Archives of Virology

, Volume 156, Issue 7, pp 1143–1150

Genetic characterization of a novel calicivirus from a chicken

Authors

    • Department of Biology, Institute for MicrobiologyDresden University of Technology
  • Jochen Reetz
    • Federal Institute of Risk Assessment
  • Peter Otto
    • Institute for Bacterial Infections and Zoonoses, Friedrich-Loeffler-InstitutFederal Research Institute for Animal Health
Original Article

DOI: 10.1007/s00705-011-0964-5

Cite this article as:
Wolf, S., Reetz, J. & Otto, P. Arch Virol (2011) 156: 1143. doi:10.1007/s00705-011-0964-5

Abstract

We describe the identification and genetic characterization of a novel enteric calicivirus, detected by transmission electron microscopy and RT-PCR in two clinically normal chickens and in a chicken with runting and stunting syndrome from different flocks in southern Germany. Positive findings were confirmed by sequencing. The complete nucleotide sequence and genome organization of one strain (Bavaria/04V0021) was determined. The genome of the Bavaria virus is 7,908 nt long and contains two coding open reading frames. Phylogenetic analysis of the deduced partial 2C helicase/NTPase, 3C cysteine protease, RNA-dependent RNA polymerase and complete VP1 capsid protein amino acid sequences showed that the virus is genetically related to but distinct from sapoviruses and lagoviruses. Morphologically, the Bavaria virus particles are 37-42 nm in diameter and exhibit characteristic cup-shaped surface depressions.

Introduction

Caliciviruses (CVs) are small, non-enveloped viruses of 27 to 36 nm in diameter. The family Caliciviridae consists of five recognized genera: Norovirus, Sapovirus, Lagovirus, Vesivirus and the recently ratified genus Nebovirus [1, 2]. It also includes several unclassified CV strains, including Tulane virus (proposed genus Recovirus) and St-Valérien-like viruses (proposed genus Valovirus) [3, 4]. The single-stranded, positive-sensed RNA genome of the CVs varies from 6.4 to 8.4 kb in length and encodes a polyprotein precursor for non-structural proteins (NS proteins), a major structural capsid protein (VP1) and a minor structural protein (VP2). The genomes of CVs differ in parts in their genomic organization and contain either two or three open reading frames (ORFs). ORF1 encodes the NS proteins and the VP1 capsid protein in lagoviruses, sapoviruses, neboviruses and the St-Valérien-like viruses, whereas the VP1 capsid protein is encoded in a separate ORF, ORF2, in noroviruses, vesiviruses and Tulane virus. The VP2 protein is encoded in a separate ORF near the 3′ terminus of the genome (ORF2 in lagovirus, sapovirus, nebovirus and St-Valérien-like viruses and ORF3 in norovirus, vesivirus and the Tulane virus). Several highly conserved amino acid (aa) motifs, such as GXPGXGKT (2C helicase/NTPase), G(D/Y)CGXP (3C cysteine protease) and DYSKWDST, GLPSG and YGDD (RNA-dependent RNA polymerase) are common to all CVs.

Caliciviruses have a worldwide distribution, and many of them are important human and veterinary pathogens. In addition to humans, CVs have been found to infect different animals including pigs, cattle, sheep, cats, dogs, mink, hares and a number of marine mammals [411]. Viruses that are suspected to be CVs based on their size and capsid morphology using electron microscopy have also been detected in chickens and a number of other birds, including goldfinch (Carduelis carduelis), Guinea fowl (Numida meleagris), pheasant (Phasianus colchicus) and white tern (Gygis alba rothschildi) [1218]. In chickens, CV-like particles have been associated with gastroenteritis, infectious stunting syndrome, poor feathering, sticky vents and other symptoms [14, 15, 19]. However, to date, little is known about their genetic characteristics, diversity, and their prevalence among chicken flocks or their actual host range.

Here, we report on the first detection of CVs in German chicken flocks using electron microscopy and highly degenerate reverse transcription PCR and on the full genomic sequence of one chicken CV strain.

Materials and methods

Specimens

Intestinal contents were obtained from five chickens at the age of 15 to 25 days from flocks in southern Germany (Bavaria) between 2004 and 2005. One chicken from flock M (Bavaria/04V0027) suffered from runting and stunting syndrome, but none of the other chickens from flocks E (Bavaria/04V0021) and X (Bavaria/05V0013, Bavaria/05V0016 and Bavaria/05V0018) showed any clinical symptoms.

Transmission electron microscopy (TEM)

The intestinal contents were investigated by TEM using conventional negative staining. Briefly, supernatants of the samples were applied to polioform-carbon-coated, 400-mesh copper grids (Plano GmbH, Wetzlar, Germany), stained with 2% aqueous uranyl acetate solution and examined by TEM (JEM-1010 JEOL, Tokyo, Japan) at 80 kV accelerating voltage.

RNA extraction

The intestinal contents were suspended 1:5 in phosphate-buffered saline (PBS, pH 7.2) and then clarified at 2,200 xg for 20 min. These supernatants were used for RNA preparation. Viral RNA was extracted from 200 µl of 10% (wt/vol) suspensions of intestinal contents using a High Pure Viral Nucleic Acid Kit (Roche Molecular Biochemicals Ltd, Mannheim, Germany) as per manufacturer’s instructions.

Calicivirus detection by RT-PCR

Based on the highly conserved CV RNA-dependent RNA polymerase (RdRp) motifs DYSKWDST and YGDD, a set of degenerate primers was used to amplify a ~320-bp fragment (including primers) of the CV genome (Table 1). Reverse transcription was carried out using SuperScript III First Strand Synthesis System for RT-PCR (Invitrogen, Darmstadt, Germany). The reverse primers pYGDDa and pYGDDb were used to generate specific cDNA with a non-complementary flap sequence at the 5′ end. The 10-µl RT reaction mixture contained 100 units SuperScript III reverse transcriptase, 10 units RNase inhibitor (RNaseOUTTM, Invitrogen), 0.5 mM of each dNTP (dATP, dCTP, dGTP and dTTP), 1x First Strand RT buffer (50 mM Tris-HCl pH 8.3, 75 mM KCl, 3 mM MgCl2), 5 mM DTT, 0.5 μM of each primer pYGDDa/b and 5 µl viral RNA. RT was carried out at 44°C for 20 min and was terminated at 85°C for 3 min. For PCR, each 25 µl reaction contained 1 µl of cDNA, 1x PCR buffer minus MgCl2, 2.5 mM MgCl2, 0.4 mM of each dNTP, 1U Platinum Taq polymerase (Invitrogen), 0.6 μM of each primer pDYSKWDSTa/b/c/d and 0.8 μM of each primer pYGDDa/b. Cycling conditions were 94°C for 2 min, followed by 40 cycles at 94°C for 30 s, 50°C for 30 s and 72°C for 20 s. PCR products were electrophoresed on a 2% (wt/vol) agarose gel and purified using a QIAquick PCR purification kit (QIAGEN, Hilden Germany) according to the manufacturer’s instructions. DNA sequencing was carried out by GATC Biotech AG, Konstanz, Germany, using an ABI 3730xl sequencer.
Table 1

Oligonucleotides used in this study

Name

Sequence 5′–3′

pDYSKWDSTa

GAYTAYTCIACGTGGGAYTC

pDYSKWDSTb

GAYTAYACIGGGTGGGAYTC

pDYSKWDSTc

GAYTAYAGYACGTGGGAYTC

pDYSKWDSTd

GAYTAYTCICGGTGGGAYTC

pYGDDa

ATCCAGCYCRTCATCACCRTAa

pYGDDb

ATCCAGCCARTCATCACCRTA

p802

GGCCMCCCKGGIWKIGGIAA

p774

ACCACACCAGGNGAYTGYGG

AAP

GGCCACGCGTCGACTAGTACGGGIIGGGIIGGGIIG

AUAP

GGCCACGCGTCGACTAGTAC

NV1882F

GGATGGTTAACTCAACACCCTACCC

NV1896F

CACCCTACCCCCTTAACTGTGACC

NV1972R

GCTAGTTGCAAAGACAACTCGTGAGT

NV1988R

AGGGGGTGGAGAGATTGCTAGTTGC

NV2023R

ATAGAAGGCACCAGCTAAGG

NV3937R

TGCCACCCAGTAGCGGGTCA

NV3981R

ACCAAGCCTGGTTGATAGGGCA

NV3918F

TGACCCGCTACTGGGTGGCA

NV4668R

ACGGAGTGCGTCAGGGGAGT

NV4749R

TGCCCAGAGGGGAGACCACG

NV5316F

GCCTTAGCACTGCGGAAGGA

NV5490R

ATGATGGTTGGAGCCTGTGGGG

NV5508R

GTGGTGATCCCACCTGAGGCTT

NV5984F

CCCTGGCCCTGACTTTGGCTTT

NV6899R

GCTGATCTCATGCGGTGTAGTCAGG

aNon-complementary flap sequence at the 5′-end of primers pYGDDa/b in boldface

Sequence extension and RACE

Attempts were made to sequence the full genome of one RT-PCR-positive specimen (Bavaria/04V0021). For this, based on the sequence obtained via degenerate RT-PCR, strain-specific primers were designed (Table 1). The 3′ end of the genome was amplified using the primer pair NV5316F/NV5984F and the 5′-T25-VN primer (T25-A/G/C-A/T/G/C) in a semi-nested RT-PCR. The PCR product was cloned (CloneJetTM PCR cloning Kit, Fermentas Life Science, St. Leon-Rot, Germany) and sequenced (GATC Biotech AG). Strain-specific reverse primers and primers p774 and p802 targeting the conserved aa motifs G(D/Y)CGXP (protease) and GXPGXGKT (NTPase) were used to amplify two fragments, of ~1 kb (p774 with NV4749 and NV4668) and ~2.7 kb (p802 with NV3981R and NV3937R) in length, in subsequent semi-nested PCR reactions as described elsewhere [3]. Partial PCR products were sequenced directly.

The remaining ~1.1-kb 5′-end sequence of the Bavaria/04V0021 virus was determined using 5′ RACE: first-strand synthesis was carried out using the Bavaria/04V0021 virus strain-specific reverse primer NV2023R, after which dCTPs or dATPs were added to the 3′ hydroxyl terminus of the cDNA in separate reactions using recombinant TdT (Fermentas Life Science). Subsequently, separate semi-nested and nested PCRs were carried out using Bavaria/04V0021 virus strain-specific reverse primers (NV1988R and NV1972R), 5′ RACE primers AAP and AUAP for poly-C-tailed cDNA, and the 5′-T25-VN primer for poly-A-tailed cDNA. PCR products were subsequently cloned and sequenced.

The final Bavaria/04V0021 virus sequence was confirmed by high-fidelity reamplification (High Fidelity PCR Enzyme Mix, Fermentas Life Science) using strain-specific primers, cloning (CloneJetTM PCR cloning Kit, Fermentas Life Science) and sequencing in both directions of five overlapping fragments. The full genome sequence of the Bavaria/04V0021 virus has been submitted to the GenBank database under the accession number HQ010042. The partial RdRp sequences of strains Bavaria/04V0013 and Bavaria/05V0027 were submitted under accession numbers HQ616593 and HQ616594.

Phylogenetic analysis

Phylogenetic and molecular evolutionary analyses of multiple sequence alignments of deduced amino acid sequences of the partial NTPase, protease and RdRp genes, and the complete capsid protein gene were conducted using Molecular Evolutionary Genetics Analysis (MEGA) software version 5 [20]. Amino acid alignments were completed with the CLUSTALW algorithm (gap opening penalty 12, gap extension penalty 0.2, protein weight matrix: BLOSUM). For the NS proteins (NTPase, protease and RdRp), initial alignments were analysed for the presence of regions that are flanked by conserved motifs. These regions were then realigned and corrected manually if necessary. The lengths and locations of the partial NS proteins and the capsid protein sequences that were used are shown in Table 2.
Table 2

Virus strains, tree notations, and capsid and partial NS polyprotein sequences used for phylogenetic comparison

Calicivirus strain

GenBank accession no.

Tree notationa

NTPase

Protease

RdRp

VP1

Norwalk virus

M87661

NoV Norwalk

163 (550–712)b

143 (1115–1257)

409 (1323–1731)

530

Bovine enteric Calicivirus strain Jena

AJ011099

NoV Jena

163 (472–634)

143 (1008–1150)

409 (1215–1623)

519

Murine norovirus 1

DQ285629

MNV–1

163 (494–656)

143 (1009–1151)

408 (1224–1631)

541

Sapovirus Manchester

X86560

SaV Manchester

153 (470–622)

116 (1069–1184)

412 (1249–1660)

559

Sapovirus Mc10

AY237420

SaV Mc10

153 (471–623)

116 (1071–1186)

412 (1250–1661)

556

Porcine enteric Calicivirus strain Cowden

AF182760

PEC

153 (454–606)

116 (1063–1178)

411 (1242–1652)

542

Feline calicivirus

L40021

FCV

155 (474–628)

114 (1095–1208)

415 (1286–1700)

668

VESV-like Calicivirus strain Pan-1

AF091736

Pan-1

155 (580–734)

114 (1208–1321)

417 (1400–1816)

709

Walrus calicivirus

AF321298

WCV

155 (580–734)

114 (1207–1320)

417 (1399–1815)

708

European brown hare syndrome virus

Z69620

EBHSV

155 (507–661)

108 (1113–1220)

415 (1294–1708)

574

Rabbit hemorrhagic disease virus-FRG

M67473

RHDV

156 (512–667)

108 (1120–1227)

415 (1301–1715)

577

Newbury agent 1 virus

DQ013304

BEC-Newbury1

154 (446–599)

109 (1010–1118)

423 (1182–1604)

549

Calicivirus strain NB

AY082891

BEC-NB

154 (446–599)

109 (1010–1118)

423 (1182–1604)

549

St-Valérien Calicivirus pig/AB90/CAN

FJ355928

St-Valerien

156 (390–545)

132 (844–975)

395 (1031–1425)

517

Tulane virus

EU391643

Tulane

154 (378–531)

135 (820–954)

395 (1008–1402)

534

Bavaria/04V0021

HQ010042

Bava/021

154 (335–488)

115 (1079–1193)

415 (1257–1671)

578

aCalicivirus notation used in phylogenetic trees

bLengths and locations (in parentheses) of aligned amino acid sequences in their respective ORF-1

Phylogenetic trees were constructed using the maximum-likelihood method based on the JTT matrix-based model. For each analysis, the phylogeny was tested using 1,000 bootstrap replications. Evolutionary distances between sequences were computed based on pairwise analysis. Molecular weights from deduced aa sequences were calculated using the Molecular Weight Calculator of the Protein Information Resource (PIR), Georgetown University Medical Center (http://pir.georgetown.edu/pirwww/search/comp_mw.shtml).

Results

Transmission electron microscopy

TEM of chicken intestinal contents revealed the presence of virus particles of 37-42 nm in diameter in samples Bavaria/04V0013, Bavaria/04V0021, and Bavaria/05V0027. The three TEM-positive samples originated each from different flocks. The virus particles showed cup-shaped surface depressions that are characteristic of a number of so-called “classic” CVs, including chicken CVs, as reported by Cubitt and Barrett [14] (Fig. 1).
https://static-content.springer.com/image/art%3A10.1007%2Fs00705-011-0964-5/MediaObjects/705_2011_964_Fig1_HTML.jpg
Fig. 1

TEM micrographs showing calicivirus-like particles with characteristic surface depressions present in the intestinal contents of the chicken Bavaria/04V0021, negatively stained with uranyl acetate (bars 50 nm)

Detection of CVs in intestinal contents by RT-PCR

Viral RNA was amplified by RT-PCR using primers targeting the conserved CV motifs DYSKWDST and YGDD. PCR products of the expected size were obtained for the three TEM-positive samples (Bavaria/04V0013, Bavaria/04V0021, and Bavaria/05V0027). Sequencing of the PCR products yielded 289-bp fragments (without primers) that differed in 18 to 26 nt positions. However, the deduced amino acid (aa) sequences of the three strains were identical. The 96-aa-long sequences contained the GLPSG motif, which further confirmed the CV nature of the viruses. Comparison with other sequences in the GenBank database revealed that these viruses were related to other CVs. TBLASTN results showed the highest aa identity score (82.8) with the RdRp gene of sapovirus swine/OH-QW152/03/US (GenBank accession no. AY826425).

Genomic organization of the Bavaria/04V0021 virus

Subsequently, sequence fragments of the Bavaria/04V0021 virus were amplified by several rounds of semi-nested or nested RT-PCRs by using strain-specific primers and published CV primers or 3′- and 5′-RACE primers. Finally, the full genome was sequenced.

The positive single-stranded RNA genome of the Bavaria/04V0021 virus is 7,908 nt long, excluding the poly-A tail. It has a ribonucleotide composition of 23.9% G, 21.9% A, 24.1% U, and 30.1% C. Similar to sapo-, lago- and neboviruses, the genome is predicted to contain two coding ORFs. A first AUG (predicted start of ORF1) is observed at nucleotide 5. ORF1 is 6,936 nt long and encodes a protein of 2,311 aa with a calculated molecular weight of 249.7 kDa. A smaller ORF2 is initiated at nucleotide 6,940, is 864 nt long, and encodes a protein of 287 aa with a calculated molecular weight of 30.4 kDa, which is the largest among the CVs. ORF1 is terminated by an UAA codon, and ORF2 by a UGA codon. Between ORF1 and ORF2, a -1 frameshift was present, and ORF2 overlaps the stop codon of ORF1 by 1 nt (GGUAAUGGC; ORF1 stop codon in bold letters and ORF2 start codon in underlined letters). The 5′ and 3′ non-translated regions (NTRs) are 4 nt and 105 nt long, respectively.

ORF1 of the Bavaria/04V0021 virus codes for a continuous polyprotein including the NS-proteins and capsid protein. Amino acid motifs that are conserved in CV NS-proteins were identified within ORF1. These included the 2C helicase/NTPase (NTPase) motif GXPGXGKT at position 345, the 3C-like cysteine protease (protease) motif G(D/Y)CGXP at position 1,176 and the RNA-dependent RNA polymerase (RdRp) motifs DYSKWDST, GLPSG and YGDD at positions 1457, 1,512 and 1,560, respectively. Using published cleavage maps of lagoviruses and the porcine sapovirus strain PEC/Cowden [2123], a putative 3C cysteine protease cleavage site for the VP1 capsid protein between amino acids 1,733 and 1,734 (ME/GV) was identified. The putative VP1 protein is predicted to be 578 aa long and to have a molecular weight of 59.9 kDa, which is similar to other CVs.

Phylogenetic analysis

Phylogenetic analysis of the partial NTPase, protease and RdRp and complete VP1 aa sequences confirmed that the Bavaria/04V0021 virus is a new member of the CV family. Pairwise analysis of the partial NS proteins and the VP1 protein revealed the highest similarity with sapoviruses and lagoviruses. The NTPase, RdRp and capsid regions of the Bavaria/04V0021 strain appeared to be closely related to the human and swine sapovirus strains SaV Mc10 and PEC (49.3-50.0%, 43.5-43.6% and 23.9-24.1% aa identity, respectively), whereas the protease region was more closely related to the lagovirus strains EBHSV and RHDV (35.0-35.9%) (Table 3). Phylogenetic consensus trees of the NTPase, protease, RdRp and capsid regions using the maximum-likelihood method were constructed, and these consistently placed the Bavaria/04V0021 strain on separate branches from members of other CV genera and therefore in a unique position (Fig. 2).
Table 3

Amino acid identity of the Bavaria/04V0021 virus with other caliciviruses in regions aligned for phylogenetic analysis

Calicivirus straina

% aa identity

NTPase

Protease

RdRp

Capsid

NoV Norwalk

34.0

18.4

34.8

15.5

NoV Jena

32.6

19.4

35.0

15.5

MNV-1

31.9

19.4

32.4

16.4

SaV Manchester

44.4

28.2

40.3

23.0

SaV Mc10

50.0

29.1

43.5

24.1

PEC

49.3

25.2

43.6

23.9

FCV

45.8

20.4

40.0

17.5

Pan-1

43.1

29.1

37.7

17.8

WCV

43.2

30.0

39.0

20.0

EBHSV

43.7

35.9

40.9

21.6

RHDV

42.4

35.0

42.0

22.1

BEC-Newbury1

33.2

27.2

37.7

20.0

BEC-NB

31.9

26.2

36.9

20.3

St-Valerien

27.8

14.6

32.1

16.5

Tulane

30.6

18.4

32.4

14.6

Sequence alignments are based on partial and complete amino acid sequences as shown in Table 2

aFull virus names and GenBank accession numbers are shown in Table 2

https://static-content.springer.com/image/art%3A10.1007%2Fs00705-011-0964-5/MediaObjects/705_2011_964_Fig2_HTML.gif
Fig. 2

Phylogenetic consensus trees of partial NTPase (A), protease (B), RdRp (C) and complete major capsid (D) amino acid sequences of caliciviruses based on ClustalW alignments. The lengths and locations of the amino acid sequences used in multiple sequence alignments are shown in Table 2. The trees were constructed using the maximum-likelihood method based on the JTT matrix-based model of MEGA5. The percentages of replicate trees in which the associated viruses clustered together in the bootstrap test (1,000 replicates) are shown next to the branches. Bars, substitutions per site

Discussion

Calicivirus infections are widespread among vertebrates and are associated with a range of economically important diseases, including gastroenteritis in humans, calves and pigs (norovirus and sapovirus) [8, 2426] and respiratory, vesicular and hemorrhagic diseases in rabbits, cats, pigs and cattle (lagovirus and vesivirus) [2730]. Caliciviruses were also found to be present in a broad spectrum of other vertebrates, such as cetaceans, dogs, lions, minks, primates, reptiles, rhesus macaques, sheep, skunks and walruses, with various disease patterns [3, 6, 3134].

Calicivirus-like particles in fecal material from chickens have been observed by the use of EM in a limited number of studies [14, 15, 19]. However, sequence information and genomic characteristics of these viruses have not been investigated. The present paper describes the detection and genetic characterization of a novel enteric CV obtained from chicken flocks in Germany. Genetically, the Bavaria/04V0021 strain showed significant differences from other CVs, and the genetic distance was sufficient for differentiation of genera. The genetic distances between the Bavaria/04V0021 strain and other CVs were smallest in the NTPase, RdRp and capsid regions with human and porcine sapoviruses and in the protease region with lagoviruses. The Bavaria/04V0021 strain also resembled these two genera in terms of genomic organization. The capsid gene of strain Bavaria/04V0021 is — like in all sapo-, lago- and neboviruses—fused to the NS protein genes in a single ORF.

Although one of the chicken samples that tested positive for CV was associated with runting and stunting syndrome, the two other positive samples (including the Bavaria/04V0021 strain) originated from chickens without any clinical symptoms. It is therefore unknown whether the described chicken CV strain itself can cause disease in chickens. Many infections with CVs in animals are thought to be benign or asymptomatic but may also cause serious clinical symptoms. Norovirus has been detected both in asymptomatic cattle and in calves with diarrhea [35], whereas in swine and sheep, norovirus has only been detected in asymptomatic animals [6, 36]. In contrast, murine norovirus-1 (MNV-1) can lead to severe diseases in immunocompromised mice, including encephalitis, hepatitis, meningitis, pneumonia and vasculitis [37]. Sapovirus also causes diarrhea in swine and minks [11, 36]. Feline calicivirus, a vesivirus, can cause oral and upper respiratory tract infections characterised by vesicular lesions and that rupture to form ulcers, and some strains also cause systemic infections, which are frequently fatal [38, 39]. Rabbit haemorrhagic disease virus (RHDV), a lagovirus, causes necrotizing hepatitis with high mortality rates in rabbits [40]. The real impact of CVs in chickens will therefore remain unknown until epidemiological data are available, which, to the best of our knowledge to date do not yet exist. In a study from the Netherlands, pooled fecal samples from 48 chicken farms tested negative for chicken CVs using RT-PCR [41]. However, this finding may be due to the fact that the primer pair that was used had originally been developed for the detection of human noroviruses [42], and these primers may not bind efficiently to other CV strains, including chicken CVs.

In conclusion, we have provided TEM and genomic evidence of the presence of CV in the intestinal contents of chickens. Genetically, the Bavaria/04V0021 strain is distinct from other CVs and may represent a new CV genus. The screening of larger numbers of samples may help to confirm the occurrence of this virus in chicken flocks and to determine its genetic diversity. Epidemiological data will also help to reveal transmission routes and elucidate the clinical and economic significance of CVs in chicken flocks. The establishment of sensitive and specific chicken CV RT-PCR and RT-qPCR systems is highly desirable and will assist in achieving these goals. Although it is very likely that the CV-like particles found by TEM contain the genetic sequence described, this is not absolutely certain. However, further work is now in progress, including the establishment of a cell culture system, to characterize the novel chicken CV in more detail.

Acknowledgments

We gratefully acknowledge the excellent technical assistance of Maria-Margarida Vargas in sample preparation for electron microscopy.

Copyright information

© Springer-Verlag 2011