Amplification of Herpes simplex type 1 and Human Herpes type 5 viral DNA from formalin-fixed Alzheimer brain tissue

Neuroscience Letters 390 (2005) 37–41

Amplification of Herpes simplex type 1 and Human Herpes type 5 viral DNA from formalin-fixed Alzheimer brain tissue John D. Rodriguez a , Donald Royall b , Luke T. Daum a , Kathleen Kagan-Hallet c , James P. Chambers a,∗ b

a Department of Biology, The University of Texas at San Antonio, 6900 North Loop 1604 West, SB Rm# 1.01.28, San Antonio, TX 78249, USA Department of Psychiatry, The University of Texas Health Science Center at San Antonio, 7703 Floyd Curl Drive, San Antonio, TX 78229, USA c Department of Pathology, Audie Murphy Memorial Veterans Hospital, 7400 Merton Minter, San Antonio, TX 78229, USA

Received 17 May 2005; received in revised form 28 July 2005; accepted 29 July 2005

Abstract It is known that nucleic acids from formalin-fixed tissues are not nearly as good templates for DNA amplification as those extracted from fresh tissues. However, specimens stored in most pathologic archives are initially fixed in formalin. The possibility of an infectious etiology of several diseases including Alzheimer’s underscores the usefulness of archived tissue in assessing the association of infectious agents with specific pathology. In this report, we describe in detail a method resulting in robust amplification of HSV1 and Human Herpes type (HHV) 5 viral DNA targets using formalin-fixed Alzheimer brain frontal and temporal tissue as source of amplification template. Herpes simplex type 2 viral DNA was not detected in the limited samples examined in this study. Amplicons were verified by sequence analysis. Brain tissue stored in formalin longer than 1 year prior to post-formalin-fixation analysis gave rise to significantly shorter amplicons consistent with the observation that template DNA integrity decreases significantly with increasing time of storage in formalin. Thus, this report should be useful in PCR-based investigations assessing the regional presence of viral DNAs in formalin-fixed brain tissue. © 2005 Elsevier Ireland Ltd. All rights reserved. Keywords: Alzheimer’s; Herpes; Formalin-fixed; Brain

Although 10% formalin (3.7% formaldehyde) treatment of tissue is associated with deleterious interactions of varying degree with DNA [3,4,6,11,14,15], formalin provides the most adequate means of tissue preservation insuring optimal features for histological purposes [13]. We were interested in determining the presence of viral DNA in formalin-fixed brain tissue. In 1990, Nicoll et al. detected viral DNA using formalin-fixed, paraffin-embedded brain samples from patients with acute HSV1 encephalitis by PCR [12]. Bertrand et al. subsequently described PCRbased detection of HSV1 viral DNA in specific formalin-fixed brain tissue regions of Alzheimer patients [1]. Recently, Hemling et al. detected HSV1, HHV6 and Varicella zoster viral DNAs using PCR in human brain albeit the source of PCR template DNA is unclear [7]. However, using the methods ∗

Corresponding author. Tel.: +1 210 458 5663; fax: +1 210 458 6692. E-mail address: [email protected] (J.P. Chambers).

0304-3940/$ – see front matter © 2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.neulet.2005.07.052

of Nicoll et al. [12], Bertrand et al. [1] and Hemling et al. [7], we were unable to detect by PCR a key glycolytic enzyme commonly used as a marker to exclude false negative results. Assay variation could arise from differences in inherent molecular profiles; however, such intrinsic tissue factors affecting PCR amplification of template DNA in archived tissue are not easily identified. Considering the known deleterious effects of formalin on DNA, a more probable and easily evaluated variable is that of post-formalin-fixing treatment prior to thermal cycling, i.e., xylene extraction and subsequent removal of xylene and protein digestion. Although our study is limited to a small number of archived brain samples, we describe in detail, post-formalin-fixation treatment of frontal and temporal tissue samples resulting in robust, sequence analysis validated amplification of (1) the important marker gene target, 3-phosphoglyceraldehyde dehydrogenase in all samples tested and (2) HSV1 or HHV5 viral DNA, respectively.

38

J.D. Rodriguez et al. / Neuroscience Letters 390 (2005) 37–41

The presence of HSV2 viral DNA was not detected. Brain tissue stored in formalin longer than 1 year prior to embedding and post-formalin-fixation analysis gave rise to shorter amplicons suggesting that brain template DNA integrity decreases significantly with increasing time of storage in formalin necessitating the design of primers specific for shorter DNA stretches. Thus, this report should be useful in PCRbased investigations assessing the regional presence of viral DNAs in formalin-fixed brain tissue. Source of brain tissue: Formalin-fixed Alzheimer brain tissue samples (frontal and temporal cortex stored in formalin 1, 2 and 3 years prior to paraffin embedding) were obtained from Dr. Kathleen Kagan-Hallet Department of Pathology, The Audie Murphy Veteran’s Hospital, San Antonio, TX, USA. Xylene extraction: Xylene extraction of formalin-fixed brain tissue was accomplished by modification of the procedures of Nicoll et al. [12] and Bielawski et al. [2]. Tissue (10–15 mg) was minced by hand using a sterile scalpel blade. The finely minced tissue was transferred to a 2 ml microcentrifuge tube and to each sample 1 ml xylene (Sigma–Aldrich, St. Louis, MO, USA) was added. The xylene tissue suspension was vortexed vigorously for 5 s followed by incubation at room temperature for 15 min. Bulk xylene was decanted and discarded using a pipette. This step was repeated twice. Residual xylene removal was accomplished by addition of 1 ml ACS grade, 100% ethanol (Sigma–Aldrich, St. Louis) vortexing for 5 s followed by incubation at room temperature for 5 min. This was repeated a second time prior to extraction of DNA. DNA extraction: DNA extraction was accomplished using modification of the Epicentre MasterPure reagent DNA extraction procedure [5]. To each sample was added 300 ␮l Tissue and Cell Lysis Solution (Epicentre, Madison, WI, USA) containing 2 ␮l (50 mg/ml Proteinase K, Epicentre, Madison) and 10 ␮l 50 ␮g/␮l Proteinase K (100 ␮g, Qiagen Valencia, CA, USA) in a total volume of 312 ␮l. The reaction mixture containing the tissue sample and Proteinase K cocktail was mixed thoroughly and incubated at 65 ◦ C overnight. Following incubation, samples were placed on ice for 4 min. To the lysed sample, 150 ␮l MPC Protein Precipitation Reagent (Epicentre, Madison) was added, gently agitated for 10 s and resulting debris removed by centrifuga-

tion for 10 min at 9000 × g using a microcentrifuge. Supernatant material was carefully decanted and transferred to a 2 ml microcentrifuge tube. To the supernatant material, 500 ␮l 99% (v/v) molecular biology grade isopropyl alcohol (Sigma–Aldrich, St. Louis) was added, inverting each sample multiple times (∼50). DNA was pelleted by centrifugation at 5 ◦ C for 10 min at 9000 × g. Isopropyl alcohol was carefully aspirated without dislodging the DNA pellet. The pellet material was washed twice with 35 ␮l 75% (v/v) molecular biology grade ethanol, gently tapping the tube for approximately 8 s each wash. Xylene (200 ␮l) was added to the respective DNA pellets, incubated at room temperature for 15 min followed by bulk removal of xylene using a pipette. Residual xylene was removed by adding to each sample 500 ␮l (ACS grade) 100% ethanol followed by incubation at room temperature for 5 min. Samples were desiccated for 30 min in order to remove residual ethanol followed by resuspension in 20 ␮l nuclease free water (Ambion, Austin, TX, USA). This material served as DNA template for PCR amplification. PCR assay: Control DNA obtained from HSV1, 2 and HHV5 cultures and tissue extracted DNA was quantitated spectrophotometrically. Ten-fold dilutions ranging from 100 ng/␮l to 1 pg/␮l were made for each of the templates using Nuclease Free water. Detection of HSV1, 2 and HHV5 template control DNA was observed in a range of 100 ng/␮l to 1 pg/␮l under the conditions described below. To each reaction was added 2 ␮l of each DNA dilution per 20 ␮l total reaction volume. A universal primer set was derived from the Herpes simplex virus types 1, 2 and HHV5 viral polymerase gene sequence producing expected PCR amplicon products of 235 (HSV1 and 2) and 238 (HSV5) base pairs using the forward and reverse primer sequences 5 -TCATCTACGGGGACACGGAC-3 and 5 TTGCGCACCAGATCCACG-3 , respectively (Integrated DNA Technologies, Coralville, IA, USA). The nucleotide sequence from the 3-phosphoglyceraldehyde dehydrogenase gene served as an internal control and was amplified using a series of forward and reverse primers (Integrated DNA Technologies) giving rise to amplicons of varying size (59–545 bp). Primer sequences are listed in Table 1. PCR reactions (20 ␮l total volume) were carried out using 2X Superscript II one Step-RT-PCR with Platinum Taq sys-

Table 1 3-Phosphoglyceraldehyde dehydorgenase forward and reverse primers Bp

Forward primer

Reverse primer

59 88 150 268 300 379 395 472 525 545

5 -AGATGCTGCATTCGCCCTCTTA-3 5 -CTCCAAACAGCCTTGCTTGCTT-3 5 -TGTTCGTCATGGGTGTGAACCA-3 5 -TGCCTTCTTGCCTCTTGTCTCT-3 5 -ACGCTTTCTTTCCTTTCGCGCT-3 5 -TTGGCAAATCAAAGCCCTGGGA-3 5 -ATGGCAAATTCCATGGCACCGT-3 5 -ATGGCAAATTCCATGGCACCGT-3 5 -TTTCTCCTCCGGGTGATGCTTT-3 5 -TTGGCAAATCAAAGCCCTGGGA-3

5 -TCCAGAATATGTGAGCAGCCCT-3 5 -TCTGAAGTAGTGGTGCCAGCTT-3 5 -TTGCTGCAAAGAAAGAGGGAGC-3 5 -TTGACGGTGCCATGGAATTTGC-3 5 -AAGCATCACCCGGAGGAGAAAT-3 5 -GCTTAAGGCATGGCTGCAACTGAA-3 5 -ATGAGCCTACAGCAGAGAAGCAGA-3 5 -TGGTTCACACCCATGACGAACA-3 5 -TTTGCCAAGTTGCCTGTCCTTC-3 5 -AGAGACAAGAGGCAAGAAGGCA-3

J.D. Rodriguez et al. / Neuroscience Letters 390 (2005) 37–41

tem reaction master mix (Invitrogen, Carlsbad, CA, USA) at a final 1× concentration containing 0.4 nM of each primer and 0.1 U/␮l Platinum Taq DNA polymerase (Invitrogen, Carlsbad). Reaction volumes were all brought to 20 ␮l with nuclease-free water. Conditions for amplification of Herpes virus gene targets were as follows: initial denaturation for 3 min at 95 ◦ C followed by 35 cycles consisting of DNA melting for 30 s at 95 ◦ C, primer-annealing for 15 s at 54 ◦ C and primer elongation at 68 ◦ C for 30 s. The final elongation reaction was carried out for 7 min at 68 ◦ C. Conditions for amplification of 3-phosphoglyceraldehyde dehydrogenase were as follows: initial denaturation for 2 min at 95 ◦ C followed by 40 cycles consisting of DNA melting for 20 s at 94 ◦ C, primer-annealing for 20 s at 50 ◦ C and primer elongation at 72 ◦ C for 45 s. The final elongation reaction was carried out for 10 min at 72 ◦ C. All PCR reactions were carried out using a Robocylcer 96 Gradient Cycler with hot top (Stratagene La Jolla, CA, USA). PCR products were separated electrophoretically on preformed 2% (w/v) E-Gel agarose gels (Invitrogen, Carlsbad) containing ethidium bromide and visualized under UV Transilluminator (Fisher Biotech, Vernon Hills, IL, USA). The respective amplicon products were size characterized using markers of known size (100 bp, Bio-Rad Hercules, CA, USA). Stained gels were archived using a photodocumentation camera (Fisher Biotech, Vernon Hills, CA, USA). DNA reamplification: Reamplification of amplicon material was carried out after carefully removing the respective gel areas containing the bands from the gel using a scalpel. Each gel slice was transferred to an Ultracentrifuge-MC centrifugal filter device (Amicon Billerica, MA, USA) and subjected to centrifugation at 9000 × g for 10 min using an Eppendorf centrifuge. Respective filters were discarded and the eluent was used for reamplification. Reamplification was carried out using 2 ␮l gel slice eluent material under conditions identical to those described above. Reamplification of product was verified by agarose gel electrophoresis as previously described. Amplified products were cleaned up using a QIAquick PCR Purification Kit (Qiagen Valencia) per manufacturer’s instructions resulting in removal of doublestranded DNA fragments from the PCR reaction mixture. Sequencing reaction: Typical sequencing reactions consisted of 4 ␮l Big Dye Terminator v3.1 Cycle Sequencing RR-100 (Applied Biosystems, Foster City, CA, USA), 3.5 ␮l nuclease free water, 500 nM primer and 2 ␮l template. Cycle conditions were 96 ◦ C for 1 min followed by 25 cycles at 96 ◦ C for 10 s, 50 ◦ C for 5 s and 60 ◦ C for 4 min. Excess dye terminator was removed using a DyeExTM 2.0 Spin Kit (Qiagen, Valencia) per manufacturer’s instructions. All centrifugation steps were carried out using an Eppendorf centrifuge at 700 × g. Respective spin column eluates were dehydrated using a vacuum centrifuge followed by resuspension in 10 ␮l formamide. All samples were sequenced using an ABI-Systems 3100 automated sequenator (Applied Biosystems, Foster City).

39

Fig. 1. Amplification of genomic 3-phosphoglyceraldehyde dehydrogenase (GAPDH) and Herpes simplex type 1 and Human Herpes type 5 viral polymerase gene targets from formalin-fixed human brain. Frame A: PCR amplification of specific gene targets from formalin-fixed tissue extracts (stored in formalin less than 1 year) was carried out as previously described. Lane 2: 100 bp ladder; lane 3: 300 bp GAPDH amplicon from control human genomic DNA; lanes 4 and 5: 300 bp GAPDH amplicon from formalin-fixed brain; lanes 6–8: amplicons from HSV1 (235 bp)-, HSV2 (235 bp)- and HHV5 (238 bp)-extracted DNA from respective virus-infected cultures; lane 9: formalin-fixed brain HSV1 amplicon; lane 10: formalin-fixed brain HHV5 amplicon; lane 11: 100 bp ladder. Frame B: Reamplification and sequencing of GAPDH and HSV1 and HHV5 viral polymerase gene target amplicons was carried out. GAPDH sequence analysis: Sequence shown unbolded represents 289 out of 300 nucleotides determined by sequencing. Underlined regions represent the 300-bp forward and reverse primer compliment, respectively, with bolded regions not detected by sequence analysis. The 289 nucleotide sequence was subjected to PubMed BLAST search and the resulting search correlated to 100% homology to GAPDH. 5 ACGCTTTCTTTCCTTTCGCGCTctgcggggtcacgtgtcgcagaggagcccctcccccacggcctccggcaccgcaggccccgggatgctagtgcgcagcgggtgcatccctgtccggatgctgcgcctgcggtagagcggccgccatgttgcaaccgggaaggaaatgaatgggcagccgttaggaaagcctgccggtgactaaccctgcgctcctgcctcgatgggtggagtcgcgtgtggcggggaagtcaggtggagcgaggctagctggcccgTAAAGAGGAGGCCCACTACGAA-3 . HSV1 and HHV5 sequence analysis: Sequence shown unbolded represents nucleotides determined (HSV1, 230 out of 235 and HHV5, 236 out of 238) by sequence analysis. All nucleotide sequences were subjected to a PubMed BLAST search and all sequences correlated to 100% homology to known HSV and HHV sequences from multiple sources within GenBank. Underlined sequence represents the HSV1 and 2, and HHV5 universal primer pair sequence corresponding to the sense strand, respectively, with bolded regions not detected by sequence analysis. Subtype 1: 5 TCATCTACGGGGACACGGACtccatatttgtgctgtgccgcggcctcacggccgccgggctgacggccatgggcgacaagatggcgagccacatctcgcgcgcgctgtttctgccccccatcaaactcgagtgcgaaaagacgttcaccaagctgctgctgatcgccaagaaaaagtacatcggcgtcatctacgggggtaagatgctcatcaagggCGTGGATCTGGTGCGCAA-3 . Subtype 5: 5 TCATCTACGGGGACACGGACagcgtgtttgtccgctttcgtggcctgacgccgcaggctctggtggcgcgtgggcccagcctggcgcactacgtgacggcctgtctttttgtggagcccgtcaagctggagtttgaaaaggtcttcgtctctcttatgatgatctgcaagaaacgttacatcggcaaagtggagggcgcctcgggtctgagcatgaagggCGTGGATCTGGTGCGCAA-3 .

Shown in Fig. 1, frame A is a representative electrophoretic analysis of amplicons produced by amplification of genomic 3-phosphoglyceraldehyde dehydrogenase (300 bp target comparable in size to that of the viral polymerase target) and HSV1 (235 bp) and HHV5 viral poly-

40

J.D. Rodriguez et al. / Neuroscience Letters 390 (2005) 37–41

Fig. 2. Amplification of 3-phosphoglyceraldehyde dehydrogenase (GAPDH) DNA template from 1- to 3-year-old formalin-fixed brain tissue. PCR amplification of GAPDH gene targets from formalin-fixed (1–3 years) brain template DNA was carried out as previously described using specific primer sets. Lane 1: 100 bp ladder; lanes 2–6 (1-year-old formalin-fixed brain): 300, 268,150, 88 and 59 bp; lanes 7–9 (2-year-old formalin fixed brain): 150, 88, and 59 bp; lanes 10–11 (3-year-old formalin-fixed brain): 88 and 59 bp; and lane 12: 100 bp ladder. The largest amplicon observed for brain stored in formalin for 2 and 3 years was 150 bp (lane 7) and 88 bp (lane 10), respectively.

merase (238 bp) gene targets from formalin-fixed, paraffinembedded human brain. Reamplification of respective gel slice extracted bands and subsequent sequence analysis confirmed the identity of the amplicons (Fig. 1, frame B). Storage in formalin in excess of 1 year appears to be accompanied by significant template, i.e., DNA degradation, as evidenced by inability to amplify genomic targets longer than 150 bp (Fig. 2, lane 7) and 88 bp (Fig. 2, lane 10). The 545 bp genomic amplicon is observed only in brain stored in formalin no longer than 1 year (data not shown). Due to template degradation, amplification of the respective viral polymerase gene targets (235 and 238 bp in length) was not evaluated using samples stored in formalin in excess of 1 year. We describe in detail, post-formalin-fixation treatment resulting in robust, sequence analysis validated amplification of (1) the housekeeping gene 3-phosphoglyceraldehyde dehydrogenase and (2) Herpes viral DNA targets in archived brain tissue. Although template integrity and thus amplification is dependent upon a myriad of factors, the post-analysis events that readily stand out are xylene treatment and removal of nucleic acids following protein digestion. Xylene is typically used for dewaxing purposes; however, residual, poorly understood/characterized inhibitors derived from formalin fixation apparently are removed by our xylene treatment protocol as evidenced by the fact that in the absence of subsequent xylene treatment of the precipitated DNA following extended proteinase K treatment, we were unable to amplify genomic DNA templates in formalin-fixed brain samples using short (59 bp) to moderately long (545 bp) primers (data not shown). Following xylene extraction, Nicoll et al. [12]

removed residual xylene by two consecutive 400 ␮l changes of ethanol. Bertrand et al. do not describe xylene extraction and Hemling et al. utilized the High Pure Viral Nucleic Acid Kit (Roche Diagnostics), a commercially available product designed principally for DNA extraction from blood and fresh tissue [1,7]. Brain tissue is a heterogenous tissue comprising different types of cells and phenotypes, thus constituting a complex proteinaceous milieu. Proteinase treatment significantly improves release of nucleic acids trapped within formalin-induced protein:protein and protein:nucleic acid cross-linked complexes [4]. Proteinase K digestion conditions used here are similar to those used by Douglas and Rogers but are considerably different than those of Hemling et al. [7]. Details of Proteinase K treatment are not given by Bertrand et al. [1] and Nicoll et al. [12] did not proteinase treat their respective samples [12]. It could be argued that proteinase treatment is not required when dealing with very high template number arising from a fulminating viral infection, e.g., HSV1 encephalitis. However, template copy number could be significantly reduced by formalininduced nucleic acid cross-linking, protein adduct formation and/or base modifications in patients with neurological diseases other than fulminating HSV1 encephalitis. Thus, in the absence of proteinase treatment, this could account for the absence of HSV1 viral DNA in control brain samples in the Nicoll study [12]. In contrast to sequence data presented here ruling out any possibility of primer dimer formation, identity of amplicons generated in Bertrand’s study was addressed using a restriction enzyme digestion approach in contrast to Hemling’s study that employed Southern blot hybridization. The possibility of an infectious etiology of several diseases including Alzheimer’s Disease has long been debated. Jamieson et al. have shown that the majority of elderly individuals as well as those with Alzheimer’s disease harbour a latent from of HSV1 in brain [8,9]. More recently, Wozniak et al. using autopsy brain tissue have shown the complete functional HSV1 genome to be present in the brain of Alzheimer disease patients and elderly normal brains [16]. Little et al. have shown Chlamlydia pneumoniae induces Alzheimerlike amyloid plaques in brains of BalB/C mice [10] supporting the hypothesis that bacterial (non-viral) agents give rise to acute infections possibly resulting in dementia [10]. Considering the copious amount of archived tissue available for study and the problems attendant to formalinfixation, the simple but detailed procedure described here could be very useful in assessing the association of infectious agents with specific pathology in archived brain tissue.

Acknowledgement This work was supported by NIH Grant GM 00819 (J.P.C.).

J.D. Rodriguez et al. / Neuroscience Letters 390 (2005) 37–41

References [1] P. Bertrand, D. Guillaume, K. Hellauer, D. Dea, J. Lindsay, S. Kogan, S. Gauthier, J. Poirier, Distribution of Herpes Simplex Virus Type 1 DNA in selected areas of normal and Alzheimer’s disease brains: a PCR study, Neurodegeneration 2 (1993) 201–208. [2] K. Bielawski, A. Zaczek, U. Lisowska, A. Dybikowska, A. Kowalska, B. Falkiewicz, The suitability of DNA extracted from formalinfixed, paraffin-embedded tissues fro double differential polymerase chain reaction analysis, Int. J. Mol. Med. 8 (2001) 573–578. [3] S.J. Diaz-Cano, S.P. Brady, DNA extraction from formalin fixed, paraffin embedded tissues: protein digestion as a limiting step for retrieval of high quality DNA, Diagn. Mol. Pathol. 6 (1997) 342–346. [4] M.P. Douglas, S.O. Rogers, DNA damage caused by common cytological fixatives, Mut. Res. 401 (1998) 77–88. [5] Epicentre. Isolation of DNA from paraffin-embedded tissue using the MasterPureTM complete DNA and RNA purification kit. http://www.epicentre.com/f5 3pa.asp. [6] H. Fraenkel-Conrat, Reaction of nuclei acid with formaldehyde, Biochem. Biophys. Acta 15 (1954) 307–309. [7] N. Hemling, M. Roytta, J. Rinne, P. Pollanen, E. Broberg, V. Tapio, T. Vahlberg, V. Hukkanen, Herpes viruses in Brains in Alzheimer’s and Parkinson’s diseases, Ann. Neurol. 54 (2003) 267–271. [8] G.A. Jamieson, N.J. Maitland, G.K. Wilcock, J. Craske, R.F. Itzhaki, Latent Herpes Simplex Virus Type 1 in normal and Alzheimer’s disease brains, J. Med. Virol. 33 (1991) 224–227. [9] G.A. Jamieson, N.J. Maitland, G.K. Wilcock, C.M. Yates, R.F. Itzhaki, Herpes simplex virus type 1 DNA is present in specific

[10]

[11]

[12]

[13]

[14] [15]

[16]

41

regions of brain from aged people with and without senile dementia of the Alzheimer type, J. Pathol. 167 (1992) 365–368. C.S. Little, C.J. Hammond, A. MacIntyre, B. Balin, D.M. Appelt, Chlamydia pneumoniae induces Alzheimer-like amyloid plaques in brains of BALB/c mice, Neurobiol. Aging 25 (2004) 419– 429. J.D. McGhee, P.H. von Hippel, Formaldehyde as a probe of DNA structure. Mechanism of the initial reaction of formaldehyde with DNA, Biochemistry 16 (1977) 3276–3293. J.A.R. Nicoll, N.J. Maitland, S. Love, Use of the polymerase chain reaction to detect herpes simplex virus DNA in paraffin sections of human brain at necropsy, J. Neurol. Neurosurg. Psychiatry 54 (1991) 167–168. C. Paska, K. Bogi, L. Szilak, A. Tokes, E. Szabo, I. Sziller, J. Rigo, G. Sobel, I. Szabo, P. Novak, A. Kiss, Z. Schaff, Effect of formalin, acetone, and RNAlater fixatives on tissue preservation and different size amplicons by real-time PCR from paraffin-embedded tissue, Diagn. Mol. Pathol. 13 (2004) 234–240. M. Staehelin, Reaction of tobacco mosaic virus nucleic acid with formaldehyde, Biochem. Biophys. Acta 29 (1958) 410–417. C. Williams, F. Ponten, C. Moberg, P. Soderkvist, M. Uhlen, J. Ponten, G. Sitbon, J. Lundeberg, A high frequency of sequence alterations is due to formalin fixation of archival specimens, Am. J. Pathol. 155 (1999) 1467–1471. M.A. Wozniak, S.J. Shipley, M. Combrinck, G.K. Wilcock, R.F. Itzhaki, Productive Herpes Simplex Virus in brain of elderly normal subjects and Alzheimer’s disease patients, J. Med. Virol. 75 (2005) 300–306.