Reports on HIV/AIDS: 1990

JULY 27, 1990, MMWR, Vol. 39, No. 29, pp. 489-493
regions (V3, V4, and V5) and a constant region (C3) of the major external glycoprotein
gp120 were selectively amplified using the polymerase chain reaction (PCR) (1).
Amplified HIV DNA was molecularly cloned, and nucleotide sequences of multiple
clones were determined. The relatedness of the sequences was analyzed by several
computer-based methods in collaboration with Los Alamos National Laboratory.t
This multifaceted analysis showed a similarity between the sequences from the
patient and the dentist that was comparable to what has been observed for cases that
have been epidemiologically linked (Los Alamos National Laboratory, unpublished
data). Although the viral sequences from the dentist and the patient could be
distinguished from each other, they were closer than what has been observed for
pair-wise comparisons of sequences taken from the other North American isolates
studied (3).
Reported by: Div of HIV/AIDS and Hospital Infections Program, Center for Infectious Diseases;
Dental Disease Prevention Activity, Center for Prevention Svcs; National Institute for Occupational Safety and Health, CDC.
tViral sequences obtained from the samples taken from the dentist and the patient were shown
to be distinct by the following criteria:
1. Each PBMC sample was split into two before extraction of DNA. PCR amplification of human
leukocyte antigen (DQO) sequences was performed on each sample. The sequences were the
same between samples from the same person, but the dentist and patient DNA samples
were clearly different.
2. The average difference (4.6%, range: 2.0%-7.2%) between all viral V4-C3-V5 sequences
present in the patient versus all those in the dentist was higher than the average difference
between the viral sequences present within the dentist alone (3.5%, range: 1.2%-6.0%) and
within the patient alone (2.0%, range: 0.4%-3.6%).
3. Viral sequences in the patient possessed some unique substitutions not found in the viral
sequences from the dentist, and vice versa.
Viral sequences obtained from the samples taken from the dentist and the patient were judged
to be closely related by the following criteria:
1. Individual consensus sequences deduced from single base substitutions (excluding insertions and deletions) in the patient's and dentist's viral sequence sets over the V3-V4-C3-V5
regions of the envelope gene differed by 1.2%. Corresponding DNA regions from 17 other
distinct North American isolates gave pair-wise differences to the dentist's consensus viral
sequence of 5.1%-10.2%, with an average of 8.1%. Similarly, comparison of the patient's
consensus viral sequence to these 17 gave pair-wise differences of 5.9%-10.7%, with an
average of 8.8%. The range of all pair-wise differences among the 17 was 4.7%-12.9%, with
an average of 9.2%.
2. Unique patterns of nucleotide substitutions not found in any other virus isolate examined
were shared between viral sequences found in the dentist and patient.
3. The average difference (4.6%) between all of the patient's viral sequences and all of the
dentist's viral sequences over the V4-C3-V5 regions falls into a class of differences
(3.4%-5.8%) similarly determined for viruses from known epidemiologically linked cases (2;
Los Alamos National Laboratory, unpublished data). These include two instances of sexual
transmission, one instance of perinatal transmission, and an instance in which a group of
persons with hemophilia became infected from a single batch of factor VIII concentrate.

84