- Email: [email protected]
Improved detection of simian immunodeficiency virus RNA by in situ hybridization in fixed tissue sections: combined effects of temperatures for tissue fixation and probe hybridization
Journal of Virological Methods 99 (2002) 23 – 32 www.elsevier.com/locate/jviromet
Improved detection of simian immunodeficiency virus RNA by in situ hybridization in fixed tissue sections: combined effects of temperatures for tissue fixation and probe hybridization Beth A. Fallert, Todd A. Reinhart * Department of Infectious Diseases and Microbiology, Graduate School of Public Health, Uni6ersity of Pittsburgh, Pittsburgh, PA 15261, USA Received 21 May 2001; received in revised form 26 July 2001; accepted 27 July 2001
Abstract In situ hybridization detection of viral RNAs in formaldehyde-fixed tissue specimens is used frequently to characterize the extent of viral replication within host tissues. The ability to determine the level of expression of viral RNAs in situ is dependent upon many factors including the extent of cross-linking during fixation, the pretreatment regimen utilized to relieve the effects of cross-linking, and the hybridization and wash protocols. In efforts to improve our ability to detect cells infected productively by simian immunodeficiency virus (SIV) in rhesus macaque tissues, the effects of unconventionally high (40 °C) and more standard low (4 °C) temperature fixation in 4% paraformaldehyde/phosphate buffered saline were tested empirically on in situ hybridization signals. In addition, hybridization temperatures ranging between 37 and 75 °C were utilized to determine the optimal hybridization conditions for detection of SIV productively infected cells. Fixation conditions of 40 °C and hybridization conditions of 50 – 55 °C were identified as providing the greatest sensitivity for detecting RNA+ cells and for quantitating the signal per cell, while still allowing antigenic epitopes to be detected by immunohistochemical staining. These data indicate that the signal intensity following in situ hybridization for viral RNAs is dependent upon the combined effects of tissue fixation and in situ hybridization temperatures. © 2002 Elsevier Science B.V. All rights reserved. Keywords: Simian immunodeficiency virus; In situ hybridization; Immunohistochemistry; Formalin fixation; Temperature
1. Introduction
* Corresponding author. Tel.: + 1-412-648-2341; fax: +1412-383-8926. E-mail address: [email protected] (T.A. Reinhart).
In situ hybridization is an extremely versatile and useful method for determining the numbers and locations of individual cells harboring a specific nucleic acid amongst all of the cells in a tissue section. In situ, hybridization is used for
0166-0934/02/$ - see front matter © 2002 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 6 - 0 9 3 4 ( 0 1 ) 0 0 3 7 8 - 0
24
B.A. Fallert, T.A. Reinhart / Journal of Virological Methods 99 (2002) 23–32
many purposes, including the identification of transgene-expressing cells, the analysis of gene expression during development, and quantitation of gene expression during disease processes. A frequent application for which in situ hybridization is used in infectious disease research is the identification of the numbers, locations, and types of cells productively infected by different viruses (Haase et al., 1996; Ozden et al., 1990; Reinhart et al., 1997; Kim and Chae, 2001; Shimakage and Sasagawa, 2001). These data can be used not only to determine whether a pathogen is present in a tissue specimen, but can also provide critical data on the extent to which the pathogen is replicating, where it is replicating, and how strongly its RNAs are being expressed. Accurate data on the extent and location of viral replication in tissues are crucial for determining the effects of local viral replication on host tissues and for identifying the mechanisms by which viral replication leads to pathogenic outcomes. The success of any in situ hybridization detection assay is critically dependent upon multiple factors including the conditions of tissue preservation, tissue section pretreatment, and hybridization and wash conditions, all of which can affect the integrity of the target RNAs, the ability of the probe to hybridize with the RNA, and therefore, the success of the assay (Yang et al., 1999). Typical protocols for the detection of RNAs in tissue sections via in situ hybridization have followed well defined nucleic acid hybridization parameters, which include 50% formamide, 10% dextran sulfate, and a hybridization temperature of 45– 50 °C (Wilcox, 1993). Fixation parameters, including the fixative itself, and the duration and temperature of fixation can affect the outcome of in situ hybridization experiments. Shorter fixation times can lead to improved signals (Wilcox, 1993), and protocols frequently call for fixation at 4 °C, but there is thus far no tissue preservation protocol that is universally used for all experimental systems. The primate lentivirus, simian immunodeficiency virus (SIV), is genetically, structurally, and pathogenically related to human immunodeficiency virus type 1 (HIV-1), which causes immunodeficiency in humans, and can cause diseases
in rhesus macaques that are very similar to those observed in HIV-1 infected humans (Desrosiers, 1990; Whetter et al., 1999). The amount of viral RNA present in plasma in humans (Mellors et al., 1996) and in lymphoid tissues in rhesus macaques (Hirsch et al., 1996) is thought to be proportional to the rate of disease progression. Therefore, reliable and sensitive methods of detecting SIV/HIV1 and cellular mRNAs by in situ hybridization in biopsies or post-mortem tissues are crucial for obtaining an accurate understanding of the virus/ host relationship as it relates to disease. Our interests in determining the extent of SIV replication in fixed tissue sections prompted us to re-examine our fixation and tissue preservation protocol in efforts to improve the sensitivity of the in situ hybridization assay for SIV RNA, while not compromising its specificity. Based on an initial observation that tissues inadvertently fixed at 40 °C provided exceptional in situ hybridization signal, we describe here comparative analyses of the effects of high and low temperature fixation in 4% paraformaldehyde/phosphate buffered saline (4% PF/PBS) on the in situ hybridization signal provided in rhesus macaque lymph node tissue sections after hybridization with a pool of SIV-specific riboprobes.
2. Materials and methods
2.1. Tissue preser6ation, sectioning, and pretreatment The cryopreserved, pathological tissue specimen used for these studies was an inguinal lymph node obtained from a rhesus macaque 2 weeks after intravenous inoculation with the SIV/DeltaB670 isolate (Murphey-Corb et al., 1986). The study that included the animal from which this lymph node was obtained was performed under the approval of the Institutional Animal Care and Use Committee at The University of Pittsburgh. The tissue specimen (:0.5×0.5× 0.25 cm3) was hemisected to facilitate diffusion of the fixative, placed into fresh 4% PF/PBS (pH 7.3; Sigma, St. Louis, MO), and fixed at 4 or 40 °C for 5 h on a rotary shaker. The tissues were then cryoprotected
B.A. Fallert, T.A. Reinhart / Journal of Virological Methods 99 (2002) 23–32
by washing in successive overnight incubations of 1× PBS, fresh 1× PBS, 10% sucrose/PBS, and 20% sucrose/PBS, all at 4 °C with shaking. The specimens were then frozen by immersion for 5– 10 s in methylbutane (Fisher, Pittsburgh, PA) cooled to −65 °C on dry ice, and then stored at − 85 °C. Cryopreserved tissues were pre-warmed at − 20 °C for 2 h before sectioning. Tissues were sectioned on a cryostat at a thickness of 14 mm, thaw mounted at 43 °C for 20 min onto SuperFrost Plus microscope slides (Fisher), and then stored at − 85 °C until use. Mounted tissue sections were post-fixed immediately out of − 85 °C storage in fresh 4% PF/PBS for 20 min, washed for 20 min in 70% ethanol, and dehydrated in graded ethanols. The tissues were then microwaved in citrate buffer (pH 6.0; Sigma), for a total of 10 min at 2 min intervals with power settings between 20 and 60% using a Sharp Carousel microwave (1500 W). Buffer lost due to boiling over was replaced between the 2 min intervals. The slides were then removed from the microwave and allowed to cool for 30 min on the bench, rinsed in water, acetylated in 0.1 M triethanolamine (Sigma) with 0.25% acetic anhydride (Sigma) for 6 min, rinsed in water and dehydrated in graded ethanols.
2.2. In situ hybridization The riboprobes utilized in these studies encompassed sequences from four regions of the SIVmacBK28 molecular clone (Kornfeld et al., 1987) spanning portions of the gag, pol, en6, and nef genes, and included positions 47– 1130, 1676– 3121, 6600– 8286, and 8453– 9267 (Genbank accession no. M19499). Riboprobes were synthesized by in vitro transcription of linearized pGEM4z plasmid templates containing the appropriate insert, using T7 or SP6 RNA polymerase to generate the antisense or corresponding sense probes. In vitro, transcription was performed using the Maxiscript SP6/T7 kit (Ambion, Austin, TX), with plasmid DNA (0.2 mg/ml) and 35S-UTP (0.05 mCi; ICN, Costa Rica, CA). Transcription was allowed to proceed at 37 °C for 3 h and DNAse was added for an additional 20 min.
25
Unincorporated nucleotides were removed by ethanol precipitating twice with 25–50 ug of yeast tRNA as carrier. After re-suspension in RNAsefree water, the probe was heated at 65 °C for 5 min to facilitate re-suspension. Equivalent amounts of the four antisense and sense riboprobes were then pooled, respectively, to a final concentration of 0.5×106 cpm/ml. Following microwave pretreatment, in situ hybridization was carried out essentially as described (Reinhart et al., 1997) in the following hybridization mix: 20 mM Hepes (Sigma) pH 7.2; 1 mM EDTA (BioWhitaker, Walkersville, MD); 1× Denhart’s reagent; 0.1 mg/ml PolyA (Amersham/Pharmacia, Piscataway, NJ); 0.6 M NaCl (BioWhitaker); 100 mg/ml yeast tRNA; 10% dextran sulfate; 50% formamide; and riboprobes at 50 000 cpm/ml. The hybridization mix (5–10 ml) was placed on each tissue section, covered with a siliconized coverslip, and sealed with rubber cement. Hybridizations were performed at temperatures ranging from 37 to 75 °C for : 18 h. Coverslips were then removed under 5× standard saline citrate (SSC) at room temperature and moved progressively through: 5× SSC/10 mM DTT, 42 °C, 30 min; 2 × SSC/50% formamide/ 10 mM DTT, 60 °C, 20 min; riboprobe wash solution (0.1 M Tris, pH 7.5, 50 mM ethylenediaminetetraacetic acid, 0.4 M NaCl), 37 °C, 10 min, twice; riboprobe wash solution containing 12.5 U/ml RNAse T1 and 12.5 mg/ml RNAse A, 37 °C, 30 min; riboprobe wash solution, 37 °C, 15 min; 2 × SSC, 37 °C, 15 min; 0.2× SSC, 37 °C, 15 min. Tissue sections were then dehydrated in graded ethanols containing 0.3 M ammonium acetate, air dried, coated with NTB-2 emulsion (Kodak, Rochester, NY) and exposed at 10 °C for 1 day.
2.3. Enumeration of SIV 6RNA+ cells and in situ hybridization signal per cell SIV vRNA+ cells were counted in each of 20 random microscopic fields viewed through a 40× Plan Apochromat objective. The in situ hybridization signal over individual cells was determined by capturing images with a SPOT camera (Diagnostic Instruments, Sterling Heights, MI) mounted
26
B.A. Fallert, T.A. Reinhart / Journal of Virological Methods 99 (2002) 23–32
on a Nikon E600 fluorescence microscope, through a 60× Plan Apochromat objective after illumination with polarized light directed through an IGS polarizing filter cube (Omega Optical, Brattleboro, VT). Images were captured and analyzed using the Metaview software package (Universal Imaging Corporation, West Chester, PA). Individual cells in captured images were selected using the region tool and then thresholded to include all of the silver grains within that region. The surface area (pixels) of light reflected by the silver grains over each vRNA+ cell was then determined using the measure function in Metaview, and the data values exported to Microsoft Excel.
2.4. Immunohistochemical staining and simultaneous in situ hybridization and immunohistochemical staining Immunohistochemical staining was carried out essentially as described (Reinhart et al., 1997) with either a rabbit polyclonal serum specific for CD3o (Dako, Carpenteria, CA; catalogue no. A0452) or a murine monoclonal antibody specific for CD68 (Dako; clone KP1). Tissue sections were blocked by incubation of sections in 1× PBS containing 5% non-fat dry milk for 1– 3 h at room temperature. Excess blocking solution was then dabbed off and the primary antibodies diluted in PBS/5% non-fat dry milk was added at a dilution of 1:50 for CD68 and 1:100 for CD3o. Antibodies were detected using the avidin– biotin complex method with the Vectastain Elite system (Vector Labs, Burlingame, CA), using 3%,3-diaminobenzidine (Sigma) as the final substrate. Identification of the cell types expressing SIV RNA was performed by combining in situ hybridization with 35S-UTP-labeled riboprobes and immunohistochemical staining with antigen-specific antibodies. Following in situ hybridization, tissue sections were immediately rinsed in 1× PBS prior to initiation of the immunohistochemical staining. Following incubation with the 3%,3diaminobenzidine substrate, sections were then dehydrated in graded ethanols containing 0.3 M ammonium acetate, air dried and coated with NTB-2 emulsion and exposed at 10 °C for 1 day.
Antigen-positive cells were identified by the deposition of a brown precipitate and RNA+ cells were identified by collections of silver grains over individual cells.
3. Results
3.1. In situ hybridization for SIV+ cells: detection of 6RNA+ cells depends upon fixation and hybridization temperatures Equal portions of an inguinal lymph node were obtained from an SIV-infected rhesus macaque 2 weeks post-infection. At this time, referred to as the acute phase of infection, viral replication is widespread and peaks (Hirsch et al., 1996; Reimann et al., 1994), usually subsiding to lower levels very soon thereafter. These specimens were fixed in 4% PF/PBS for 5 h on a rotary shaker either at 4 or 40 °C, and then cryoprotected and snap frozen. In situ hybridization detection of SIV productively infected cells with a pool of SIV-specific riboprobes in these two portions of lymph node at different hybridization temperatures ranging from 37 to 75 °C yielded drastically different results. Representative high-power microscopic fields from sections of the tissue fixed at 40 °C and hybridized at different temperatures are shown in Fig. 1, and demonstrate that hybridization at 50–55 °C resulted in very strong signal, whereas hybridization at lower or higher temperatures resulted in reduced signal intensity. As shown in Fig. 2, the numbers of vRNA+ cells/ mm2 in the portion of tissue fixed at 40 °C increased from 0.4 after hybridization at 37 °C to 9.9 after hybridization at 75 °C, peaking at 41.3 after hybridization at 50 °C. The numbers of vRNA+ cells obtained after hybridization at 50, 55, and 60 °C were all between 28.5 and 41.3. In contrast, the numbers of vRNA+ cells in the portion of tissue fixed at 4 °C increased steadily with hybridization temperatures between 37 and 60 °C, peaking at 36.2 vRNA+ cells per mm2 after hybridization at 60 °C. Parallel in situ hybridizations were performed on the same slide with a pool of SIV-specific sense control riboprobes and in all instances the sense control
B.A. Fallert, T.A. Reinhart / Journal of Virological Methods 99 (2002) 23–32
probes provided only low level background in situ hybridization signal with no RNA+ cells detected on the entire tissue section (data not shown). As an additional control, in situ hybridization was performed at 50 °C on tissue sections from an uninfected macaque inguinal lymph node fixed at 40 °C, and for both the antisense and sense probes, no RNA+ cells were observed over the entire tissue section (data not shown). In summary, tissue fixation and in situ hybridization temperatures of 40 and 50 °C, respectively, were identified as providing strong autoradiographic signals without an accompanying loss of specificity.
27
3.2. In situ hybridization for cells producti6ely infected by SIV: signals per cell depend upon fixation and hybridization temperatures In situ hybridization with radioactive probes, coupled with emulsion autoradiography, allows the intensity of the signal to be quantitated by counting individual silver grains or their collective surface areas over individual RNA+ cells (Cox et al., 1984; Haase et al., 1996; Ozden et al., 1990; Reinhart et al., 1997). We therefore, utilized a quantitative image capture and analysis system to measure the signal intensity over individual cells in the same tissue sections used to enumerate the
Fig. 1. In situ hybridization detection of SIV vRNA+ cells. Representative photomicrographs are shown for sections obtained from the lymph node specimen fixed at 40 °C in fresh 4% PF/PBS and hybridized with a pool of SIV-specific riboprobes. The hybridization temperatures correspond to (A) 37 °C, (B) 45 °C, (C) 50 °C, (D) 55 °C, (E) 60 °C, and (F) 75 °C. Exposure times were 1 day. Original magnification, 300 × .
28
B.A. Fallert, T.A. Reinhart / Journal of Virological Methods 99 (2002) 23–32
Fig. 2. Enumeration of SIV vRNA+ cells following in situ hybridization. The numbers of vRNA+ cells were counted as described in Section 2 for lymph node specimens fixed in fresh 4% PF/PBS at either 4 or 40 °C. The data are presented as vRNA+ cells/mm2, with the gray bars representing the two independent experiments performed on tissue fixed at 40 °C and the white bars representing the two independent experiments performed on tissue fixed at 4 °C. Asterisks indicate the detection of no vRNA+ cells in 20 random high power fields. ND, not determined.
vRNA+ cells/mm2. Tissue sections were illuminated with epipolarized fluorescent light and microscopic images of the light reflected by the metallic silver grains were captured with a digital camera. Using the region, threshold, and measure tools of the Metaview software package the surface area of reflected light from individual vRNA+ cells was determined. Whenever possible, at least 100 vRNA+ cells per tissue section were included in these analyses. As shown in Fig. 3, the surface areas of reflected light over individual cells ranged from : 200 to 10 000 pixels/cell. The means and distributions of the surface areas of reflected light per cell indicate that fixation of the tissues at 40 °C and hybridization at 50– 55 °C provided on average the greatest signal intensity
per vRNA+ cell. The percentage of vRNA+ cells with in situ hybridization signals covering greater than 2000 pixels/cell was higher for the tissue fixed at 40 °C as compared to 4 °C, except when hybridized at the unconventionally high temperature of 60 °C, which provided nearly equivalent signals ranging between 33 and 39% (Fig. 3).
3.3. Antigenic epitopes remain a6ailable after fixation of tissues at high temperature A strength of microscope-slide based approaches is not only the ability to detect the RNA expressed by a specific virus, but also the ability to determine the numbers, locations and types of cells present within the tissue section, for both
B.A. Fallert, T.A. Reinhart / Journal of Virological Methods 99 (2002) 23–32 29
Fig. 3. Quantitative image analysis of the in situ hybridization signals per vRNA+ cell. The surface area of polarized light reflected by silver grains over vRNA+ cells was determined as described in Section 2 for lymph node specimens fixed at either 4 °C (grey symbols) or 40 °C (black symbols). Only those tissue sections that provided more than two vRNA+ cells/mm2 (Fig. 2) were examined to determine the in situ hybridization signals per vRNA+ cell; those not examined are indicated by the asterisk. Results are shown from both of the replicate experiments. In the bottom of the figure are the mean surface areas of reflected light per cell and the percentage of cells with \2000 pixels of reflected light per cell. ND, not determined.
30
B.A. Fallert, T.A. Reinhart / Journal of Virological Methods 99 (2002) 23–32
infected and other cells. Identification of the types of cells expressing viral RNAs can be accomplished by coupling immunohistochemical staining for cell-type specific antigens with in situ hybridization for viral RNA (Reinhart et al., 1997). To examine whether fixation at 40 °C rendered antigenic epitopes unavailable for immunohistochemical staining, a standard avidin– biotin complex (ABC) staining approach was applied to this tissue specimen. Immunohistochemical staining for the T-lymphocyte-specific marker, CD3o, and the monocyte/macrophage-specific marker, CD68, was undertaken. In both instances, the antigenic epitope was readily detectable (Fig. 4) and provided robust signal, even after performing in situ hybridization for SIV RNA and immunohistochemical staining for CD68 simultaneously on the same tissue section (Fig. 4C). In this instance, a CD68-negative cell is expressing SIV RNA. Application of control antibody provided only minimal background immunohistochemical staining signal (data not shown).
4. Discussion It was demonstrated that fixation of tissues at 40 °C in fresh 4% PF/PBS and probe hybridization at 50–55 °C provided optimal detection of SIV RNAs in cells infected productively via in situ hybridization. The improved in situ hybridization signal might be due to increased covalent cross-linking by the 4% PF due to high temperature induced increases in the concentration of free aldehyde from the methylene glycol, which predominates in formaldehyde solutions (Fox et al., 1985). The higher fixation and hybridization temperatures, which provided increased in situ hybridization sensitivity without compromising specificity did not destroy the antigenic epitopes on the cell-type specific markers CD3o and CD68, indicating that higher temperature fixation does not grossly reduce the availability of antigenic epitopes for immunohistochemical staining. These data demonstrate an inter-relationship between fixation and hybridization conditions, and identify high temperature (40 °C) fixation
Fig. 4. Immunohistochemical detection of antigenic epitopes in lymph node sections from a tissue specimen fixed at high temperature. Standard avidin – biotin complex (ABC) immunohistochemical staining approaches were used to detect CD3opositive T-lymphocytes (A) or CD68-positive monocytes/macrophages (B). Antigen positive cells are identifiable by the deposition of an insoluble brown product within the cell. Simultaneous immunohistochemical staining for CD68 and in situ hybridization for SIV RNA was performed to identify the target cells infected by SIV (C). Original magnifications: 400 × (A, B) and 600 × (C).
B.A. Fallert, T.A. Reinhart / Journal of Virological Methods 99 (2002) 23–32
conditions as optimal, when the ensuing tissue sections are hybridized at routine temperatures (45 – 55 °C). Higher hybridization temperatures have been demonstrated to be appropriate, when using riboprobes due to the greater stability of RNA/RNA hybrids (Cox et al., 1984), and our determination that : 50 °C is an optimal temperature for in situ hybridization is very similar to that determined in one of the initial publications examining the effects of hybridization temperature on in situ hybridization signals using riboprobes (Cox et al., 1984). Importantly, there is a marked difference in the in situ hybridization signal intensity obtained after hybridization at 45 versus 50 °C, emphasizing that even moderate changes in this parameter can drastically affect the final data output. Equally as important is our finding that in tissues fixed under conditions suspected of being sub-optimal, reasonable in situ hybridization signals might be achieved by hybridizing at unconventionally high temperatures, such as 60 °C (Fig. 2). It is unlikely in all instances of sub-optimal tissue preservation, that high-temperature hybridization will provide otherwise missed in situ hybridization signals, but our data suggest it is a parameter that can be evaluated quickly. Further analysis is required to identify the hybridization temperature that will ultimately prevent hybrid formation and maintenance in tissues fixed at 4 °C. These data also indicate that not only are the numbers of vRNA+ cells observed by in situ hybridization dependent upon the tissue fixation and hybridization temperatures, but the signals per cell are also similarly dependent on these parameters. These are, of course related outputs for the assay, as classification of a cell as ‘positive’ or ‘negative’ is based upon a visual density of silver grains over individual cells. There are clear differences in the distributions of signals per cell after in situ hybridization under the conditions examined, and these likely are underestimates of the differences. This would be expected, because as silver grain density increases over a cell, although the increased numbers of silver grains lead to an increase in the surface area of reflected light, the clustering of the silver grains in different planes of focus also leads to a decrease in surface
31
area of reflected light per silver grain in captured images, due to the inability of each silver grain to present 100% of its surface area for illumination and reflection. In summary, this study demonstrates that the temperatures of tissue fixation and probe hybridization contribute in a co-ordinated fashion to the success of the in situ hybridization assay. Although we have obtained these data using radioactively labeled riboprobes to detect SIV RNA, we propose that these optimized in situ hybridization parameters will be broadly applicable to studies of host cellular mRNAs, other viral RNAs, and other riboprobe detection strategies. The mRNAs expressed by SIV and other retroviruses are essentially cellular mRNAs, insofar as they are transcribed by RNA polymerase II and processed by the host cellular machinery. This aspect of the SIV model system used here extend the relevance of these findings to detection of other mRNAs, whether cellular or viral, by in situ hybridization. Finally, these studies suggest that empiric determination of optimal in situ hybridization temperatures might increase greatly an investigator’s ability to detect viral and cellular mRNAs in tissue and cellular specimens especially, when there is an incompletely understood or sub-optimal fixation history.
Acknowledgements We thank Dr Michael Murphey-Corb, Dr Saverio Capuano III, and Dawn McClemens-McBride for assistance with animal care and project co-ordination. This work was supported in part, by grants from the National Institutes of Health (HL62056) and the University of Pittsburgh Competitive Medical Research Fund, Office of Research.
References Cox, K.H., DeLeon, D.V., Angerer, L.M., Angerer, R.C., 1984. Detection of mRNAs in sea urchin embryos by in situ hybridization using asymmetric RNA probes. Dev. Biol. 101, 485 – 502.
32
B.A. Fallert, T.A. Reinhart / Journal of Virological Methods 99 (2002) 23–32
Desrosiers, R.C., 1990. The simian immunodeficiency viruses. Ann. Rev. Immunol. 8, 557 –578. Fox, C.H., Johnson, F.B., Whiting, J., Roller, P.P., 1985. Formaldehyde fixation. J. Histochem. Cytochem. 33, 845 – 853. Haase, A.T., Henry, K., Zupancic, M., Sedgewick, G., Faust, R.A., Melroe, H., Cavert, W., Gebhard, K., Staskus, K., Zhang, Z.Q., Dailey, P.J., Balfour, H.H., Erice, A., Perelson, A.S., 1996. Quantitative image analysis of HIV-1 infection in lymphoid tissue. Science 274, 985 –989. Hirsch, V.M., Fuerst, T.R., Sutter, G., Carroll, M.W., Yang, L.C., Goldstein, S., Piatak, M. Jr., Elkins, W.R., Alvord, W.G., Montefiori, D.C., Moss, B., Lifson, J.D., 1996. Patterns of viral replication correlate with outcome in simian immunodeficiency virus (SIV)-infected macaques: effect of prior immunization with a trivalent SIV vaccine in modified vaccinia virus Ankara. J. Virol. 70, 3741 –3752. Kim, J., Chae, C., 2001. Optimized protocols for detection of porcine circovirus 2 DNA from formalin-fixed paraffin-embedded tissues using nested polymerase chain reaction and comparison of nested PCR with in situ hybridization. J. Virol. Meth. 92, 105 –111. Kornfeld, H., Riedel, N., Viglianti, G.A., Hirsch, V., Mullins, J.I., 1987. Cloning of HTLV-4 and its relation to simian and human immunodeficiency viruses. Nature 326, 610 – 613. Mellors, J.W., Rinaldo, C.R. Jr., Gupta, P., White, R.M., Todd, J.A., Kingsley, L.A., 1996. Prognosis in HIV-1 infection predicted by the quantity of virus in plasma. Science 272, 1167 – 1170.
Murphey-Corb, M., Martin, L.N., Rangan, S.R., Baskin, G.B., Gormus, B.J., Wolf, R.H., Andes, W.A., West, M., Montelaro, R.C., 1986. Isolation of an HTLV-III-related retrovirus from macaques with simian AIDS and its possible origin in asymptomatic mangabeys. Nature 321, 435 – 437. Ozden, S., Aubert, C., Gonzalez-Dunia, D., Brahic, M., 1990. Simultaneous in situ detection of two mRNAs in the same cell using riboprobes labeled with biotin and 35S. J. Histochem. Cytochem. 38, 917 – 922. Reimann, K.A., Tenner-Racz, K., Racz, P., Montefiori, D.C., Yasutomi, Y., Lin, W., Ransil, B.J., Letvin, N.L., 1994. Immunopathogenic events in acute infection of rhesus monkeys with simian immunodeficiency virus of macaques. J. Virol. 68, 2362 – 2370. Reinhart, T.A., Rogan, M.J., Huddleston, D., Rausch, D.M., Eiden, L.E., Haase, A.T., 1997. Simian immunodeficiency virus burden in tissues and cellular compartments during clinical latency and AIDS. J. Infect. Dis. 176, 1198 – 1208. Shimakage, M., Sasagawa, T., 2001. Detection of Epstein-Barr virus-determined nuclear antigen-2 mRNA by in situ hybridization. J. Virol. Meth. 93, 23 – 32. Whetter, L.E., Ojukwu, I.C., Novembre, F.J., Dewhurst, S., 1999. Pathogenesis of simian immunodeficiency virus infection. J. Gen. Virol. 80, 1557 – 1568. Wilcox, J.N., 1993. Fundamental principles of in situ hybridization. J. Histochem. Cytochem. 41, 1725 – 1733. Yang, H., Wanner, I.B., Roper, S.D., Chaudhari, N., 1999. An optimized method for in situ hybridization with signal amplification that allows the detection of rare mRNAs. J. Histochem. Cytochem. 47, 431 – 446.
Improved detection of simian immunodeficiency virus RNA by in situ hybridization in fixed tissue sections: combined effects of temperatures for tissue fixation and probe hybridization Beth A. Fallert, Todd A. Reinhart * Department of Infectious Diseases and Microbiology, Graduate School of Public Health, Uni6ersity of Pittsburgh, Pittsburgh, PA 15261, USA Received 21 May 2001; received in revised form 26 July 2001; accepted 27 July 2001
Abstract In situ hybridization detection of viral RNAs in formaldehyde-fixed tissue specimens is used frequently to characterize the extent of viral replication within host tissues. The ability to determine the level of expression of viral RNAs in situ is dependent upon many factors including the extent of cross-linking during fixation, the pretreatment regimen utilized to relieve the effects of cross-linking, and the hybridization and wash protocols. In efforts to improve our ability to detect cells infected productively by simian immunodeficiency virus (SIV) in rhesus macaque tissues, the effects of unconventionally high (40 °C) and more standard low (4 °C) temperature fixation in 4% paraformaldehyde/phosphate buffered saline were tested empirically on in situ hybridization signals. In addition, hybridization temperatures ranging between 37 and 75 °C were utilized to determine the optimal hybridization conditions for detection of SIV productively infected cells. Fixation conditions of 40 °C and hybridization conditions of 50 – 55 °C were identified as providing the greatest sensitivity for detecting RNA+ cells and for quantitating the signal per cell, while still allowing antigenic epitopes to be detected by immunohistochemical staining. These data indicate that the signal intensity following in situ hybridization for viral RNAs is dependent upon the combined effects of tissue fixation and in situ hybridization temperatures. © 2002 Elsevier Science B.V. All rights reserved. Keywords: Simian immunodeficiency virus; In situ hybridization; Immunohistochemistry; Formalin fixation; Temperature
1. Introduction
* Corresponding author. Tel.: + 1-412-648-2341; fax: +1412-383-8926. E-mail address: [email protected] (T.A. Reinhart).
In situ hybridization is an extremely versatile and useful method for determining the numbers and locations of individual cells harboring a specific nucleic acid amongst all of the cells in a tissue section. In situ, hybridization is used for
0166-0934/02/$ - see front matter © 2002 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 6 - 0 9 3 4 ( 0 1 ) 0 0 3 7 8 - 0
24
B.A. Fallert, T.A. Reinhart / Journal of Virological Methods 99 (2002) 23–32
many purposes, including the identification of transgene-expressing cells, the analysis of gene expression during development, and quantitation of gene expression during disease processes. A frequent application for which in situ hybridization is used in infectious disease research is the identification of the numbers, locations, and types of cells productively infected by different viruses (Haase et al., 1996; Ozden et al., 1990; Reinhart et al., 1997; Kim and Chae, 2001; Shimakage and Sasagawa, 2001). These data can be used not only to determine whether a pathogen is present in a tissue specimen, but can also provide critical data on the extent to which the pathogen is replicating, where it is replicating, and how strongly its RNAs are being expressed. Accurate data on the extent and location of viral replication in tissues are crucial for determining the effects of local viral replication on host tissues and for identifying the mechanisms by which viral replication leads to pathogenic outcomes. The success of any in situ hybridization detection assay is critically dependent upon multiple factors including the conditions of tissue preservation, tissue section pretreatment, and hybridization and wash conditions, all of which can affect the integrity of the target RNAs, the ability of the probe to hybridize with the RNA, and therefore, the success of the assay (Yang et al., 1999). Typical protocols for the detection of RNAs in tissue sections via in situ hybridization have followed well defined nucleic acid hybridization parameters, which include 50% formamide, 10% dextran sulfate, and a hybridization temperature of 45– 50 °C (Wilcox, 1993). Fixation parameters, including the fixative itself, and the duration and temperature of fixation can affect the outcome of in situ hybridization experiments. Shorter fixation times can lead to improved signals (Wilcox, 1993), and protocols frequently call for fixation at 4 °C, but there is thus far no tissue preservation protocol that is universally used for all experimental systems. The primate lentivirus, simian immunodeficiency virus (SIV), is genetically, structurally, and pathogenically related to human immunodeficiency virus type 1 (HIV-1), which causes immunodeficiency in humans, and can cause diseases
in rhesus macaques that are very similar to those observed in HIV-1 infected humans (Desrosiers, 1990; Whetter et al., 1999). The amount of viral RNA present in plasma in humans (Mellors et al., 1996) and in lymphoid tissues in rhesus macaques (Hirsch et al., 1996) is thought to be proportional to the rate of disease progression. Therefore, reliable and sensitive methods of detecting SIV/HIV1 and cellular mRNAs by in situ hybridization in biopsies or post-mortem tissues are crucial for obtaining an accurate understanding of the virus/ host relationship as it relates to disease. Our interests in determining the extent of SIV replication in fixed tissue sections prompted us to re-examine our fixation and tissue preservation protocol in efforts to improve the sensitivity of the in situ hybridization assay for SIV RNA, while not compromising its specificity. Based on an initial observation that tissues inadvertently fixed at 40 °C provided exceptional in situ hybridization signal, we describe here comparative analyses of the effects of high and low temperature fixation in 4% paraformaldehyde/phosphate buffered saline (4% PF/PBS) on the in situ hybridization signal provided in rhesus macaque lymph node tissue sections after hybridization with a pool of SIV-specific riboprobes.
2. Materials and methods
2.1. Tissue preser6ation, sectioning, and pretreatment The cryopreserved, pathological tissue specimen used for these studies was an inguinal lymph node obtained from a rhesus macaque 2 weeks after intravenous inoculation with the SIV/DeltaB670 isolate (Murphey-Corb et al., 1986). The study that included the animal from which this lymph node was obtained was performed under the approval of the Institutional Animal Care and Use Committee at The University of Pittsburgh. The tissue specimen (:0.5×0.5× 0.25 cm3) was hemisected to facilitate diffusion of the fixative, placed into fresh 4% PF/PBS (pH 7.3; Sigma, St. Louis, MO), and fixed at 4 or 40 °C for 5 h on a rotary shaker. The tissues were then cryoprotected
B.A. Fallert, T.A. Reinhart / Journal of Virological Methods 99 (2002) 23–32
by washing in successive overnight incubations of 1× PBS, fresh 1× PBS, 10% sucrose/PBS, and 20% sucrose/PBS, all at 4 °C with shaking. The specimens were then frozen by immersion for 5– 10 s in methylbutane (Fisher, Pittsburgh, PA) cooled to −65 °C on dry ice, and then stored at − 85 °C. Cryopreserved tissues were pre-warmed at − 20 °C for 2 h before sectioning. Tissues were sectioned on a cryostat at a thickness of 14 mm, thaw mounted at 43 °C for 20 min onto SuperFrost Plus microscope slides (Fisher), and then stored at − 85 °C until use. Mounted tissue sections were post-fixed immediately out of − 85 °C storage in fresh 4% PF/PBS for 20 min, washed for 20 min in 70% ethanol, and dehydrated in graded ethanols. The tissues were then microwaved in citrate buffer (pH 6.0; Sigma), for a total of 10 min at 2 min intervals with power settings between 20 and 60% using a Sharp Carousel microwave (1500 W). Buffer lost due to boiling over was replaced between the 2 min intervals. The slides were then removed from the microwave and allowed to cool for 30 min on the bench, rinsed in water, acetylated in 0.1 M triethanolamine (Sigma) with 0.25% acetic anhydride (Sigma) for 6 min, rinsed in water and dehydrated in graded ethanols.
2.2. In situ hybridization The riboprobes utilized in these studies encompassed sequences from four regions of the SIVmacBK28 molecular clone (Kornfeld et al., 1987) spanning portions of the gag, pol, en6, and nef genes, and included positions 47– 1130, 1676– 3121, 6600– 8286, and 8453– 9267 (Genbank accession no. M19499). Riboprobes were synthesized by in vitro transcription of linearized pGEM4z plasmid templates containing the appropriate insert, using T7 or SP6 RNA polymerase to generate the antisense or corresponding sense probes. In vitro, transcription was performed using the Maxiscript SP6/T7 kit (Ambion, Austin, TX), with plasmid DNA (0.2 mg/ml) and 35S-UTP (0.05 mCi; ICN, Costa Rica, CA). Transcription was allowed to proceed at 37 °C for 3 h and DNAse was added for an additional 20 min.
25
Unincorporated nucleotides were removed by ethanol precipitating twice with 25–50 ug of yeast tRNA as carrier. After re-suspension in RNAsefree water, the probe was heated at 65 °C for 5 min to facilitate re-suspension. Equivalent amounts of the four antisense and sense riboprobes were then pooled, respectively, to a final concentration of 0.5×106 cpm/ml. Following microwave pretreatment, in situ hybridization was carried out essentially as described (Reinhart et al., 1997) in the following hybridization mix: 20 mM Hepes (Sigma) pH 7.2; 1 mM EDTA (BioWhitaker, Walkersville, MD); 1× Denhart’s reagent; 0.1 mg/ml PolyA (Amersham/Pharmacia, Piscataway, NJ); 0.6 M NaCl (BioWhitaker); 100 mg/ml yeast tRNA; 10% dextran sulfate; 50% formamide; and riboprobes at 50 000 cpm/ml. The hybridization mix (5–10 ml) was placed on each tissue section, covered with a siliconized coverslip, and sealed with rubber cement. Hybridizations were performed at temperatures ranging from 37 to 75 °C for : 18 h. Coverslips were then removed under 5× standard saline citrate (SSC) at room temperature and moved progressively through: 5× SSC/10 mM DTT, 42 °C, 30 min; 2 × SSC/50% formamide/ 10 mM DTT, 60 °C, 20 min; riboprobe wash solution (0.1 M Tris, pH 7.5, 50 mM ethylenediaminetetraacetic acid, 0.4 M NaCl), 37 °C, 10 min, twice; riboprobe wash solution containing 12.5 U/ml RNAse T1 and 12.5 mg/ml RNAse A, 37 °C, 30 min; riboprobe wash solution, 37 °C, 15 min; 2 × SSC, 37 °C, 15 min; 0.2× SSC, 37 °C, 15 min. Tissue sections were then dehydrated in graded ethanols containing 0.3 M ammonium acetate, air dried, coated with NTB-2 emulsion (Kodak, Rochester, NY) and exposed at 10 °C for 1 day.
2.3. Enumeration of SIV 6RNA+ cells and in situ hybridization signal per cell SIV vRNA+ cells were counted in each of 20 random microscopic fields viewed through a 40× Plan Apochromat objective. The in situ hybridization signal over individual cells was determined by capturing images with a SPOT camera (Diagnostic Instruments, Sterling Heights, MI) mounted
26
B.A. Fallert, T.A. Reinhart / Journal of Virological Methods 99 (2002) 23–32
on a Nikon E600 fluorescence microscope, through a 60× Plan Apochromat objective after illumination with polarized light directed through an IGS polarizing filter cube (Omega Optical, Brattleboro, VT). Images were captured and analyzed using the Metaview software package (Universal Imaging Corporation, West Chester, PA). Individual cells in captured images were selected using the region tool and then thresholded to include all of the silver grains within that region. The surface area (pixels) of light reflected by the silver grains over each vRNA+ cell was then determined using the measure function in Metaview, and the data values exported to Microsoft Excel.
2.4. Immunohistochemical staining and simultaneous in situ hybridization and immunohistochemical staining Immunohistochemical staining was carried out essentially as described (Reinhart et al., 1997) with either a rabbit polyclonal serum specific for CD3o (Dako, Carpenteria, CA; catalogue no. A0452) or a murine monoclonal antibody specific for CD68 (Dako; clone KP1). Tissue sections were blocked by incubation of sections in 1× PBS containing 5% non-fat dry milk for 1– 3 h at room temperature. Excess blocking solution was then dabbed off and the primary antibodies diluted in PBS/5% non-fat dry milk was added at a dilution of 1:50 for CD68 and 1:100 for CD3o. Antibodies were detected using the avidin– biotin complex method with the Vectastain Elite system (Vector Labs, Burlingame, CA), using 3%,3-diaminobenzidine (Sigma) as the final substrate. Identification of the cell types expressing SIV RNA was performed by combining in situ hybridization with 35S-UTP-labeled riboprobes and immunohistochemical staining with antigen-specific antibodies. Following in situ hybridization, tissue sections were immediately rinsed in 1× PBS prior to initiation of the immunohistochemical staining. Following incubation with the 3%,3diaminobenzidine substrate, sections were then dehydrated in graded ethanols containing 0.3 M ammonium acetate, air dried and coated with NTB-2 emulsion and exposed at 10 °C for 1 day.
Antigen-positive cells were identified by the deposition of a brown precipitate and RNA+ cells were identified by collections of silver grains over individual cells.
3. Results
3.1. In situ hybridization for SIV+ cells: detection of 6RNA+ cells depends upon fixation and hybridization temperatures Equal portions of an inguinal lymph node were obtained from an SIV-infected rhesus macaque 2 weeks post-infection. At this time, referred to as the acute phase of infection, viral replication is widespread and peaks (Hirsch et al., 1996; Reimann et al., 1994), usually subsiding to lower levels very soon thereafter. These specimens were fixed in 4% PF/PBS for 5 h on a rotary shaker either at 4 or 40 °C, and then cryoprotected and snap frozen. In situ hybridization detection of SIV productively infected cells with a pool of SIV-specific riboprobes in these two portions of lymph node at different hybridization temperatures ranging from 37 to 75 °C yielded drastically different results. Representative high-power microscopic fields from sections of the tissue fixed at 40 °C and hybridized at different temperatures are shown in Fig. 1, and demonstrate that hybridization at 50–55 °C resulted in very strong signal, whereas hybridization at lower or higher temperatures resulted in reduced signal intensity. As shown in Fig. 2, the numbers of vRNA+ cells/ mm2 in the portion of tissue fixed at 40 °C increased from 0.4 after hybridization at 37 °C to 9.9 after hybridization at 75 °C, peaking at 41.3 after hybridization at 50 °C. The numbers of vRNA+ cells obtained after hybridization at 50, 55, and 60 °C were all between 28.5 and 41.3. In contrast, the numbers of vRNA+ cells in the portion of tissue fixed at 4 °C increased steadily with hybridization temperatures between 37 and 60 °C, peaking at 36.2 vRNA+ cells per mm2 after hybridization at 60 °C. Parallel in situ hybridizations were performed on the same slide with a pool of SIV-specific sense control riboprobes and in all instances the sense control
B.A. Fallert, T.A. Reinhart / Journal of Virological Methods 99 (2002) 23–32
probes provided only low level background in situ hybridization signal with no RNA+ cells detected on the entire tissue section (data not shown). As an additional control, in situ hybridization was performed at 50 °C on tissue sections from an uninfected macaque inguinal lymph node fixed at 40 °C, and for both the antisense and sense probes, no RNA+ cells were observed over the entire tissue section (data not shown). In summary, tissue fixation and in situ hybridization temperatures of 40 and 50 °C, respectively, were identified as providing strong autoradiographic signals without an accompanying loss of specificity.
27
3.2. In situ hybridization for cells producti6ely infected by SIV: signals per cell depend upon fixation and hybridization temperatures In situ hybridization with radioactive probes, coupled with emulsion autoradiography, allows the intensity of the signal to be quantitated by counting individual silver grains or their collective surface areas over individual RNA+ cells (Cox et al., 1984; Haase et al., 1996; Ozden et al., 1990; Reinhart et al., 1997). We therefore, utilized a quantitative image capture and analysis system to measure the signal intensity over individual cells in the same tissue sections used to enumerate the
Fig. 1. In situ hybridization detection of SIV vRNA+ cells. Representative photomicrographs are shown for sections obtained from the lymph node specimen fixed at 40 °C in fresh 4% PF/PBS and hybridized with a pool of SIV-specific riboprobes. The hybridization temperatures correspond to (A) 37 °C, (B) 45 °C, (C) 50 °C, (D) 55 °C, (E) 60 °C, and (F) 75 °C. Exposure times were 1 day. Original magnification, 300 × .
28
B.A. Fallert, T.A. Reinhart / Journal of Virological Methods 99 (2002) 23–32
Fig. 2. Enumeration of SIV vRNA+ cells following in situ hybridization. The numbers of vRNA+ cells were counted as described in Section 2 for lymph node specimens fixed in fresh 4% PF/PBS at either 4 or 40 °C. The data are presented as vRNA+ cells/mm2, with the gray bars representing the two independent experiments performed on tissue fixed at 40 °C and the white bars representing the two independent experiments performed on tissue fixed at 4 °C. Asterisks indicate the detection of no vRNA+ cells in 20 random high power fields. ND, not determined.
vRNA+ cells/mm2. Tissue sections were illuminated with epipolarized fluorescent light and microscopic images of the light reflected by the metallic silver grains were captured with a digital camera. Using the region, threshold, and measure tools of the Metaview software package the surface area of reflected light from individual vRNA+ cells was determined. Whenever possible, at least 100 vRNA+ cells per tissue section were included in these analyses. As shown in Fig. 3, the surface areas of reflected light over individual cells ranged from : 200 to 10 000 pixels/cell. The means and distributions of the surface areas of reflected light per cell indicate that fixation of the tissues at 40 °C and hybridization at 50– 55 °C provided on average the greatest signal intensity
per vRNA+ cell. The percentage of vRNA+ cells with in situ hybridization signals covering greater than 2000 pixels/cell was higher for the tissue fixed at 40 °C as compared to 4 °C, except when hybridized at the unconventionally high temperature of 60 °C, which provided nearly equivalent signals ranging between 33 and 39% (Fig. 3).
3.3. Antigenic epitopes remain a6ailable after fixation of tissues at high temperature A strength of microscope-slide based approaches is not only the ability to detect the RNA expressed by a specific virus, but also the ability to determine the numbers, locations and types of cells present within the tissue section, for both
B.A. Fallert, T.A. Reinhart / Journal of Virological Methods 99 (2002) 23–32 29
Fig. 3. Quantitative image analysis of the in situ hybridization signals per vRNA+ cell. The surface area of polarized light reflected by silver grains over vRNA+ cells was determined as described in Section 2 for lymph node specimens fixed at either 4 °C (grey symbols) or 40 °C (black symbols). Only those tissue sections that provided more than two vRNA+ cells/mm2 (Fig. 2) were examined to determine the in situ hybridization signals per vRNA+ cell; those not examined are indicated by the asterisk. Results are shown from both of the replicate experiments. In the bottom of the figure are the mean surface areas of reflected light per cell and the percentage of cells with \2000 pixels of reflected light per cell. ND, not determined.
30
B.A. Fallert, T.A. Reinhart / Journal of Virological Methods 99 (2002) 23–32
infected and other cells. Identification of the types of cells expressing viral RNAs can be accomplished by coupling immunohistochemical staining for cell-type specific antigens with in situ hybridization for viral RNA (Reinhart et al., 1997). To examine whether fixation at 40 °C rendered antigenic epitopes unavailable for immunohistochemical staining, a standard avidin– biotin complex (ABC) staining approach was applied to this tissue specimen. Immunohistochemical staining for the T-lymphocyte-specific marker, CD3o, and the monocyte/macrophage-specific marker, CD68, was undertaken. In both instances, the antigenic epitope was readily detectable (Fig. 4) and provided robust signal, even after performing in situ hybridization for SIV RNA and immunohistochemical staining for CD68 simultaneously on the same tissue section (Fig. 4C). In this instance, a CD68-negative cell is expressing SIV RNA. Application of control antibody provided only minimal background immunohistochemical staining signal (data not shown).
4. Discussion It was demonstrated that fixation of tissues at 40 °C in fresh 4% PF/PBS and probe hybridization at 50–55 °C provided optimal detection of SIV RNAs in cells infected productively via in situ hybridization. The improved in situ hybridization signal might be due to increased covalent cross-linking by the 4% PF due to high temperature induced increases in the concentration of free aldehyde from the methylene glycol, which predominates in formaldehyde solutions (Fox et al., 1985). The higher fixation and hybridization temperatures, which provided increased in situ hybridization sensitivity without compromising specificity did not destroy the antigenic epitopes on the cell-type specific markers CD3o and CD68, indicating that higher temperature fixation does not grossly reduce the availability of antigenic epitopes for immunohistochemical staining. These data demonstrate an inter-relationship between fixation and hybridization conditions, and identify high temperature (40 °C) fixation
Fig. 4. Immunohistochemical detection of antigenic epitopes in lymph node sections from a tissue specimen fixed at high temperature. Standard avidin – biotin complex (ABC) immunohistochemical staining approaches were used to detect CD3opositive T-lymphocytes (A) or CD68-positive monocytes/macrophages (B). Antigen positive cells are identifiable by the deposition of an insoluble brown product within the cell. Simultaneous immunohistochemical staining for CD68 and in situ hybridization for SIV RNA was performed to identify the target cells infected by SIV (C). Original magnifications: 400 × (A, B) and 600 × (C).
B.A. Fallert, T.A. Reinhart / Journal of Virological Methods 99 (2002) 23–32
conditions as optimal, when the ensuing tissue sections are hybridized at routine temperatures (45 – 55 °C). Higher hybridization temperatures have been demonstrated to be appropriate, when using riboprobes due to the greater stability of RNA/RNA hybrids (Cox et al., 1984), and our determination that : 50 °C is an optimal temperature for in situ hybridization is very similar to that determined in one of the initial publications examining the effects of hybridization temperature on in situ hybridization signals using riboprobes (Cox et al., 1984). Importantly, there is a marked difference in the in situ hybridization signal intensity obtained after hybridization at 45 versus 50 °C, emphasizing that even moderate changes in this parameter can drastically affect the final data output. Equally as important is our finding that in tissues fixed under conditions suspected of being sub-optimal, reasonable in situ hybridization signals might be achieved by hybridizing at unconventionally high temperatures, such as 60 °C (Fig. 2). It is unlikely in all instances of sub-optimal tissue preservation, that high-temperature hybridization will provide otherwise missed in situ hybridization signals, but our data suggest it is a parameter that can be evaluated quickly. Further analysis is required to identify the hybridization temperature that will ultimately prevent hybrid formation and maintenance in tissues fixed at 4 °C. These data also indicate that not only are the numbers of vRNA+ cells observed by in situ hybridization dependent upon the tissue fixation and hybridization temperatures, but the signals per cell are also similarly dependent on these parameters. These are, of course related outputs for the assay, as classification of a cell as ‘positive’ or ‘negative’ is based upon a visual density of silver grains over individual cells. There are clear differences in the distributions of signals per cell after in situ hybridization under the conditions examined, and these likely are underestimates of the differences. This would be expected, because as silver grain density increases over a cell, although the increased numbers of silver grains lead to an increase in the surface area of reflected light, the clustering of the silver grains in different planes of focus also leads to a decrease in surface
31
area of reflected light per silver grain in captured images, due to the inability of each silver grain to present 100% of its surface area for illumination and reflection. In summary, this study demonstrates that the temperatures of tissue fixation and probe hybridization contribute in a co-ordinated fashion to the success of the in situ hybridization assay. Although we have obtained these data using radioactively labeled riboprobes to detect SIV RNA, we propose that these optimized in situ hybridization parameters will be broadly applicable to studies of host cellular mRNAs, other viral RNAs, and other riboprobe detection strategies. The mRNAs expressed by SIV and other retroviruses are essentially cellular mRNAs, insofar as they are transcribed by RNA polymerase II and processed by the host cellular machinery. This aspect of the SIV model system used here extend the relevance of these findings to detection of other mRNAs, whether cellular or viral, by in situ hybridization. Finally, these studies suggest that empiric determination of optimal in situ hybridization temperatures might increase greatly an investigator’s ability to detect viral and cellular mRNAs in tissue and cellular specimens especially, when there is an incompletely understood or sub-optimal fixation history.
Acknowledgements We thank Dr Michael Murphey-Corb, Dr Saverio Capuano III, and Dawn McClemens-McBride for assistance with animal care and project co-ordination. This work was supported in part, by grants from the National Institutes of Health (HL62056) and the University of Pittsburgh Competitive Medical Research Fund, Office of Research.
References Cox, K.H., DeLeon, D.V., Angerer, L.M., Angerer, R.C., 1984. Detection of mRNAs in sea urchin embryos by in situ hybridization using asymmetric RNA probes. Dev. Biol. 101, 485 – 502.
32
B.A. Fallert, T.A. Reinhart / Journal of Virological Methods 99 (2002) 23–32
Desrosiers, R.C., 1990. The simian immunodeficiency viruses. Ann. Rev. Immunol. 8, 557 –578. Fox, C.H., Johnson, F.B., Whiting, J., Roller, P.P., 1985. Formaldehyde fixation. J. Histochem. Cytochem. 33, 845 – 853. Haase, A.T., Henry, K., Zupancic, M., Sedgewick, G., Faust, R.A., Melroe, H., Cavert, W., Gebhard, K., Staskus, K., Zhang, Z.Q., Dailey, P.J., Balfour, H.H., Erice, A., Perelson, A.S., 1996. Quantitative image analysis of HIV-1 infection in lymphoid tissue. Science 274, 985 –989. Hirsch, V.M., Fuerst, T.R., Sutter, G., Carroll, M.W., Yang, L.C., Goldstein, S., Piatak, M. Jr., Elkins, W.R., Alvord, W.G., Montefiori, D.C., Moss, B., Lifson, J.D., 1996. Patterns of viral replication correlate with outcome in simian immunodeficiency virus (SIV)-infected macaques: effect of prior immunization with a trivalent SIV vaccine in modified vaccinia virus Ankara. J. Virol. 70, 3741 –3752. Kim, J., Chae, C., 2001. Optimized protocols for detection of porcine circovirus 2 DNA from formalin-fixed paraffin-embedded tissues using nested polymerase chain reaction and comparison of nested PCR with in situ hybridization. J. Virol. Meth. 92, 105 –111. Kornfeld, H., Riedel, N., Viglianti, G.A., Hirsch, V., Mullins, J.I., 1987. Cloning of HTLV-4 and its relation to simian and human immunodeficiency viruses. Nature 326, 610 – 613. Mellors, J.W., Rinaldo, C.R. Jr., Gupta, P., White, R.M., Todd, J.A., Kingsley, L.A., 1996. Prognosis in HIV-1 infection predicted by the quantity of virus in plasma. Science 272, 1167 – 1170.
Murphey-Corb, M., Martin, L.N., Rangan, S.R., Baskin, G.B., Gormus, B.J., Wolf, R.H., Andes, W.A., West, M., Montelaro, R.C., 1986. Isolation of an HTLV-III-related retrovirus from macaques with simian AIDS and its possible origin in asymptomatic mangabeys. Nature 321, 435 – 437. Ozden, S., Aubert, C., Gonzalez-Dunia, D., Brahic, M., 1990. Simultaneous in situ detection of two mRNAs in the same cell using riboprobes labeled with biotin and 35S. J. Histochem. Cytochem. 38, 917 – 922. Reimann, K.A., Tenner-Racz, K., Racz, P., Montefiori, D.C., Yasutomi, Y., Lin, W., Ransil, B.J., Letvin, N.L., 1994. Immunopathogenic events in acute infection of rhesus monkeys with simian immunodeficiency virus of macaques. J. Virol. 68, 2362 – 2370. Reinhart, T.A., Rogan, M.J., Huddleston, D., Rausch, D.M., Eiden, L.E., Haase, A.T., 1997. Simian immunodeficiency virus burden in tissues and cellular compartments during clinical latency and AIDS. J. Infect. Dis. 176, 1198 – 1208. Shimakage, M., Sasagawa, T., 2001. Detection of Epstein-Barr virus-determined nuclear antigen-2 mRNA by in situ hybridization. J. Virol. Meth. 93, 23 – 32. Whetter, L.E., Ojukwu, I.C., Novembre, F.J., Dewhurst, S., 1999. Pathogenesis of simian immunodeficiency virus infection. J. Gen. Virol. 80, 1557 – 1568. Wilcox, J.N., 1993. Fundamental principles of in situ hybridization. J. Histochem. Cytochem. 41, 1725 – 1733. Yang, H., Wanner, I.B., Roper, S.D., Chaudhari, N., 1999. An optimized method for in situ hybridization with signal amplification that allows the detection of rare mRNAs. J. Histochem. Cytochem. 47, 431 – 446.