2 Methods
2.1 Patients and Samples
The study was approved by the Internal Review
Board of Warsaw Medical University in Poland.
Cerebrospinal fluid (CSF) was obtained from a
HIV-positive 28-year-old woman hospitalized
with sepsis and impaired consciousness at
Warsaw Municipal Hospital for Infectious
Diseases, in March 2012. HIV viral load in
serum and CSF was 10^6 copies and 1.5 105
copies/ml, respectively (ABBOTT Real Time
PCR HIV CE Abbott Molecular Inc., Des
Plaines, IL). A control CSF sample was obtained
as part of a routine diagnostic procedure from a
69-year-old man with motor neuron disease.
HSV-1 positive samples consisted of DNA
isolated from Vero cell line (ATCC CCL81)
and provided by the Reference Virus Isolation
Service Laboratory (New York, NY). Viral load
was 4.4 103 copy/μl (LightCycler HSV 1/2
Qualitative Kit Roche Diagnostics, Mannheim,
Germany). As a negative control (sample W1 and
W2) ultrapure water (Water Molecular Biology
Reagent, Sigma Aldrich) was used.
The HIV-positive CSF sample was diluted in
the HIV-negative CSF to obtain concentrations
of 10^4 ,10^3 ,10^2 , and 10^1 viral copies per reaction.
Serial dilutions of HSV were obtained by spiking
negative CSF with control HSV DNA. Likewise,
dilutions were adjusted to 10^4 ,10^3 ,10^2 , and 10^1
viral copies per reaction.
2.2 Nucleic Acid Isolation
and Amplification
Total RNA was extracted from 250μl of CSF by
the Chomczynski method (Chomczynski and
Sacchi 1987 ). The yield and purity of isolated
RNA was estimated using a NanoDrop Spectro-
photometer and Qubit 2.0 Fluorometer with RNA
HS Assay Kit (Thermo Scientific, Waltham,
MA).
The extracted RNA and DNA were subjected
to isothermal RNA amplification (Ribo-SPIA)
using commercially available kit RNA-Seq Ova-
tion V2 kit (NuGEN, San Carlos, CA). This
system requires a minimal amount of input
RNA (500 pg) and generates a smaller back-
ground of human sequences then other commer-
cially available kits (Malboeuf et al. 2013 ; Kurn
et al. 2005 ). The manufacturer’s protocol was
closely followed and the only modification was
a substitution of random primers for oligo-dT
mix. The amplified products were purified using
AMPure XP beads (Beckman Coulter, Pasadena,
CA) and measured on a Qubit 2.0 Fluorometer
(Life Technology, Carlsbad, CA). In our earlier
study we showed that the above approach
enables the detection of both RNA and DNA
pathogens (Perlejewski et al. 2015 ).2.3 NGS Library Construction,
Sequencing and Bioinformatic
AnalysisThe library for each sample was generated using
an Nextera XT Kit (Illumina, San Diego, CA)
according to the manufacturer’s protocol.
Briefly, cDNA was fragmented using
transposon-based method followed by dual
indexes addition with PCR. To eliminate short
cDNA inserts within sequences and to increase
the final length of sequences in the library, the
following modifications were introduced:
(1) final concentration of DNA digesting enzyme
(transposase) was reduced by one fourth (2) PCR
products were purified twice using 1.8 and 1.6
volumes of AMPure XP beads.
The quality and average length of sequence
library for each sample was assessed using a
Bioanalyzer (Agilent Technologies, Santa
Clara, CA) with a high sensitivity DNA Assay.
Six indexed samples were pooled equimolarly
and sequenced on a single lane of the Illumina
HiSeq 1500 (100 base paired-end reads).
A bioinformatic analysis was conducted as
described elsewhere (Perlejewski et al. 2015 ).
In short, raw Illumina high-throughput sequenc-
ing reads were analyzed with a pipeline that
included quality control, trimming, reference
alignment to the human reference sequenceSensitivity of Next-Generation Sequencing Metagenomic Analysis for Detection... 55