their influence in geometric mophrometric studies.     

Effects of tissue preservation on murine bone mechanical properties

Murine bone specimens are used extensively in skeletal research to assess the effects of environmental, physiologic and pathologic factors on their mechanical properties. Given the destructive nature of mechanical testing, it is normally performed as a terminal procedure, where specimens must be preserved without affecting their mechanical properties. To this end, we aimed to study the effects of tissue preservation (freezing and formalin fixation) on the elastic and viscoelastic mechanical properties of murine femur and vertebrae. A total of 120 femurs and 180 vertebral bodies (L3–L5) underwent non-destructive cyclic loading to assess their viscoelastic properties followed by mono-cyclic loading to failure to assess their elastic properties. All specimens underwent re-hydration in 0.9% saline for 30 min prior to mechanical testing. Analysis indicated that stiffness, modulus of elasticity, yield load, yield strength, ultimate load and ultimate strength of frozen and formalin-fixed femurs and vertebrae were not different from fresh specimens. Cyclic loading of both femurs and vertebrae indicated that loss, storage and dynamic moduli were not affected by freezing. However, formalin fixation altered their viscoelastic properties. Our findings suggest that freezing and formalin fixation over a 2-week period do not alter the elastic mechanical properties of murine femurs and vertebrae, provided that specimens are re-hydrated for at least half an hour prior to testing. However, formalin fixation weakened the viscoelastic properties of murine bone by reducing its ability to dissipate viscous energy. Future studies should address the long-term effects of both formalin fixation and freezing on the mechanical properties of murine bone.     

Accuracy of body mass estimates of formalin-preserved fish – a review

This paper highlights possible effects of physical and chemical mechanisms of formalin fixation and preservation on biological tissue and reviews the consequent potential inaccuracies on estimates of body mass of small fishes fixed and preserved in formalin. Twenty-six papers including 65 independent experiments with 35 species which examine effects of formalin on body mass estimates on small fishes are included. The effect of the formalin on the specimens depends on the salinity of the water used to dilute the commercial formalin (usually 1:9 formalin: water) before being used to fixate and preserve fish. Mean wet body mass of the specimens from the studies using seawater or fresh water diluted formalin deceases by 13% and increases by 7%, respectively, from before to after being immersed in formalin. The same trend is found with condition factor in the few papers that report this parameter. Body length decreases on average by c. 2% in fixated and preserved fish regardless of whether the formalin is diluted in seawater or fresh water.     

Effects of fixation for immunohistochemistry on tissue concentrations of substance P and somatostatin determined by radioimmunoassay

Concentrations of substance P and somatostatin were measured in preparations of the myenteric plexus (plus longitudinal muscle) of the guinea-pig ileum after fixation and processing for immunohistochemistry and compared with concentrations measured in fresh tissue. Two fixative solutions were used: (i) 4% formalin in phosphate buffer (0.1 M, pH 7.0); and (ii) a mixture of aqueous picric acid with 2% formalin in phosphate buffer (0.1 M, pH 7.0). Tissues were extracted in boiling aqueous acetic acid (2.0 M) either immediately after fixation and processing or after storage for up to four weeks in phosphate-buffered saline (PBS) with or without sodium azide. The concentrations of substance P and somatostatin in these extracts were measured by radioimmunoassay and compared to the concentrations in extracts of fresh tissue. The concentration of substance P in fixed tissue was the same as that found in fresh tissue, whereas the concentration of somatostatin in fixed tissue was half that found in fresh tissue (P<0.01). If the tissue was not subjected to the extensive washing for immunohistochemistry, somatostatin concentrations in fresh and fixed tissue were not significantly different. The concentration of substance P did not change on storage of the fixed tissue in PBS, either with or without sodium azide. The concentration of somatostatin decreased on storage of the fixed tissue in PBS over four weeks to 40% of its original value, but the presence of sodium azide maintained the concentration at 60% at four weeks. Neither fixative solution interfered with the radioimmunoassay except at very high concentrations. Fixation for 24h gave the highest estimates of each of the peptides. It is concluded that fixation can be a useful alternative to freezing for preservation of peptides in tissue for radioimmunoassay.     

Cost/benefit and the effect of sample preservation procedures on quantitative patterns in benthic ecology

Some benthic assemblages studies have tested the effects of different preservation procedures on biomass, but their influence on quantitative patterns (number of species and abundance) is still unclear. We evaluated the influence of two sample preservation procedures on quantitative patterns in benthic ecology. Ten sampling points were systematically interspersed on two types of sediment (sandy and muddy). At each sediment type, samples from five sampling points were fixed in 10% formalin, and the other five points were preserved in 70% ethanol (without previous fixation). Three replicates were collected at each sampling point, and samples were washed in 0.5 mesh size and sorted in laboratory. A cost/benefit analysis was performed considering the washing time in laboratory and the costs of substances. A total of 1970 individuals were collected (muddy sediment: 132; sandy sediment: 1838), belonging to 121 taxa (muddy: 49; sandy: 83). Assemblages preserved in ethanol were composed of 795 individuals and 80 taxa, while those fixed with formalin had 1173 individuals and 94 taxa. Polychaeta predominated as the most abundant group for both preservation procedures. For the whole benthic community, significant differences occurred only between sediment types. Significant differences in the number of individuals of polychaetes were observed for the different preservation procedures in sandy sediment. Ethanol has the best cost/benefit ratio in both sediment types due to additional costs to attend safety requirements for formalin-fixed samples. Further studies should evaluate how quantitative patterns are affected by exposure time of preservation, anesthesia interaction, and morphological deformations (e.g. impossibility of identification).     

Effects of fixation and preservation conditions on immunohistochemical profiles of the skeletal muscle fibers in Japanese macaques

Effects of fixation and preservation conditions of muscle tissues on immunohistochemical profiles are investigated. Samples of the hind limb and epaxial muscles were removed from 4 adult female Japanese macaques (Macaca fuscata) fixated with 10% formalin and preserved in the same solution under different conditions for 6 months to 4 years and 6 months. Sections were stained with indirect immunofluorescence and avidin-biotin peroxidase complex methods using an antibody against fast myosin (Mouse Monoclonal Anti-skeletal Myosin-Fast, clone MY-32, Sigma) as a primary antibody. Clear responses to the antibody were demonstrated in the samples from the specimens fixated by injection or immersion with 10% formalin and preserved in the same solution for 6 months to 1 year and 6 months. Distribution patterns of the fibers reacting to the antibody coincided with that of the fast twitch fibers determined using enzyme-histochemical techniques in these samples. Clear responses to the antibody were not demonstrated in the samples from the specimen repeatedly rinsed in water for gross anatomical dissections during the preservation period. The results of this study warrant applications of immunohistochemical techniques to the study of fiber type composition in muscle samples from specimens fixated with formalin and preserved in the same solution for a long term.     

Shrinkage of Prochilodus lineatus (Valenciennes, 1847) larvae preserved in either ethyl‐alcohol or formalin in relation to their developmental stage and feeding condition

The effects of preservation in 95% ethyl‐alcohol and 5% formalin were analysed for 3 months on standard length of Prochilodus lineatus larvae from hatching to the end of the flexion process. Unyolked stages were raised under two feeding regimes: unfed and daily fed. All developmental stages that were preserved in formalin as well as the yolked and flexion‐postflexion larvae stored in alcohol shrank significantly (2–6%). In contrast, unyolked preflexion larvae showed a slight but significant enlargement after storage in alcohol (1%). Shrinkage of preflexion stages was 2.5% higher when stored in formalin, while both preservative agents caused similar shrinkage in flexion‐postflexion larvae (ca. 3%). Shrinkage levels after storage in alcohol were dependent on live length, decreasing or increasing with increasing length in yolked and flexion‐postflexion larvae, respectively. The feeding regime did not affect length changes after preservation in either preservative agent.     

Changes in body dimensions of larval Chaoborus in ethanol and formalin

The effects of fixation and preservation in formalin and ethanolon various body size measurements of the planktonic midge, Chaoborus,were determined. The magnitude of the observed change variedwith the species and the fixative/preservative used. Failureto consider changes in body dimensions induced by fixation andpreservation may significantly bias the estimation of body weightfrom the length of preserved specimens.     

Preserving young‐of‐the‐year Perca fluviatilis in ethanol,formalin, or in a frozen state and the consequences for measuring morphometrics

As it is often not possible to immediately analyse individuals sampled in the field, captured fish are preserved and stored for later investigation. The objective of this study was to assess the effects of various preservation methods on subsequent changes in morphometric parameters while also providing correction factors to re‐calculate the original body dimensions when sampled fish are measured at a later date. In this study, juvenile perch (Perca fluviatilis, 66.5 ± 10.2 mm total length) were measured directly after capture, then either frozen at ?20°C, preserved in 70% ethanol, or in 4% formalin. They were again measured after 24 h, 3 days, 7 days, and thereafter on a weekly basis for 8 weeks. Ethanol‐preserved perch greatly decreased length and weight; formalin preservation also led to a comparable length reduction, but increased the weight of the perch. In contrast, frozen perch showed less shrinkage and only moderate weight loss. Of the three preservation methods, freezing clearly caused the fewest distortions. Hence, freezing is recommend as the most preferable preservation method, especially in multi‐disciplinary studies on fish ecology; for all other preservation methods the correction factor over time will have to be pre‐determined for each species and size class.     

Shrinkage of Gobiocypris rarus, Procypris rabaudi, and Sinilabeo rendahli preserved in formalin

Length measurements of preserved fishes are necessary in many types of fish surveys because logistics often do not allow for fish measurement immediately after catch. If the fixative causes significant shrinkage, then the preserved lengths cannot be directly used to indicate accurate live lengths. The objective of this study was to determine how preservation in formalin affects standard length of Gobiocypris rarus larvae (24‐day‐old and newly hatched), larval Procypris rabaudi (4‐day‐old), and larval Sinilabeo rendahli (12‐day‐old). Fishes were measured (to nearest 0.01 mm) and individually fixed in the appropriate formalin solution (2.5% or 5.0% formalin), then re‐measured at 0.5, 1, 3, 7, 14, 30, 45 and 75 days after preservation to follow the time course of shrinkage. Most of the shrinkage occurred within the first half day after preservation. The 5.0% formalin caused a higher relative shrinkage rate than did the 2.5% solution; however, the difference was not statistically significant. In G. rarus, initial shrinkage of newly hatched larvae was higher than that of 24‐day‐old larvae.     

Assessing DNA for fish identifications from reference collections: the good,bad and ugly shed light on formalin fixation and sequencing approaches

Natural history collections are repositories of biodiversity and are potentially used by molecular ecologists for comparative taxonomic, phylogenetic, biogeographic and forensic purposes. Specimens in fish collections are preserved using a combination of methods with many fixed in formalin and then preserved in ethanol for long-term storage. Formalin fixation damages DNA, thereby limiting genetic analyses. In this study, the authors compared the DNA barcoding and identification success for frozen and formalin-fixed tissues obtained from specimens in the CSIRO Australian National Fish Collection. They studied 230 samples from fishes (consisting of >160 fish species). An optimized formalin-fixed, paraffin-embedded DNA extraction method resulted in usable DNA from degraded tissues. Four mini barcoding assays of the mitochondrial DNA (mtDNA) were characterized with Sanger and Illumina amplicon sequencing. In the good quality DNA (without exposure to formalin), up to 88% of the specimens were correctly matched at the species level using the cytochrome oxidase subunit 1 (COI) mini barcodes, whereas up to 58% of the specimens exposed to formalin for less than 8 weeks were correctly identified to species. In contrast, 16S primers provided higher amplification success with formalin-exposed tissues, although the COI gene was more successful for identification. Importantly, the authors found that DNA of a certain size and quality can be amplified and sequenced despite exposure to formalin, and Illumina sequencing provided them with greater power of resolution for taxa identification even when there was little DNA present. Overall, within parameter constraints, this study highlights the possibilities of recovering DNA barcodes for identification from formalin-fixed fish specimens, and the authors provide guidelines for when successful identification could be expected.     

Impact of tissue preservation on collagen fiber architecture

Microarchitectural features of collagen-rich extracellular matrices provide the mechanical foundation for tissue function and exhibit topographical cues that influence cellular behavior including proliferation, migration and protein expression. Preservation of tissue microarchitecture is required for accurate evaluation of tissue characteristics and pathology. It is unclear whether common tissue preservation methods possess equal ability to preserve microarchitecture. We investigated collagen microarchitecture in samples that had been flash frozen, fixed in formalin or preserved in RNAlater®, and which contained both collagen-rich and collagen-sparse regions. Fibrillar collagen organization was characterized using picrosirius red staining and second harmonic generation (SHG) microscopy. Maintenance of collagen fiber characteristics compared to the gold standard of flash freezing depended on both the method of preservation and the local collagen content of the tissue. Both formalin fixation and RNAlater® preserved collagen fiber characteristics similar to flash freezing in collagen-rich areas of the tissue, but not in collagen-sparse regions. Analysis using picrosirius red staining indicated preservation-dependent changes in overall tissue architecture and suprafibrillar organization. Together with considerations of cost, ease of use, storage conditions and ability to use the preserved tissue for RNA or protein analysis, our quantitative characterization of the effects of preservation method on collagen microarchitecture may help investigators select the most appropriate preservation approach for their needs.     

Cytofluorometric nuclear DNA content analysis of breast tissues after frozen section diagnosis

OBJECTIVE: To establish a suitable method for measurement of nuclear DNA content in breast tissues from frozen storage after frozen section diagnosis. STUDY DESIGN: For fundamental research, rat liver samples preserved in a deep freezer were used. Four protocols were used (1. fixation with 70% ethanol followed by naked nuclei preparation; 2. fixation with 10% neutral buffered formalin followed by naked nuclei preparation; 3. preparation for naked nuclei prior to fixation with 70% ethanol; and 4. preparation for naked nuclei prior to fixation with 70% neutral buffered formalin). For clinical research, 13 separate fresh frozen breast tissue samples were analyzed after frozen section diagnosis. One contained a malignant phyllodes tumor (MPT) consisting of 2 components, benign epithelial cells and malignant stromal cells; 3 were benign tumors containing fibroadenoma; and 9 cases were carcinomas, consisting of 5 scirrhous, 3 papillotubular and 1 mucinous. RESULTS: Protocols 1, 2 and 3 were not suitable methods for our purpose because remaining cytoplasm or cohesive nuclei were observed. In protocol 4 the cytoplasm was completely undetectable, and nuclei were suitably separated for nuclear DNA content measurement. Benign epithelial cell component nuclei presented a diploid pattern, and the malignant stromal cell component nuclei indicated a euploid pattern in MPT. All 3 cases of benign constituents in fibroadenoma showed a diploid pattern, as did the 3 carcinoma cases (1 mucinous, 1 scirrhous and 1 papillary). Four scirrhous and 2 papillary carcinomas showed an aneuploid pattern. CONCLUSION: Our findings show that it is possible to measure nuclear DNA content of human frozen storage tissues after frozen section diagnosis.     

The Effect of Antigen Retrieval on Cells Fixed in 10% Neutral Buffered Formalin Followed by Transfer to 70% Ethanol

Fixation in 10% neutral buffered formalin prior to transfer to 70% ethanol for one week has been shown to adequately preserve immunorecognition of PCNA, cytokeratins AE1/AE3 and EGFr. This study investigated whether 12 hrs fixation in 10% NBF plus transfer to 70% ethanol for 4 weeks would similarly preserve immunorecognition to an extent where antigen retrieval (AR) used to reverse the masking effects of fixation on some antigens would not be necessary. Two cell lines, DU145 and SKOV3 were grown on coverslips and fixed either for 684 hrs in 10% NBF or for 12 hrs in 10% NBF which was then replaced with 70% ethanol for 672 hrs. The second experiment had the same design except an additional set of cells were subjected to heat-induced AR concomitantly. PCNA, cytokeratins AE1/AE3, and EGFr (membrane and cytoplasmic) were used to evaluate the effects of immunorecognition. Fixation in 10% NBF for 12 hrs plus transfer to 70% ethanol for 672 hrs did not preserve immunorecognition of PCNA adequately in either cell lines. Cytokeratins immunoreactivity was preserved by transfer to 70% ethanol. Cytoplasmic EGFr antigens were not adversely affected by 10% NBF fixation in either cell line and transfer to 70% ethanol had limited effects. With AR, there was little recovery of PCNA immunorecognition on cells fixed in only 10% NBF, but almost complete recovery for cells transferred to 70% ethanol. For cytokeratins there was complete recovery of immunorecognition either with only 10% NBF or 12 hrs plus transfer to 70% ethanol. For EGFr, AR resulted in complete loss of immunorecognition following either treatment. This study indicated that 12 hrs of fixation in 10% NBF plus transfer to 70% ethanol for 4 weeks with AR resulted in recovery of immunorecognition for PCNA and cytokeratins, but standard methods of AR caused loss of immunorecognition of EGFr.     

Effect of conservation of fish in formalin and ethanol on length‐weight relationships and condition factor in Tlaloc labialis (Günther, 1866)

The effect of the preserver on the length‐weight relationship and condition factor were estimated using 50 specimens of Tlaloc labialis; a small stream fish from the Grijalva River basin, México. They were measured at different stages of preservation: freshly collected, fixed in 10% formalin and preserved in 70% ethanol for over a 2 years period. All fish decreased in weight and condition factor, with very significant differences compared to the fresh material. The value of the coefficient (b) increased and stabilized from the second month of preservation in ethanol.     

Evaluation of fixative composition,fixative storage,and fixation duration on the fine structure and volume of root-cell nucleoli

Summary— The effect of various combinations of three fixative compositions (glutaraldehyde buffered in veronal acetate, cacodylate, and piperazine-N, N'-bis[2-ethanesulfonic acid]—PIPES], two fixative storage times (fresh vs 6 weeks), and two fixation durations (3 h vs 9 days) on nucleolar fine structure and nucleolar volume in three root cell-types of oat seedlings (Avena sativa L, cv Seger) were evaluated. All fixatives show overall good preservation of fine structure. Nucleolar components are distinct and well delineated in cells fixed in solutions buffered with either cacodylate or veronal acetate; the components are more condensed when preserved in fixative buffered with PIPES. Nucleolar volume is greatest in cells fixed in the cacodylate fixative, and smallest in those preserved in the PIPES fixative. Among the treatments tested, the PIPES fixative evidently best maintains nucleolar volume. Distracting particulate deposits are abundant on nuclei and nucleoli in cells preserved in the veronal-acetate fixative. Contrary to common assumptions, aging of buffered fixative at room temperature for 6 weeks seems to affect neither the general quality of cellular preservation nor the pH of the fixatives, although nucleolar volume is reduced by such treatment. Long-period fixation (9 days) results in destruction of membrane integrity (mitochondria, plastids, ER), and shrinkage of organelles from the cytoplasm. Nucleolar volume is reduced with prolonged fixation.     

The search for an optimal DNA, RNA, and protein detection by in situ hybridization, immunohistochemistry, and solution-based methods

Clinical trials and correlative laboratory research are increasingly reliant upon archived paraffin-embedded samples. Therefore, the proper processing of biological samples is an important step to sample preservation and for downstream analyses like the detection of a wide variety of targets including micro RNA, DNA and proteins. This paper analyzed the question whether routine fixation of cells and tissues in 10% buffered formalin is optimal for in situ and solution phase analyses by comparing this fixative to a variety of cross linking and alcohol (denaturing) fixatives. We examined the ability of nine commonly used fixative regimens to preserve cell morphology and DNA/RNA/protein quality for these applications. Epstein-Barr virus (EBV) and bovine papillomavirus (BPV)-infected tissues and cells were used as our model systems. Our evaluation showed that the optimal fixative in cell preparations for molecular hybridization techniques was "gentle" fixative with a cross-linker such as paraformaldehyde or a short incubation in 10% buffered formalin. The optimal fixatives for tissue were either paraformaldehyde or low concentration of formalin (5% of formalin). Methanol was the best of the non cross-linking fixatives for in situ hybridization and immunohistochemistry. For PCR-based detection of DNA or RNA, some denaturing fixatives like acetone and methanol as well as "gentle" cross-linking fixatives like paraformaldehyde out-performed other fixatives. Long term fixation was not proposed for DNA/RNA-based assays. The typical long-term fixation of cells and tissues in 10% buffered formalin is not optimal for combined analyses by in situ hybridization, immunohistochemistry, or--if one does not have unfixed tissues--solution phase PCR. Rather, we recommend short term less intense cross linking fixation if one wishes to use the same cells/tissue for in situ hybridization, immunohistochemistry, and solution phase PCR.     

Detection of proliferating cell nuclear antigen in diagnostic histopathology.

We describe the effects of tissue preservation, fixation time, and hydrolytic treatment on the detection of proliferating cell nuclear antigen (PCNA) by immunoperoxidase staining with three commercial anti-PCNA antibodies (19A2, 19F4, PC10). Our goal was to provide guidelines for PCNA immunohistochemistry in formalin-fixed, paraffin-embedded specimens. In proliferative cell compartments, nuclear staining was achieved with all three antibodies. In some cases PCNA was also expressed in non-proliferative, histologically normal tissues associated with tumors or other lesions elsewhere. In most autopsy specimens PNCA immunoreactivity was markedly diminished as compared with similar surgical specimens. Incubation overnight with primary antibody at 4 degrees C enhanced PCNA immunoreactivity over incubation at 42 degrees C for 45 min. Pre-treatment with 2 N HCl did not increase staining. Staining with the PC10 antibody was much better preserved than staining with the antibodies 19A2 and 19F4 after prolonged formalin fixation of surgical specimens and in tissues obtained at autopsy. With all three antibodies, however, PCNA immunoreactivity was well preserved during formalin fixation for 8-24 hr and during fixation delays for 8 hr at room temperature. This indicates that PCNA is stable under conditions routinely encountered in diagnostic surgical pathology and facilitates its potential use as a diagnostic proliferation marker.