WO2011050050A2 - Use of a high-resolution melting assay to measure genetic diversity - Google Patents

Abstract

A method is provided for determining the level of genetic diversity in a sample using a high-resolution melting assay.

Description

USE OF A HIGH-RESOLUTION MELTING ASSAY TO MEASURE GENETIC

DIVERSITY

CROSS-REFERENCE TO RELATED APPLICATION

This application claims the benefit of the following U.S. Provisional Application No.:

61/253,329, filed October 20, 2009 and U.S. Provisional Application No.: 61/314,235, filed March 16, 2010, the entire contents of which are incorporated herein by reference.

STATEMENT OF RIGHTS TO INVENTIONS MADE UNDER FEDERALLY SPONSORED RESEARCH

This work was supported by the following grants from the National Institutes of Health, Grant Nos: U01-AI068613 and U01-AI068632. The government has certain rights in the invention.

BACKGROUND OF THE INVENTION

Current studies suggest that one or a few HIV variants usually initiate infection, and that the immune response and other selective forces then drive the evolution of viral variants within an infected person, generating a complex population of related viral quasispecies. Several factors promote rapid HIV evolution, including large viral population size, rapid viral turnover, lack of proofreading by HIV reverse transcriptase, and a high rate of genetic recombination. Changes in HIV diversity have been associated with different stages of HIV disease. The rapid evolution and high rate of genetic diversity of HIV viruses also complicate HIV therapy and vaccine development.

HIV diversity is usually studied by analyzing sequences from individual HIV variants using phylo genetic or other methods. Individual viral sequences can be obtained by analyzing HIV after cloning, by single genome sequencing, or by high-throughput sequencing methods, such as "deep" pyrosequencing. HIV diversity has also been studied using heteroduplex mobility assays. Use of these methods may be limited by cost, effort required, or the complexity of analysis. The availability of a simple, rapid method for quantifying the level of HIV diversity could facilitate studies of HIV transmission and pathogenesis. Novel approaches for analysis of HIV diversity that are rapid and simple are needed. SUMMARY OF THE INVENTION

As described below, the present invention features a method of measuring the genetic diversity of a sample.

The invention provides compositions and methods for measuring the genetic diversity of a sample. Compositions and articles defined by the invention were isolated or otherwise manufactured in connection with the examples provided below. Other features and advantages of the invention will be apparent from the detailed description, and from the claims.

In one aspect, the invention generally features a method of measuring genetic diversity involving receiving a melting curve for a polynucleotide, ascertaining a left margin temperature and a right margin temperature for the melting curve, calculating a high-resolution melting (HRM) score for the sample by subtracting the left margin temperature from the right margin temperature; and correlating the HRM score with genetic diversity.

In another aspect, the invention generally features a method for measuring genetic diversity, the method involving amplifying a region of the genome of an organism in the presence of a detectable moiety, such that the amplicon comprises the detectable moiety; heating the polynucleotide comprising the detectable moiety; detecting an alteration in the signal generated by the detectable moiety in response to heating, wherein the detected alteration when plotted as a function of temperature defines a melting curve for said polynucleotide, and the difference in temperature between a left and a right margins of the melting curve is a HRM score; and correlating the HRM score with genetic diversity.

In yet another aspect, the invention generally features a method of discriminating between recent and non-recent HIV infections involving determining an HRM score for an HIV virus isolated from a subject, wherein the HRM score correlates with the length of infection.

In further aspects, the invention generally features a method of identifying virus from a subject having a viral infection involving; determining a melting curve for a virus isolated from the subject, and correlating the melting curve with a particular viral fingerprint.

In various embodiments of any of the above aspects or any other aspect of the invention delineated herein, the method also involves the step of generating a melting curve for a polynucleotide. In another embodiment, the polynucleotide is an amplicon comprising a detectable moiety incorporated into the amplicon in a polymerase chain reaction. In yet another embodiment, the polynucleotide is selected from the group consisting of dsDNA, dsRNA, and DNA/RNA hybrid. In other embodiments the melting curve is ascertained using a fluorescent assay. In additional embodiments the polynucleotide is isolated from an organism. In certain embodiments the organism is selected from a human, an animal, a plant, a virus, a bacterium, a fungus, and a protozoa. In some embodiments the virus is human immunodeficiency virus (HIV).

In various embodiments of any of the above aspects or any other aspect of the invention delineated herein, the detectable moiety is a fluorescent dye. In other embodiments the detected alteration is fluorescence which changes as a function of temperature. In another embodiment, the amplicon is heated over a melt range of 68°C to 98°C. In additional embodiments the melting curve displays a change in fluorescence as a function of temperature. In further embodiments the HRM score reflects the temperatures over which melting occurred.

In various embodiments of any of the above aspects or any other aspect of the invention delineated herein, the amplicon comprises at least a portion of a viral gag coding region. In additional embodiments the amplicon comprises at least a portion of HIV gag p7 and gag pi. In further embodiments the amplicon comprises at least a portion of the coding regions for gag p7, gag pi, and/or gag p6. In additional embodiments the amplicon comprises at least a portion of a viral env coding region. In certain embodiments the amplicon comprises at least a portion of a viral gp41 coding region. In yet another embodiment the gp41 amplicon comprises at least a portion of the coding regions for gp41 HR1, gp41 HR2, and/or gp41 IDR. In further

embodiments the HRM score for a region in HIV gag varies with the stage of HIV disease. In yet another embodiment the HRM score for HIV gp41 HR1 or gp41 HR2 region varies with the stage of HIV disease. In an additional embodiment an increased HRM score in an adult relative to a control is indicative of an increased severity of HIV. In another embodiment an increased HRM score is correlated with acute HIV, recent acquisition of HIV, chronic HIV, and AIDS. In further embodiments the amplicon comprises at least a portion of a viral pol coding region. In certain embodiments the pol coding region comprises a portion of the coding regions for protease and reverse transcriptase. In yet another embodiment the polynucleotide is purified from a source selected from the group comprising viruses, bacteria, fungi, cancer cells, tissue, and bodily fluids. In other embodiments the range of melting temperatures of DNA duplexes provides a measure of HIV diversity. In an embodiment the amplicon is generated using primers designed to amplify HIV subtypes A, B, C, and D.

In various embodiments of any of the above aspects or any other aspect of the invention delineated herein, a method of determining the severity of a viral infection involving, determining an HRM score for a virus, wherein the HRM score is determined according to any of the methods disclosed herein, and correlating the HRM score with the severity of the infection. In another embodiment is a method of determining the length of time a subject has had a viral infection involving; determining the HRM score for a virus isolated from the subject, wherein the HRM score is determined according to any of the methods disclosed herein, and correlating the HRM with the length of infection. In another embodiment HRM score increases in acute HIV, recent acquisition of HIV, chronic HIV, and AIDS. Another embodiment is a method of determining prognosis in a subject having a viral infection involving; determining the HRM score for a virus isolated from the subject, wherein the HRM score is determined according to any of the methods provided herein, and correlating the HRM with prognosis. Yet another embodiment is a method of determining the efficacy of a treatment for a viral infection involving; determining the HRM score for a virus isolated from the subject before and after treatment, wherein the HRM score is determined according to any of the methods described herein, and correlating the HRM with the efficacy of treatment. Another embodiment is a method of determining the duration of viral infection in a subject involving; determining the HRM score for a virus isolated from the subject, wherein the HRM score is determined according to a method described herein, and correlating the HRM with the duration of infection. An additional embodiment is a method of determining a cross-sectional incidence of a viral infection involving; determining the HRM scores for virus from samples isolated from a population of subjects, wherein the HRM scores are determined according to a method provided herein, and correlating the HRM with the incidence of infection. In further

embodiments the melting curve is generated for a mixture of at least two polynucleotides. In additional embodiments the HRM score of at least two polynucleotides are determined when the polynucleotides are analyzed as a mixture. In yet another embodiment the melting curve is generated by plotting the negative derivative of fluorescence/temperature [-d(fluorescence/dT)] against temperature. In additional embodiments the left and right margin temperatures are determined by measuring where the slope of the melting curve achieves a 30 degree angle. Definitions

By "melting curve" is meant a graphical display that provides data about the progressive melting of nucleic acid duplexes over a range of temperatures. This includes, but is not limited to, a plot of fluorescence versus temperature, or a plot of the negative derivative of fluorescence divided by a derivative of temperature [-d(melting)/d( temperature)] plotted against temperature, where melting is determined through the use of a fluorescent molecule

By "left margin temperature" is meant a first or lower temperature at which melting begins.

By "right margin temperature" is meant a second or higher temperature at which melting ends.

By "high-resolution melting score" or "HRM score" is meant the distance between the left margin temperature and the right margin temperature expressed in degrees temperature.

By "alteration" is meant a change (increase or decrease).

By "viral fingerprint" is meant a melting curve that can be used to distinguish a particular viral strain.

In this disclosure, "comprises," "comprising," "containing" and "having" and the like can have the meaning ascribed to them in U.S. Patent law and can mean " includes," "including," and the like; "consisting essentially of" or "consists essentially" likewise has the meaning ascribed in U.S. Patent law and the term is open-ended, allowing for the presence of more than that which is recited so long as basic or novel characteristics of that which is recited is not changed by the presence of more than that which is recited, but excludes prior art embodiments.

"Detect" refers to identifying the presence, absence or amount of an analyte to be detected.

By "detectable moiety" is meant a composition that when linked to a molecule of interest renders the latter detectable, via spectroscopic, photochemical, biochemical, immunochemical, radiological, or chemical means. For example, useful labels include radioactive isotopes, magnetic beads, metallic beads, colloidal particles, fluorescent dyes, electron-dense reagents, enzymes (for example, as commonly used in an ELISA), biotin, digoxigenin, or haptens.

By "disease" is meant any condition or disorder that damages or interferes with the normal function of a cell, tissue, or organ. By "portion" is meant a portion of a polypeptide or nucleic acid molecule. This portion contains, preferably, at least 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, or 90% of the entire length of the reference nucleic acid molecule or polypeptide. A fragment may contain 10, 20, 30, 40, 50, 60, 70, 80, 90, or 100, 200, 300, 400, 500, 600, 700, 800, 900, or 1000 nucleotides or amino acids.

"Hybridization" means hydrogen bonding, which may be Watson-Crick, Hoogsteen or reversed Hoogsteen hydrogen bonding, between complementary nucleobases. For example, adenine and thymine are complementary nucleobases that pair through the formation of hydrogen bonds.

By "isolated polynucleotide" is meant a nucleic acid (e.g., DNA) that is synthesized or free of the genes which, in the naturally-occurring genome of the organism from which the nucleic acid molecule of the invention is derived, flank the gene or DNA fragment. The term therefore includes, for example, a recombinant DNA that is incorporated into a vector; into an autonomously replicating plasmid or virus; or into the genomic DNA of a prokaryote or eukaryote; or that exists as a separate molecule (for example, a cDNA,genomic DNA, or DNA fragment produced by PCR or restriction endonuclease digestion) independent of other sequences. In addition, the term includes an RNA molecule that is isolated from a cell or virus, or transcribed from a DNA molecule, as well as a recombinant DNA that is part of a hybrid gene encoding additional polypeptide sequence.

"Primer set" means a set of oligonucleotides that may be used, for example, for PCR. A primer set would consist of at least 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 30, 40, 50, 60, 80, 100, 200, 250, 300, 400, 500, 600, or more primers.

By "reduces" is meant a negative alteration of at least 10%, 25%, 50%, 75%, or 100%. By "reference" is meant a standard or control condition.

By "subject" is meant a mammal, including, but not limited to, a human or non-human mammal, such as a bovine, equine, canine, ovine, or feline.

Ranges provided herein are understood to be shorthand for all of the values within the range. For example, a range of 1 to 50 is understood to include any number, combination of numbers, or sub-range from the group consisting 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, or 50. Unless specifically stated or obvious from context, as used herein, the term "or" is understood to be inclusive. Unless specifically stated or obvious from context, as used herein, the terms "a", "an", and "the" are understood to be singular or plural.

Unless specifically stated or obvious from context, as used herein, the term "about" is understood as within a range of normal tolerance in the art, for example within 2 standard deviations of the mean. About can be understood as within 10%, 9%, 8%, 7%, 6%, 5%, 4%, 3%, 2%, 1%, 0.5%, 0.1%, 0.05%, or 0.01% of the stated value. Unless otherwise clear from context, all numerical values provided herein are modified by the term about.

The recitation of a listing of chemical groups in any definition of a variable herein includes definitions of that variable as any single group or combination of listed groups. The recitation of an embodiment for a variable or aspect herein includes that embodiment as any single embodiment or in combination with any other embodiments or portions thereof.

Any compositions or methods provided herein can be combined with one or more of any of the other compositions and methods provided herein.

BRIEF DESCRIPTION OF THE DRAWINGS

Figures 1A to 1C show representative data from the HRM assay. Figure 1A is a graph HRM data from a representative clinical sample (fluorescence vs. temperature). In this figure, the start of melting (left margin) occurs where the slope has a downward turn (at approximately 88°C) and the end of melting (right margin) occurs where the slope flattens (at approximately 92.5°C). Figure IB is a melting curve for the sample shown in Figure 1A (- d[fluorescence]/d[temperature] plotted against temperature). The left and right margins of the melting curve are defined as the temperatures where the curve achieves an angle of 30° compared to a horizontal baseline (red and blue lines, respectively). Figure 1C is a plot of the melting curves for six control plasmids, two of each subtype (subtype A, red; subtype C, green; subtype D, blue).

Figure 2 shows HRM data from nine mother-infant pairs. Melting curves for each of the nine mother-infant pairs are shown panels MI-1 to MI-9. In each panel, the melting curve from the mother' s sample is shown in red and the melting curve from the corresponding infant is shown in blue. Figures 3A and 3B shows representative regions of the HIV genome analyzed using the HRM assay. Figure 3A: Gagl, Gag2, and Pol; Figure 3B: HRl, HR2 and the IDR region of the transmembrane protein, gp41, which is encoded by the env gene. The HIV genome is represented by a non-shaded bar. The positions of the HRM amplicons are shown by shaded bars. LTR: long terminal repeat; PR: protease; RT: reverse transcriptase; HR: helical region.

Figures 4A to 4E show the Kaplan Meier analysis for survival compared to HRM scores. For this analysis, infants with HRM scores above the 75th percentile (above the third quartile, >Q3) were characterized as having high HRM scores (black dashed line), and infants below that cutoff were characterized as having low HRM scores (grey line). The X axis shows the time since birth in days (infant age); the Y axis shows the survival probability. The number of infants still alive in each group (<Q3, >Q3) at each time point is shown below each graph. (Fig. 4A) Gagl region, (Fig. 4B) Gag2 region, (Fig. 4C) Pol region, (Fig. 4D) mean of the two gag regions (Gagl/Gag2), (Fig. 4E) mean of all three regions (Gagl/Gag2/Pol).

Figure 5 shows a comparison of the HRM scores in the Gagl, Gag2, and Pol regions of the HIV genome in samples from HIV-infected infants.

Figure 6 shows a scatter plot of the relationship of genetic diversity in three distinct regions of the HIV genome and different stages of HIV infection in adults.

Figure 7 shows that HRM scores increase with increasing progression of HIV disease.

Figure 8 shows that the HRM assay can be used to assess diversity in more than one region simultaneously in a multiplex format. Regions of eight plasmids derived from subtype B HIV that contain different portions of the gp41 gene. The HRl and HR2 regions were amplified in the presence of LC Green Plus Dye. The amplified products were melted using a high- resolution melting instrument (LightScanner), and melt curves were derived from the resulting data. Two single amplicon curves were generated for the HR2 and HRl regions of gp41. Data from one sample is shown. Figures 8A and 8B show results obtained by performing the HRM assay on a single region (HR2 and HRl, respectively). Figure 8C shows results obtained by simultaneously analyzing the HRl and HR2 regions in a multiplexed HRM assay (in a single HRM reaction). All three analyses (data shown in Figures 8A-8C) were conducted on the same plate in parallel. For single amplicon analysis (Figures 8A and 8B), 9 ul of mastermix designed to amplify either HRl or HR2 was added to each well. For the multiplexed analysis, 4.5 ul of the HR2 mastermix was added to the well, followed by 4.5 ul of the HRl mastermix. Thus, each multiplex well contained the same total primer amounts as a normal reaction, but with an equimolar amount of each primer set (Figure 8C).

Figure 9 shows that the HRM assay can be used to detect individual melting domains in a large amplicon, and the position of those domains on the temperature (X) axis corresponds to the guanine/cytosine (GC) content in each domain or amplicon. HRM analysis was conducted on an amplicon that consisted of a region of HIV gp41 that included the HR1, IDR, and HR2 regions (HXB2 coordinates: 7798-8299). The resulting melting curve had three peaks (Figure 9A). HRM analysis was subsequently conducted using primers designed to amplify shorter segments of the amplicon amplified in 9A. The individual regions analyzed included: HR2 (Figure 9B), the IDR region (Figure 9C), and HR1 (Figure 9D). The amplicon sizes and respective GC contents of each amplicon are noted below each panel. The position of the melting curves for the individual amplicons shifted from lower to higher temperatures as a product of increasing GC content. This indicates that one of the largest determinants of the position of the melting curve on the X (temperature) axis is GC content. Differences in GC content in various regions of a given amplicon are likely to be a major factor determining the temperature range over which the amplicon melts. Thus, amplicons that have distinct GC-rich regions or distinct melting domains may produce HRM curves with multiple peaks, reflecting differences in local GC content within the various melting domains.

Figure 10 shows that melting curves for different individuals have different shapes that may be used as a genetic fingerprint of the virus. Melting curves are shown for three different individuals (Figures 10A and D are from one individual; figures 10B and 10E are from one individual; figures IOC and 10F are from one individual). Figure 10A (panels A-C): HR1 diversity was assayed within within 1 day of template preparation; (panels D-F): the same samples were reanalyzed after a period of three weeks. The features of the melting curves for each individual (overall shape, left and right margin) were nearly identical in the duplicate tests (most notably observed in comparison between B and E). Figure 10B: the same analysis was performed for HR2. Other examples of the variation in melting curve shape among different individuals are shown in Figure 1.

Figure 11 shows the diversity of HIV (Gag2 amplicon) from a study subject treated with interleukin-2 (IL-2) and highly active antiretro viral therapy (HAART). HIV was analyzed during three time viremic periods using the HRM assay. The periods that were studied were: pre- treatment, HAART interruption #1 (No treatment (Tx) #1), and HAART interruption #2 (No Tx #2). All had very low diversity (HRM scores <4.5) with no significant variation between the time periods, although there was a slight increase in diversity in the last two time periods.

Figure 12 shows the HRM scores (plotted on the Y axis) of three samples analyzed in triplicate reactions, varying HIV RNA viral load (c/ml: copies/ml) and volume of plasma analyzed (in ul).

DETAILED DESCRIPTION OF THE INVENTION

The invention features methods that are useful for determining the genetic diversity of a sample.

The present invention provides methods of measuring genetic diversity in a sample by generating a melting curve for the sample, ascertaining the left and right margin temperatures from the melting curve, and calculating a high-resolution melting (HRM) score for the sample from the left and right margin temperatures.

The invention is based, at least in part, on the discovery that HIV viruses usually exist as quasispecies (mixtures of genetically-related variants) and that the level of diversity and type(s) of genetic variation (e.g., type and positions of point mutations, nucleotide insertions and deletions, etc.) differ from individual to individual, in different tissue compartments (e.g., blood, breastmilk, semen, lymphnodes, mucosal tissue) and change within an infected person during the course of HIV infection. An HRM assay was developed to analyze HIV diversity without sequencing. In this assay, DNA is amplified in the presence of a fluorescent dye. A high- resolution melting instrument, such as the LightScanner (Idaho Technologies, Inc.), is then used to detect the change in fluorescence as the DNA is melted. The HRM score indicates the number of degrees over which melting occurs, and is correlated with sequencing-based measures of HIV diversity. Plasma samples were obtained from nine Ugandan mother-infant pairs (HIVNET 012 trial). DNA amplified from the HIV gag region (Gag2 fragment) was analyzed to determine the number of degrees over which the DNA melted (HRM score). Individual Gag2 amplicons were also cloned and sequenced (50 clones/mother; 20 clones/infant). The median HRM score for HIV from infant samples (4.3, range 4.2-5.3) was higher than that for control plasmids (3.4, range 3.2-3.8, p < 0.001) and lower than that for HIV from maternal mothers (5.7, range 4.4-7.7, p = 0.005, exact Wilcoxon rank sum test). The intraclass correlation coefficient reflecting assay reproducibility was 94% (95% confidence intervals [CI]: 89-98%). HRM scores were also compared to sequenced-based measures of HIV diversity; higher HRM scores were associated with higher genetic diversity (P <0.001), complexity (P=0.009), and Shannon entropy

(P=0.022), but not with length variation (P=0.111). The HRM assay provides a novel, rapid method for assessing HIV diversity without sequencing. This assay could be applied to any region of the HIV genome or to other genetic systems that exhibit DNA diversity.

The invention is also based, at least in part, on the observation that higher HIV diversity in particular regions of the HIV genome is associated with disease progression and death. The HRM assay was used to measure HIV diversity in Ugandan infants and to examine the relationship between HIV diversity and infant survival through 5 years of age. Plasma samples were obtained from 31 HIV-infected infants (HIVNET 012 trial). The HRM assay was used to measure diversity in two regions in the HIV gag gene (Gagl and Gag2) and one region in the HIV pol gene (Pol) (Figure 3A). Higher HRM scores at 6-8 weeks of age (scores above the 75^th percentile) were associated with an increased risk of death by 5 years of age (for Pol: P=0.005; for Gagl/Gag2 (mean of two scores): P=0.003; for Gagl/Gag2/Pol (mean of three scores):

P=0.002). At 6-8 weeks, low and middle range HRM scores for different regions (Gagl, Gag2, Pol) were tightly clustered, while high HRM scores in different regions were not (Figure 5). In multivariate models that included HIV viral load at 14 weeks of age and HRM score, higher HRM scores at 6-8 weeks were independently associated with death (for Gag2: adjusted OR=3.5, 95% CI: 1.1, 10.9, P=0.03; for Gagl/Gag2: OR=8.7, 95% CI: 2.6, 28.6, P=0.0004; for

Gagl/Gag2/Pol: OR=6.9, 95% CI: 2.1, 22.9, P=0.002). Kaplan Meier curves showing the relationship between high HRM score (>Q3) and death is shown in Figure 4. No association between HRM scores and other clinical and laboratory variables was found in this cohort. Higher HIV diversity in these regions (Gagl, Gag2, Pol) at 6-8 weeks of age was associated with a significantly increased risk of death by 5 years of age.

Further, the invention is also based, at least in part, on the observation that genetic diversity in HIV gag and pol increased over time during HIV infection, reflecting the duration of infection. Plasma samples were obtained from 31 HIV-infected infants (HIVNET 012 trial). The HRM assay was used to measure diversity in two regions in the gag gene (Gagl and Gag2) and one region in the pol gene (Pol) (Figure 3A). HRM scores for HIV from infant samples ranged from 3.3 to 7.3 and were higher than HRM scores for plasmid controls (P<0.0001 for each region). HRM scores in all three regions increased with age from 6-8 weeks to 12-18 months (for Gagl : P=0.005; for Gag2: P=0.006; for Pol: P=0.016). Plasma samples were also obtained from 79 HIV-infected Ugandan children enrolled in an observational study (median age 4.7 years, range 0.6-12.4 years, enrollment 2004-2006). Some of the children had been exposed to single dose nevirapine (sdNVP) prophylaxis at birth. In this cohort, the median age of the sdNVP-exposed children was lower than that of the sdNVP-unexposed children (median: 1.7 years, range 0.6-6.3 years, vs. median: 7.7 years, range 2.9-12.4 years, p<0.0001, Wilcoxon rank sum test), reflecting the fact that many of the older children were born prior to widespread availability of sdNVP prophylaxis. The median HRM score for all 79 children was 5.9 (range:

3.8-11.9). In univariate models, higher HRM scores were significantly associated with older age, lower CD4 cell count, lower CD4 cell %, and lack of sdNVP exposure. In a multivariate model including age, CD4 cell % and HIV viral load as covariates, both a lower CD4 cell % and older age remained highly associated with higher HRM score (adjusted odds ratio (OR) for CD4 cell %: 5.4, 95% confidence intervals (CI): 1.7, 17.2), p=0.004; OR for age: 1.3 (1.1, 1.6), p=0.005). As age and sdNVP exposure were highly confounded in this observational study, sdNVP from the multivariate model were excluded, as an association of increased HIV diversity with older age (reflecting a longer duration of untreated HIV infection) was more biologically plausible than an association of increased HIV diversity without prior sdNVP exposure. The relationship of HRM score and duration of HIV infection provides proof-of-principle that the HRM assay will be a useful tool for cross-sectional HIV incidence determination.

In addition, the invention is based, at least in part, on the observation that higher HRM scores reflecting higher HIV gag diversity (Gag2 region, Figure 3C) were associated with immunologic status in HIV-infected individuals and also with the immunologic response to highly active antiretroviral therapy (HAART) which is used to treat HIV infection. Plasma samples were obtained from 79 HIV-infected Ugandan children enrolled in an observational study (median age 4.7 years, range 0.6-12.4 years, enrollment 2004-2006, same cohort as described above). These children met the 2003 World Health Organization (WHO) criteria for HAART and received a regimen of stavudine (d4T), lamivudine (3TC), and nevirapine (NVP); none of the children were switched to a second-line regimen during the study. Pre-treatment HIV viral loads were measured the day of HAART initiation; pre-treatment CD4 cell counts and CD4 cell % were measured within 30 days of HAART initiation. Plasma viral load, CD4 cell count, and CD4 cell % were assessed every 12 weeks between 24 and 96 weeks after HAART initiation. Detection of HIV RNA did not lead to change of therapy if the children were clinically and immunologically stable. Most of the children had a significant decline in HIV RNA after HAART initiation, and the response to HAART was similar among children with and without prior single dose NVP (sdNVP) exposure. Prior to HAART initiation, higher HRM scores reflecting greater HIV gag region diversity were associated with lower CD4 cell % values in a univariate model (odds ratio (OR): 6.8, 95% confidence intervals (CI): 2.4, 19.0, p=0.0003) and in a multivariate model adjusting for age and pre-HAART viral load (OR: 5.4, 95% CI: 1.7, 17.2, p=0.004). Higher HRM scores were also associated with lower CD4 cell count in a univariate model (OR: 5.7, 95% CI: 2.0, 15.8, p=0.0008), but not in a multivariate model (OR: 1.7, 95% CI: 0.4, 7.8, p=0.49). Pre-HAART HRM scores were not associated with virologic response to HAART. However, in children who did not have severe immune compromise prior to treatment (pre-HAART CD4 cell % > 5%), a higher pre-HAART HRM score was associated with a reduced immunologic response to HAART, when CD4 cell count was used to evaluate treatment response.

The invention is also based, at least in part, on the observation that genetic diversity in the HIV genome correlates with the stage of infection. In adults, HIV infection is usually initiated by one or a few viral variants. The genetic diversity of HIV generally increases during the course of HIV infection, but may decline in late stages of HIV disease. The HRM assay was used to compare HIV diversity in a region in gag (the Gag2 region, Figure 3A) and two regions in gp41 (HR1 and HR2, Figure 3B) in adults with different stages of HIV disease. Samples were obtained from adults with acute HIV infection (RNA positive, antibody negative, EXPLORE Study, N=20), recent infection (median 189 days after last HIV negative test [range: 14-540 days], EXPLORE Study, N=102), and non-recent infection (HIV infected >2 years, Johns Hopkins Univ. Moore Clinic and Emergency Department, N=68). The non-recent group included 35 adults with CD4 cell counts >50 cells/ul (chronic infection) and 33 adults with CD4 cell counts <50 cells/ul (AIDS). Samples were tested with the HRM assay to generate an HRM score for each of the three regions analyzed. For all 190 samples, the median (range) of HRM scores was 4.3 for gag (3.5-10.1), 4.6 for HR1 (4.1-8.1), and 4.7 for HR2 (4.1-9.0). Median

HRM scores were higher for adults with recent vs. acute infection (P=0.0095 for gag; P=0.018 for HR1; P=0.0007 for HR2) and were higher for adults with non-recent (chronic or AIDS) vs. recent infection (P<0.0001 for all three regions). Region- specific assay cutoffs were set at the mean + 3 standard deviations of the HRM scores obtained for adults with recent infection. HRM scores above these cutoffs were highly associated with non-recent infection (p<0.0001 for each region). Discrimination between recent and non-recent infection was enhanced by combining data from more than one region of the HRM genome (within the same or different HIV genes). These data show that the HRM assay is useful for discriminating between recent and non-recent HIV infection in cross-sectional samples, particularly when multiple regions of the HIV genome are analyzed. This indicates that the HRM assay can be used alone, or as part of a multi-assay algorithm, for cross-sectional HIV incidence determination. This assay is also useful for studying the relationship between HIV diversity and disease progression, and for studies of the pathogenesis of HIV infection.

The invention is also based, at least in part, on the observation that the HRM assay can be used to simultaneously analyze genetic diversity in different regions of the HIV virus, using a multiplex assay format. Figures 8A and 8B show the individual melting curves for the HR1 and HR2 domains of HIV gp41. Figure 8C shows the melting curve produced when the two domains are analyzed in a single HRM amplification reaction containing two different sets of primers. The temperatures over which each amplicon melts (left and right margins, peak melting temperature) and the HRM score (noted as melt range in the figure) are nearly identical when the amplicons are analyzed individually or in a single multiplex reaction.

The invention is also based, at least in part, on the observation that the temperatures over which an amplicon melts (left and right margins, peak melting temperature) is influenced by the GC content in the domain analyzed (Figure 9). The melting curve of a large amplicon with three GC-rich domains is shown in Figure 9A: the melting curve has three peaks. The melting curves for each of the three corresponding domains, amplified and analyzed in separate reactions using three different primer pairs (one for each of the GC-rich domains in the large amplicon), are shown in Figures 9B, 9C, and 9D. The domain with the lowest GC content (36.2% GC) melts at a lowest temperature range (Figure 9B curve is further to the left on the temperature (X) axis); the domain with an intermediate GC content (Figure 9C, 48.24% GC) is further to the right on the temperature (X) axis; the domain with the highest GC content (51.465 GC) melts at a higher temperature range (Figure 9D, curve is the farthest to the right on the temperature (X) axis). This information is useful for the design of HRM assays that involve multiplexing (analysis of more than one domain in a single reaction). If the individual domains in the multiplex reaction are selected (through primer design) to have different GC content, the curves for each domain can be "positioned" differently along the temperature (X) axis, displaying the data from each domain separately.

The invention is also based, at least in part, on the observation that the shape of the melting curves produced in the HRM assay varies among individuals. Examples of differently shaped melting curves are provided in Figure 1 and Figure 10. Differences in the "shape" of the melting curves includes, but are not limited to: the position of the curve on the temperature (X) axis (e.g., left margin, peak melting temperature, right margin), the width of the curve (e.g.,

HRM score, width at half height), the symmetry of the curve, the number of distinct peaks of the curve, the presence of "shoulders" in the curve (shown in Figure 10A and 10D), and the height/width ratio of the curve. These distinct features indicate that the HRM assay can be used to compare viral populations in two different samples or run data. Such comparisons could be used to evaluate the possibility of sample mix-ups (misidentification of a sample during testing), to monitor shifts in viral populations due to drug, immune, or other selective pressures, to detect abrupt changes in the viral population due to HIV super-infection, or to confirm that a treatment regimen suppresses viral evolution.

The invention is also based, at least in part, on the observation that antiretroviral drugs used to treat or prevent HIV infection provide selective pressure that can cause shifts in the viral population and changes in HIV diversity. Plasma samples were obtained from 79 HIV-infected Ugandan children enrolled in an observational study (median age 4.7 years, range 0.6-12.4 years, enrollment 2004-2006, same cohort as described above). Among children who were clinically and immunologically stable on HAART but were not virologically suppressed, HRM scores were significantly reduced at 48 or 96 weeks on non-suppressive HAART compared to the pre- HAART HRM score (p=0.001). These reductions in HRM score in children on non-suppressive HAART reflect bottlenecking of the viral population due to antiretroviral drug selection. These findings suggest that the HRM assay may be useful for monitoring the effects of non-suppressive antiretroviral drug regimens, such as those that may be used to treat patients with antiretroviral drug resistant HIV (where suppressive treatment is not an option) or to treat patients during periods when they are not likely to adhere to HAART (e.g., partially suppressive monotherapy). The invention is also based, at least in part, on the observation that the HRM assay can be used to monitor HIV diversity during treatment to gain information about the effect of the therapy on viral suppression and viral evolution. HIV diversity was measured in an adult treated with interleukin-2 (IL-2 therapy, Figure 11). Development of an HIV vaccine may be facilitated by identification of immunologic mechanisms for suppressing viral replication in vivo. The rationale for treatment of HIV infection with IL-2 is that it stimulates proliferation of T-cells, absolute CD4 T-cell counts, and cytotoxic and other functions of CD8 T-cells and NK cells that may inhibit HIV replication. At standard doses (4.5-7.5 million international units (MIU)/d bid), IL-2 can be given for only 5 days every 8 weeks because of toxicity. However, at 1.2 MlU/m /d toxicity is minimal and IL-2 can be given daily for many months with stimulation of cellular immunity. This provides both a chronic effect and a larger cumulative dose than intermittent dosing. Although IL-2 at either intermittent or daily dosing had no effect on chronic HIV infection, when daily low-dose IL-2 was used to treat a patient who had received potent ART within one month of HIV seroconversion, the patient subsequently had an undetectable HIV viral load for 14 months off all therapy, an unprecedented suppression lasting about 40 standard deviations longer than the mean duration of an undetectable viral load after HAART cessation. How closely the HIV that rebounded after this 14-month period of spontaneous suppression was related to the HIV variants present in this patient at earlier time points was analyzed. One patient was studied who had received ART for three years, beginning one month after seroconversion. ART was interrupted twice, once with and once without concurrent daily injections of ultra-low- dose IL-2 (1.2 million IU/M /day). Virus rebounded quickly after the interruption without IL-2. In contrast, the patient's viral load remained < 50 c/ml for >1 year without antiretroviral therapy after the interruption with IL-2, even though HIV was detected in resting CD4 T-cells and, with a specialized test, in plasma. To compare the patient's HIV strains before and after each treatment period, plasma HIV from 3 viremic periods: post-seroconversion, first viral rebound (no treatement (No Tx) #1), and second viral rebound (No Tx #2) were analyzed. Standard methods were used to sequence HIV protease and the first 300 bases of HIV reverse transcriptase.

Sequences were analyzed using the PHYLIP package. The HRM assay was performed as follows: 1) Amplify the region of interest in the presence of a fluorescent dye; in this study, a region in HIV gag (Gag2 region) was amplified (Figure 3A); 2) Use a high-resolution melting instrument (LightScanner) to heat the DNA, releasing the dye; 3) Analyze the melting curve to determine the HRM score, defined as the number of degrees over which the sample melts. The higher the HRM score, the greater the genetic diversity in the sample. Protease and reverse transcriptase sequences from the all samples from the study subject were nearly identical. After each HAART interruption, the level of gag diversity measured with the HRM assay was low, similar to the level of diversity one month after seroconversion (Figure 11). In this unique patient (the only patient to date treated with both early HAART and then ultra-low-dose daily IL-2), the HIV that rebounded after 14 months of suppression without therapy was nearly identical to viruses present earlier in his infection, as assessed by phylogenetic analysis of HIV sequences. These data essentially rule out super-infection as an explanation for his late viral rebound. The late rebound virus also had very low diversity by HRM analysis, another indication of lack of viral evolution. These data, in conjunction with rapid viral rebound (~2 weeks) after HAART was stopped without IL-2 therapy, argue strongly that endogenous immune mechanisms stimulated by IL-2 mediated the 14 months of viral suppression. This postulated effect of IL-2 in this patient is consistent with long-lasting stimulation of antiviral immunity by a short course of daily IL-2 in the mouse LCMV model. Taken together, these data suggest strongly that study of this patient, and ideally additional patients treated with early HAART and daily low-dose IL-2, may provide insights into in vivo immunological mechanisms of HIV suppression. These insights may facilitate development of a preventive HIV vaccine. The following examples are put forth so as to provide those of ordinary skill in the art with a complete disclosure and description of how to make and use the assay, screening, and therapeutic methods of the invention, and are not intended to limit the scope of what the inventors regard as their invention. EXAMPLES

Example I: Analysis of HIV diversity using a high-resolution melting (HRM) assay

The HRM assay was used to analyze plasma samples obtained from Ugandan women and infants enrolled in the HIVNET 012 trial (Guay LA et al, 1999, Lancet, 354:795-802). In HIVNET 012, HIV- 1 -infected, antiretro viral drug-naive women received a single dose of nevirapine

(sdNVP) during labor, and their infants received sdNVP shortly after birth to reduce the risk of HIV mother-to-child transmission. The analyzed samples were collected from women prior to NVP exposure and samples from their HIV-infected infants collected at 6-8 weeks of age.

Samples from nine mother-infant pairs were available for analysis. HIV-1 plasmids were served as controls.

DNA templates for the HRM assay were prepared from the plasma samples using the

ViroSeq HIV-1 Genotyping System v2 (ViroSeq, Celera, Alameda, CA). Briefly, HIV RNA was extracted from 500 ml of maternal plasma or 100 ml of infant plasma and reverse transcribed. A non-nested PCR was used to amplify a PCR product encoding a portion of HIV gag, HIV protease, and a portion of HIV reverse transcriptase. The amplified DNA was purified using spin columns, analyzed by agarose gel electrophoresis, and diluted according to the

manufacturer's instructions. A portion of the diluted PCR products was sequenced to identify anti-retro viral drug resistance mutations. PCR products that remained after sequencing were stored at - 80°C and used as templates for the HRM assay.

A region of HIV gag (Gag2 region) was amplified from the PCR products prepared in the ViroSeq system. Each 10 ml HRM amplification reaction included 1 ml of a 1 : 10 dilution of ViroSeq PCR products or 5 ng of plasmid control (template DNA), 0.2mM forward and reverse primers, and lxLight Scanner Master Mix amplification buffer (Idaho Technologies, Salt Lake City, UT), which contains Taq polymerase and a fluorescent dye (LCGreen Plus dye), which is incorporated into the amplified PCR products. The primer sequences were forward: HRM-3F: 5'- ACTGAGAGACAGGCTAATTTTTTAG (corresponds to HXB2 2068-2092) and reverse:

HRM-3R: 5'-GGTCGTTGCC AAAGAGTGATTTG (corresponds to HXB2 2256-2278); the reverse primer differs from the HXB2 sequence at two nucleotide positions. A positive plasmid control and a negative control (no template) were included with each amplification run.

Amplification was performed on a 9700 Thermal Cycler (Applied Biosystems, Foster City, CA) using a 2-min 95 °C hold, followed by 45 two-step cycles of 94°C for 30s and 63 °C for 30s. The cycling was followed by a 94°C hold for 30s and a 28 °C hold for 30s. The resulting HIV gag amplicons (150- 190 base pairs) were then analyzed on a high-resolution melting instrument (LightScanner Model HR 96, Idaho Technologies, Salt Lake City, UT), following the

manufacturer's instructions, using a melt range of 68°C to 98°C with a hold at 65°C. Results were further analyzed using the labeled probe module of a high-resolution melting instrument (LightScanner system). The resulting data were processed to produce a melting curve for each sample representing the negative derivative of the data

(-d[fluorescence]/d[temperature] plotted against temperature). The left and right margins of each melting curve were marked at the positions at which the curve reached an angle of 30° relative to a horizontal baseline. The melting curves from duplicate analyses of a sample were almost always nearly identical, which was a prerequisite for further analysis. The difference in temperature between the left and right margins of a melting curve was defined as the HRM score.

HIV gag amplicons produced in the HRM assay were purified using the QiaQuick PCR Purification kit (Qiagen, Valencia, CA) and cloned using the TOPO TA Cloning kit (In vitro gen, Carlsbad, CA). A combined amplification/ sequencing method (AmpliSeq) was used to generate HIV gag sequences directly from bacterial colonies; AmpliSeq was performed, with 2 mM forward M13 primer (5 '-TGTAAAACGACGGCCAG) , 10 mM M13 reverse primer (5'- CAGGAAACAGCTATGACCA), 5xSequencing Buffer (Applied Biosystems), and ImM dNTPs. HIV gag sequences were obtained for 50 clones from each maternal sample and 20 clones from each infant sample.

Phylogenetic methods were used to confirm that the sequences from clones derived from each mother-infant pair grouped together (50 sequences from each maternal sample, 20 sequences from each infant sample, see above), without evidence of a sample mix-up.

Sequences from each plasma sample were aligned using MegAlign (DNAStar, Madison, WI) and manually edited to remove and/or align gaps in the sequences. The sequences from maternal samples (n = 450) and from infant samples (n = 180) were then aligned to a reference alignment (hiv.lanl.gov, subtype reference alignment version 2008) using the Profile Alignment method implemented in Clustal X2, with minimal manual editing to preserve highly conserved motifs. After alignment, sequences were trimmed to shared 5' and 3' termini. Sequence length was determined for each sequence as the number of nongap residues between those shared termini. For a given set of sequences (i.e., the set of clonal sequences from each sample), complexity was calculated as the number of unique sequence patterns n divided by the number of sequences N in that set (i.e., n/N), counting insertions or deletions as differences. Normalized Shannon entropy was calculated in a manner similar to complexity, except that for the n unique sequence patterns (individually represented i) observed at a frequency pi among N sequences, the normalized Shannon entropy was calculated as S = - (1= log N) P i = 1 pi log pi. Therefore, entropy takes into account the proportion or frequency of unique sequences in the set. For example, if a set of 50 sequences has two unique sequences at frequencies of 1/50 and 49/50, and another set of 50 sequences has two unique sequences, each at a frequency of 25/50, the complexity of both sequence sets would be the same (2/50, or 4%), but the entropy would be higher for the second set (0.025 vs. 0.177). Diversity was calculated for each set of sequences after removing sites containing gaps; diversity was calculated using Mega (version 4.1,

http://www.megasoftware.net) to apply the maximum composite likelihood distance model with rate variation among sites estimated using the gamma distribution (with alpha parameter 1.0).

The HRM assay was used for sample analysis, as follows. Using a high-resolution melting (HRM) instrument (LightScanner), the gag amplicons from a sample were heated, causing the DNA duplexes to melt and release the fluorescent dye that was incorporated into the amplicons during PCR (Figure 1A). By determining the slope of the fluorescence curve and inverting the curve (multiplying by -1), a melting curve for each sample was generated

(FigurelB, -d [fluorescence]/d[temperature] plotted against temperature). The left and right margins of the melting curve were marked, and the distance between these two margins was defined as the HRM score (Fig. IB). Figure 1C shows the melting curves for six control plasmids. The median HRM score for the plasmids was 3.4 (range 3.2-3.8). Because the plasmid templates were clonal, any diversity in the gag amplicons from these samples was likely to reflect errors introduced during PCR amplification.

The HRM assay was used to analyze the HIV samples from nine HIV-infected mother- infant pairs (Table 1 and Figure 2). The HRM scores for each woman and her infant were independent of one another (for the paired samples, Pearson correlation r = 0.204, p = 0.599). The median HRM score for HIV from the nine infants (4.3, range 4.2-5.3) was higher than that for control plasmids (p < 0.0001) and was lower than that for the HIV from nine mothers (5.7, range 4.4-7.7, p = 0.005, exact Wilcoxon rank sum test). Viral load data were available for 17 of the 18 samples; no association was found between HRM score and number of copies of HIV RNA analyzed (P = 0.081 for all 18 samples; P = 0.21 for 9 maternal samples only, exact Wilcoxon rank sum test).

Reproducibility of the HRM assay was assessed by analyzing the 18 samples four times each over the course of a year. For this analysis, DNA templates for the HRM assay (PCR products produced in the ViroSeq system) were stored at -80°C. For each run, samples were thawed and the HRM amplification and data analysis were repeated. Reproducibility of the HRM assay was high [intraclass correlation coefficient: 94% (95% CI: 89%, 98].

The results from the HRM assay were compared to the results obtained using sequence- based measures of HIV diversity. For this analysis, HIV gag amplicons produced in the HRM assay (Gag2 region) were cloned and sequenced (50 clones for each maternal sample, 20 clones for each infant sample). Sequences were analyzed for genetic diversity, complexity, Shannon entropy, and length variation (Table 1). Higher HRM scores were associated with higher genetic diversity (r = 0.76, P < 0.001), complexity (r = 0.59, P = 0.009), and Shannon entropy (r = 0.54, 0.022), but not with length variation (r = 0.39, P = 0.111).

TABLE 1

HRM Logio Genetic Median Length Length

Subject score HIV RNA diversity (%) length (bp) range (bp) diff (bp) Complexity Entropy

M-l 6.8 4.48 3.80 161 159-164 5 0.96 0.99

M-2 5.7 5.82 3.57 177 157-184 27 0.94 0.98

M-3 4.4 5.37 2.87 163 152-167 15 0.80 0.92

M-4 6.3 4.20 3.51 158 155-162 7 0.78 0.91

M-5 7.7 4.70 4.50 158 150-164 14 0.96 0.99

M-6 5.4 4.27 1.69 158 155-181 26 0.92 0.97

M-7 6.1 5.41 5.85 158 153-161 8 0.94 0.98

M-8 4.8 5.76 2.10 162 158-165 7 0.90 0.96

M-9 4.8 4.45 2.27 162 159-164 5 0.86 0.94

1-1 4.8 6.12 0.10 161 161 0 0.35 0.58

1-2 4.8 5.18 1.60 177 174-180 6 0.80 0.89

1-3 4.2 6.37 0.30 163 161-163 2 0.45 0.66

1-4 4.3 5.78 0.31 158 158 0 0.25 0.26

1-5 4.3 4.69 0.52 158 157-158 1 0.50 0.63

1-6 5.1 4.65 0.60 157.5 157-158 1 0.45 0.59

1-7 5.3 5.80 0.20 157 157 0 0.25 0.34

1-8 4.3 NA^b 0.10 162 162 0 0.15 0.13

1-9 4.2 5.50 0.51 161 160-161 1 0.55 0.65 aPlasma HIV from nine Ugandan mothers (M- 1 to M-9) and their infants (I- 1 to 1-9) was analyzed in the HRM assay. HRM scores are defined as the number of degrees centigrade over which melting of the DNA amplicons occurred (Figure IB). HIV gag amplicons produced in the HRM assay were cloned. Sequencing was performed for 50 clones from each maternal sample and 20 clones from each infant sample. For each sample, the sequences were used to determine HIV genetic diversity (%), the median length of the gag amplicons (base pairs, bp), the difference in length from the longest to the shortest gag region sequence (length diff, bp), complexity, and Shannon entropy. ^bNA, not available; insufficient sample for analysis.

Additional validation studies were performed as follows. The inventor recognized that many factors could influence the reproducibility of the HRM assay, including steps involved in preparation of the DNA templates, and steps involved in HRM analysis of those templates.

For example, there is potential to underestimate HIV diversity when analyzing samples with lower viral loads, smaller sample volumes, or lower numbers of HIV RNA used to prepare DNA templates for HRM analysis due to sampling error (bottlenecking). There is also a concern that there may be biased amplification of HIV variants during the PCR steps needed for DNA

template preparation and HRM analysis. The studies below were performed to evaluate the reproducibility of the HRM assay.

Clinical plasma samples were obtained from three individuals with high viral loads that had Gag2 HRM scores that were typical of those seen in adults with chronic HIV infection (results for the three 500 ul samples with viral loads of 50,000 copies/ml: 7.4, 6.7, 6.7). For reference, a set of plasmids (clonal DNA, no diversity) had a median Gag2 HRM score of 3.4 (range 3.2-3.8). The clinical samples were diluted with HIV-negative plasma to produce test samples that had viral loads ranging from 2,000 to 50,000 copies/ml. Those test samples were used to prepare DNA templates for the HRM assay, using either 100 ul or 500 ul of plasma for HIV RNA extraction. A total of seven test samples were prepared for each individual (21 total test samples).

The first step in preparation of DNA templates is isolation of HIV RNA from plasma. One fifth of the extracted RNA for reverse transcription (RT)/PCR. Using the test samples, the range of input HIV RNA used for RT/PCR ranged from 100 to 5,000 copies for each individual. The 21 extracted RNA samples were then subjected to RT/PCR in triplicate to generate DNA templates. The resulting 63 DNA templates were analyzed using the HRM assay (Gag2 region). Results from the RT/PCR reactions from one test sample (3 of the 63 results) were excluded from analysis for technical reasons. The results from this analysis are shown in Table 2.

Analysis of variance was used to assess the effect of HIV viral load and plasma volume on HRM scores, and separately for the maximum input number of HIV RNA copies used for RT/PCR (a number calculated from the HIV viral load and volume used for testing). Since three RT/PCR replicates were performed for each test sample (i.e., for each sample with a given HIV viral load and volume), the replicates were treated as within-factor replication, not true replicates. Therefore, the analysis used the mean HRM score for each sample (obtained by averaging results obtained for the HRM templates prepared in triplicate) to assess differences in means. The experiment was not balanced, thus Type 3 errors were used for the analysis. All of the analyses were performed in Splus using analysis of variance. Results are shown in Table 2 and Figure 12.

TABLE 2. Analysis of the effect of various factors on the reproducibility of the HRM assay.

Notes:

A and F are low volume samples; compare F to D and E (same RNA input).

D and E are replicates; compare to test the reproducibility of the extraction step

B, C, D, and G are standard volume samples with different viral loads

For each sample, compare replicates #1 -3 to test the reproducibility of the RT/PCR step There was no statistically significant change in HRM scores produced by varying the viral load of the sample or the volume of the sample used for analysis (for viral load: P=0.6; for sample volume: P=0.3; F test). There was also no statistically different change in the HRM scores obtained by varying the maximum input copies of HIV RNA used for RT/PCR, even when the lowest input copy number (100 copies of HIV RNA) was compared against all other samples (P=0.6, F-test). As described above, 60 HRM results were also obtained for twenty sets of RT/PCR reactions that were performed in triplicate. The coefficient of variation for those reactions was 0.064, which represents a 6.4% variation in the HRM score in the triplicate reactions. This indicates that variability in HRM score introduced during the RT and PCR steps of template preparation is relatively low. The peak melting temperature and shape of the melting curve were very similar among the replicates for each RNA sample, providing further evidence that there was little variation in the population of viral variants amplified in replicate RT/PCR reactions used to prepare DNA templates for HRM analysis. A study evaluating the

reproducibility of the nested HRM amplification step and melt curve analysis is described earlier in the application. In that study, the reproducibility of the HRM assay was assessed (analysis of DNA templates) by analyzing 18 DNA template samples four times during a one-year period. For that analysis, the DNA templates were stored at -80°C. For each run, the samples were thawed and HRM amplification and data analysis were repeated. That study demonstrated that reproducibility of the HRM assay was high (intra-class correlation coefficient: 94% [95% CI: 89%, 98%]).

These validation studies showed that results obtained with the HRM assay are not significantly affected by differences in HIV viral load (range: 2,000 to 50,000 copies/ml), sample volume (100 vs. 500 ul), or maximum number of HIV RNA copies used for DNA template preparation (range 100 to 5,000 copies of HIV RNA). DNA templates prepared using replicate RT/PCR reactions had similar HRM scores (coefficient of variation: 6.4%). There was very little variation in HRM scores obtained when the same DNA templates were analyzed repeatedly over the course of a year. These results provide support for the use of the HRM assay for analysis of HIV diversity using clinical plasma samples with variable viral loads, including low- volume infant samples.

The HRM assay described provides a rapid, high-throughput method for quantifying genetic diversity without sequencing. This method differs from the AmpliCot method, which has been used to analyze human genomic diversity (Baum et al., 2006, Nature Methods, 3:895-901). In AmpliCot, genetic diversity is determined by measuring the hybridization kinetics of DNA duplex formation. In contrast, the HRM assay uses the range of melting temperatures of DNA duplexes to measure HIV diversity. The HRM assay can detect low levels of sequence diversity (e.g., the level of HIV diversity in newly infected infants) and provides a simple measure of diversity, the HRM score, which is significantly associated with sequence-based diversity measures. The stability of DNA duplexes in a complex population is likely to be influenced by the number and type of nucleotide differences and insertions/deletions, as well as the proximity of these sequence differences to the ends of the duplex and to each other. For these reasons, the HRM score is a more complex measure of diversity than sequenced-based approaches that are based on relatively simple algorithms, such as the frequency of nucleotide differences in a sequence set (genetic diversity) or the frequency of unique sequences in a sequence set

(complexity).

The HRM primers used in this study were designed for amplification of several HIV subtypes, including A, B, C, and D; the HRM assay has been used to analyze HIV from >400 individuals infected with these HIV subtypes. The region selected for analysis within HIV gag (Gag2 region, Figure 3A) is poorly conserved across HIV strains and often contains both point mutations and insertions/deletions. This part of HIV gag is likely to evolve more slowly during HIV infection than the other regions, such as HIV env, but is still of interest since it contains epitopes that are targets for cytotoxic T lymphocytes. Recent studies using the HRM assay show that HRM scores based on analysis of the HIV gag p6 region vary with the stage of HIV disease in adults (acute < recent < chronic and AIDS) (Towler et al., 2010, 17^th Conf.on Retroviruses and Opportunistic Infections, San Francisco, CA, February, Abstract no. D-197). Because the portion of the gag region analyzed in this study is positioned immediately upstream of HIV protease, it is often included in amplicons produced in HIV genotyping assays used for antiretroviral drug resistance testing. Therefore, DNA remaining after genotyping can be used as a template for gag amplification in the HRM assay, which may be an advantage if primary samples are no longer available.

The median HRM score (reflecting HIV diversity in the gag region) was higher for women than infants (Towler et al, 2010, AIDS Res Human Retroviruses, 26, 913-918). The low HRM scores and low sequenced-based diversity measures obtained for infants are consistent with the fact that the infants were recently HIV infected. There is a wider range of variability in the HRM scores and sequence-based diversity measures for the women. Some of the women may have been recently infected or may have less diverse viral populations for other reasons. It was considered whether the viral populations in the infants may have been reduced by NVP exposure (genetic bottlenecking); however, the HRM scores of six infants in HIVNET 012 who were not NVP exposed (mother and infant received placebo) were similar to those of the NVP exposed infants (data not shown). Whether variations in HF/ viral load may have influenced the analysis was considered (e.g., sampling error in low viral load samples). This does not appear to be the case; there was no association between HRM score and viral load. Based on the viral load of the samples, the amount of plasma used for testing (500 μΐ for women, 100 μΐ for infants), and the amount of extracted HIV RNA used for amplification (1/5 of the RNA from each sample), it was determined that all of the samples tested had at least 889 copies of HIV RNA analyzed.

The HRM assay may facilitate studies of the complexity of HIV populations and the evolution of HIV diversity during infection. The HRM assay can be used to evaluate HIV diversity and evolution in infants, children, and adults, and to investigate the relationship between HIV diversity and other factors, such as HIV subtype and route of HIV infection. This study evaluated HIV diversity in the gag region of HIV in plasma samples, but the HRM assay could be applied to any region of the HIV genome, to HIV in other sample types, to other pathogens that can occur as mixtures of genetically related variants, and to other genetic systems that exhibit DNA diversity.

Example 2: Association of HIV diversity and survival in HIV-infected

Ugandan infants

The HRM assay was used to analyze HIV from infants (James et al, 2010, Submitted manuscript). In the HIVNET 012 trial, 37 infants in the sdNVP arm and six infants in the placebo arm were HIV-infected by 6-8 weeks of age. Samples were available for analysis from 31 of those 43 infants. Table 4 shows the characteristics of the 31 HIV-infected infants included in the sub- study and the 12 infants who did not have samples available for analysis. The only significant difference observed was that a higher proportion of infants in the group not included in the sub-study died during the follow-up period (P=0.05); none of those infants died by 6-8 weeks of age, so death of those infants was not related to their lack of inclusion in the sub-study.

For this study, summary statistics (median and range for continuous variables; frequency distributions for categorical variables) were provided for clinical and laboratory variables.

Fisher's exact test and the exact Wilcoxon rank sum test were used to compare clinical and laboratory variables between infants who were included in this sub-study and those who were infected by 6-8 weeks of age, but were not included due to lack of samples. Mean HRM scores at three different HIV genomic regions in either plasmids or infant samples from 6-8 weeks were compared using ANOVA. The Pearson correlation was used to compare the 6-8 week HRM scores in the three regions tested. Logistic regression modeled the probability of being in the high HRM group (HRM score above the 75^th percentile) and assessed the association between clinical and laboratory characteristics for HRM scores measured at 6-8 weeks. Firth's penalized likelihood approach was implemented to avoid bias in parameter estimates caused by small sample size.

For the analysis of infant survival, log rank tests were used to compare survival of infants categorized into high vs. low HRM groups based on the HRM score obtained from their 6-8 week samples; this was done separately for each region analyzed (Gagl, Gag2, Pol). The Kaplan Meier method was used to estimate survival functions for infants in the high vs. low HRM groups. Infants were censored at their last follow-up visit in the five-year follow-up of the HIVNET 012 trial. Since antiretroviral treatment may impact both infant survival and HIV diversity, the survival time for one infant who initiated antiretroviral treatment during the study was censored at the time of treatment initiation. The mean HRM score was calculated for the two gag regions and the mean HRM score for all three regions, and assessed the association of the upper quartile of the mean HRM scores with survival and other characteristics using methods stated above. The choice of breakpoint for the HRM scores was data driven, chosen amongst the 25^th, 50^th and 75^th percentile as best discriminating the 5-year mortality outcome; the 75^th breakpoint values were 5.0 for Gagl, 4.8 for Gag2, and 4.3 for Pol. A Cox proportional hazard model was used to assess the association between HRM score and survival, while adjusting for viral load at 14 weeks of age (log models 10 scale). Linear mixed - effect that took into account potential correlations among repeated measures from the same subject were used to evaluate the relationship between age and HRM scores obtained at 6-8 weeks, 12 months, and 18 months. All statistical analyses were performed using SAS version 9.2 (SAS Institute, Cary, NC, USA). Logistic regression was used to estimate odds of high vs. low HRM scores (above vs. below a median pre-HAART HRM score of 6) by age at HAART initiation, pre-treatment CD4 cell % or CD4 cell count (obtained within 30 days of treatment initiation), pre-treatment HIV viral load (obtained the day of treatment initiation), HIV subtype, and prior sdNVP exposure. HRM scores were also obtained for a subset of children who had >1,000 copies/ml HIV RNA at the 48-week or 96-week study visit; if the 48-week sample was not available, the sample collected at 96 weeks was used for analysis. A paired t-test was used to compare the difference in HRM scores before HAART vs. on non-suppressive HAART. The Wilcoxon rank sum test was used to compare the median age of children who had prior sdNVP exposure with those who were not exposed. Pearson correlation was calculated to assess the linear association between HRM score and viral load. Generalized Estimating Equations (GEE) methods were used to assess pre-HAART HRM score as a predictor of longitudinal virologic or immunologic response to HAART. In models using plasma HIV viral load, left-censored results (VL <400 copies/ml) were assigned a value of 200, and right-censored results (>750,000 copies/ml) were assigned a value of 750,000. All statistical analyses were performed using SAS version 9.1.3 on the SunOS 5.9 platform.

The HRM assay was used to measure HIV diversity in the gag and pol genes (Gagl, Gag2, and Pol amplicons, Table 3 and Figure 3). Most HIV infections in Uganda are caused by HIV subtypes A and D. As a control, the HRM scores of two subtype A and two subtype D plasmids were analyzed; because plasmids are clonal, the HRM scores of these control samples reflect the stability of the different duplexes and their melting characteristics, as well as any mutations introduced during DNA amplification. The HRM scores of the plasmid controls were low (for all three regions: median: 3.5, range: 3.2-3.8, Table 3). Plasmid HRM scores obtained for the Gagl region were slightly higher than those obtained for the Gag2 region (mean difference=0.38, P=0.008) and the Pol region (mean difference=0.35, P=0.011). HRM results for the Gagl and Gag2 regions were obtained for all 31 infants in the sub- study at 6-8 weeks of age; one infant did not have an HRM result for the Pol region due to amplification failure. The HRM scores of the infant samples were higher than those obtained for the plasmid controls, reflecting the natural genetic diversity of HIV viruses (Table 1; P<0.0001 for Gagl, P<0.0001 for Gag2, P<0.0001 for Pol). HIV viral loads were available for 11 of the 31 infant samples from 6-8 weeks of age that were tested with the HRM assay; those viral loads were high, indicating that sampling error was unlikely to account for the low level of diversity measured in some infant samples; however, some infants did not have viral load data obtained in the HIVNET 012 trial, and insufficient plasma remained to perform this testing for those infants. In the infant samples, as in the plasmid controls, the HRM scores obtained for the Gagl region were slightly higher than those obtained for the Gag2 region (mean difference=0.50,

P=0.003) and the Pol region (mean difference=0.75, P<0.0001). The correlation was evaluated between the 6-8 week HRM scores in the Gagl, Gag2, and Pol regions (Figure 5). This analysis revealed an association between the Gagl and Gag2 regions (r=0.47, p=0.008) and between the Gag2 and Pol regions (r=0.41, p<0.0001), but not between the Gagl and Pol regions (r=0.25, P=0.184). However, while this analysis revealed that while the low and middle range HRM scores were tightly clustered, high HRM scores in one region were not correlated with high HRM scores in the other region. This suggests that the degree of genetic diversity in the viral populations differed from one region to another, and supports the approach of analyzing the mean of HRM scores from two or three regions of the regions analyzed.

The association between clinical and laboratory variables and HIV diversity was assessed using the individual HRM scores obtained at 6-8 weeks for the Gagl, Gag2, and Pol regions, and also mean HRM scores of the Gagl and Gag2 regions, and mean scores of all three regions (Table 5). No significant associations were found between the HRM scores and the variables examined.

In this study, survival was reduced among infants who had higher HRM scores at 6- 8 weeks. In Kaplan Meier analyses, this association was not significant for the Gagl or Gag2 regions individually (Figure 4A and 4B), but was significant for the Pol region (P=0.005) and for the composite analyses (P=0.003 for the mean of results from Gagl and Gag2; P=0.002 for the mean of the results from all three regions). In multivariate proportional hazard models that included HIV viral load at 14 weeks of age and HRM score, higher HRM scores at 6-8 weeks (for Gag2, the mean of results from Gagl and Gag2, and the mean of results from all three regions) were independently associated with death (Table 6).

Of the 31 infants, 26 were alive at 12 months and twenty-one (81 %) of the 26 infants had samples available for testing from 12 or 18 months of age (17 and 15 infants, respectively). In the analysis of longitudinal HRM scores, it was found that higher HRM scores were associated with older age (Table 1, beta=0.37, P=0.005; for Gag2: beta=0.47, P=0.006; for Pol: beta=0.24, P=0.016; where beta is the estimated mean increase in the HRM score associated with one year increase of age). In this group of infants who were infected in utero or shortly after birth, age is likely to serve as a proxy for time since HIV infection.

TABLE 3

Use of HRM Assa to Anal ze Different Re ions of the HIV Genome.

Pr.¹ (Gagl forward): 5 ' - AAATTGC AGGGCCCCTAGGAA-3 ' ;

Pr.² (Gagl reverse): 5' -TTTCCCTAAAAAATTAGCCTGTCT-3' ;

Pr.³ (Gag 2 forward): 5 ' - ACTGAGAGAC AGGCT AATTTTTT AG-3 ' ;

Pr ⁴ (Gag2 reverse): 5 ' -GGTCGTTGCC AA AGAGTGATTTG-3 ' ;

Pr.⁵ (Pol forward): 5 ' - A A ATGG A A ACC A A A A ATG AT AG- 3 ' ;

Pr.⁶ (Pol reverse): 5 ' -CATTCCTGGCTTTAATTTTACTG-3 ' ;

a Control reagents; two subtype A plasmids and two subtype D plasmids.

b One infant did not have an HRM result for the Pol region.

c These HRM scores can be compared to those obtained in an observational study of 79 children in Uganda (Musoke PM et al., 2009, J Acquir Immune Defic Syndr, 52:560-568 ) (median age 4.7 years, range 0.6-12.4 years). In that study, the median HRM score in the Gag2 region was 5.9 (range: 3.8-11.9).

TABLE 4

Infants included Infants not included

at

Median maternal CD4 cell count at 477 ( 101, 1352) 379 ( 14.0, 1014) 0. ,22^d delivery (range)

Median maternal log 10 HIV viral 4.8 (3.9, 5.8) 5.0 (3.7, 5. ■7) 0. ,86^d load

at delivery (range)

Maternal HIV subtype

A 1 1 (35.5%) 4 (33.3%) 0. , 17^e

C 1 (3.2%) 2 (16.7%)

D 10 (32.3%) 3 (25.0%) R 3 (9.7%) 3 (25.0%)

Unknown 6 (19.4%) 0 (0.0%)

Single dose NVP exposure

Exposed 25 (80.6%) 12 (100.0%) 0.16^e

Non-Exposed ^b 6 (19.4%) 0 (0.0%)

Timing of infant HIV infection

HIV-infected in utero 20 (64.5%) 9 (75.0%) 0.72^e

HIV-infected after birth by 6-8 11 (35.5%) 3 (25.0%)

weeks

NVP resistance in infants at 6-8

weeks ^c

No resistance 10 (47.6%) 3 (100%) 0.22^e

Resistance 11 (52.4%) 0 (0.0%)

Death (5 years)

No 16 (51.6%) 2 (16.7%) 0.05^e

Yes 15 (48.4%) 10 (83.3%)

Proportion breastfeeding

At 6 months of age 27 (87.1%) 10(83.3%) 1.00^e

At 12 months of age 25 (80.6%) 10 (83.3%) 1.00^e

Initiated ART during HIVNET 012

Yes 1 (3.2%) 0 (0.0%) 1.00^{e a} Among 43 infants in the HIVNET 012 trial who were infected at birth or by 6-8 weeks, 36 had viral load data obtained at 14 weeks of age (26 included in the analysis and 10 not included in the analysis). There was insufficient viral load data available from 6-8 weeks for meaningful statistical analysis. ^b Infants with in utero HIV infection had a positive HIV DNA test at birth.

c For 19 infants (10 included in the analysis and 9 not included), either a sample was not available for

HIV genotyping or a result was not obtained.

d Two-sided exact p-value, Wilcoxon rank sum test.

e Two-sided p-value, Fisher's exact test.

TABLE 5

Association of HRM score with clinical and laboratory variables obtained for infants (HIVNET 012 trial, Uganda,

*OR: Odds ratio, CI: confidence intervals; Odds ratio, 95% confidence intervals and p-values were obtained from a logistic regression that modeled the probability of being in the high HRM group (above the 75^th percentile, >Q3) with Firth's penalized likelihood approach.

a Among the 31 infants with 6-8 week HRM results, 26 (83.9%) had viral load data obtained at 14 weeks of age.

There was insufficient viral load data available from 6-8 weeks for meaningful statistical analysis. Viral load was treated as a continuous variable in the analysis.

TABLE 6

Association of HRM score and survival in univariate models and multivariate models that included HRM score and HIV viral load.

HRM score at 6-8 weeks N Hazard Ratio P N Hazard Ratio P

(95% CI) (95% CI)

Unadjusted Adjusted¹

Gagl 31 2.0 (0.7, 6.0) 0.19 26 2.1 (0.7, 6.6) ² 0.19

2.6 (0.8, 9.2) ³ 0.12

Gag2 31 2.5 (0.8, 7.3) 0.11 26 3.5 (1.1, 10.9)² 0.03

2.4 (0.7, 7.8) ³ 0.15

Pol 30 4.7 (1.4, 15.8) 0.01 25 3.4 (1.0, 11.9) ² 0.06

2.7 (0.7, 10.6) ³ 0.15

Mean (Gagl, Gag2) 31 4.2 ( 1.5, 1 1.9) 0.006 26 8.7 (2.6, 28.6)² 0.0004

3.4 (1.1, 10.8) ³ 0.04

Mean (Gagl, Gag2, Pol) 30 4.6 ( 1.6, 13.2) 0.004 26 6.9 (2.1, 22.9)² 0.002

1.7 (0.5, 5.3)³ 0.40

¹ Two covariates were included in each multivariate model: HRM score measured at 6-8 weeks of age (binary, < 75^th percentile vs. > 75^th percentile) and HIV viral load measured at 14 weeks of age (log scale). N: number of infants included in the model; CI: confidence intervals.

2 Hazard ratio, 95% CI, and P value for HRM score at 6-8 weeks of age.

³ Hazard ratio, 95% CI, and P value for HIV viral load at 14 weeks of age (log).

A previous study of Ugandan mother-infant pairs which included nine infants in this study, confirmed that HRM scores are significantly associated with sequence-based measures of HIV diversity (genetic diversity, as well as complexity, and Shannon entropy)(Towler WI et al., 2010, AIDS Research and Human Retroviruses, 26:913-918). When the HRM assay is used to analyze complex molecular populations, such as DNA amplified from HIV in clinical samples, the melting temperatures of the DNA duplexes may be influenced by a variety of factors, including the number and type of nucleotide mismatches and insertions / deletions in the duplexes, as well as the proximity of those sequence differences to each other and to the ends of the duplex (Towler WI et al., 2010, 17th Confon Retroviruses and Opportunistic Infections, San Francisco, CA, 2010:Abstract #267). For those reasons, HRM scores provide a more comprehensive measure of diversity than traditional, sequenced-based approaches based on simple algorithms (e.g., the frequency of nucleotide differences in a sequence set). Melting curves can be generated using other instruments that measure incorporation/release of a fluorescent dye, such as those designed for real-time PCR. The LightScanner instrument was chosen for the HRM assay because it was specifically designed for melt curve analysis and includes software specifically designed for high resolution melting applications. The

LightScanner also has greater data density and greater temperature accuracy than other instruments (Herrmann MG et al., Clin Chem, 53:1544-1548). Greater data density and use of the saturating LCG+ dye are features of the LightScanner system that improve the sensitivity and accuracy of heteroduplex detection for applications such as this one that involves measuring the width of the derivative melt curve, rather than the peak melting temperature.

Numerous studies have evaluated changes in HIV diversity that occur during the course of HIV infection in adults (Frost SD et al., 2005, Proc NatlAcad Sci USA, 102:18514- 18519; Mullins JI, 2006, Curr Top Microbiol Immunol, 299:171-192; Sagar M et al., 2006, AIDS Res Hum Retroviruses, 22:430-437; Shankarappa R et al., 1999, J Virol, 73:10489-10502). In adults, while HIV infection is usually initiated by one or a few HIV variants, some individuals are infected with multiple HIV strains (Bar KJ et al., 2010, J Virol, 84:6241 -6247; Keele BF & Derdeyn CA, 2009, Curr Opin HIV AIDS, 4:352-357; Keele BF et al., 2008, Proc Natl Acad Sci USA, 105:7552-7557; Long EM et al., 2000, Nat Med, 6:71-75). When multiple HIV variants are transmitted, forces act very early in infection to select one or a few founder strains, leading to homogenization of HIV env sequences (Learn GH et al., 2002, J Virol, 76:11953-11959).

Antibody responses begin to contribute to genetic selection of HIV a few months after infection (Keele BF, 2010, Curr Opin HIV AIDS, 5:327-334). In the first few years of infection, env diversity tends to increase in a linear fashion (Mullins JI et al., 2006, Curr Top Microbiol Immunol, 299:171-192; Shankarappa R et al., 1999, J Virol, 73:10489-10502). At some point, env diversity may stabilize or even decrease. Late in disease, env sequences often become homogeneous as the immune system collapses (Mullins JI et al., 2006, Curr Top Microbiol Immunol, 299:171-192; Shankarappa R et al., 1999, J Virol, 73:10489-10502). While less information is available for other regions of the HIV genome, it is clear that different HIV genes/gene products are subjected to different selective pressures; for example, while env is the major target for anti-HIV antibodies (Baum LL, 2010, CurrHIV/AIDS Rep, 7:11-18), gag selection is mediated predominantly by cytotoxic lymphocytes (CTLs) (Piantadosi A et al., 2009, AIDS, 23:579-587). Interestingly, the homogenization that is seen very early in infection in env does not appear to occur in gag (Learn GH et al., 2002, J Virol, 76:11953-11959). Later in infection, HIV env and gag evolution is convergent in some individuals (Piantadosi A et al., 2009, AIDS, 23:579-587).

The patterns of HIV diversification and homogenization are likely to be different in infants and adults for several reasons. First, vertical transmission results from exposure of infants to a single source: HIV from the mother. In contrast, adults often have multiple independent HIV exposures leading to HIV infection, which could influence the multiplicity of HIV infection. There are also marked differences in viral dynamics in infants and adults. In adults, there is a rapid decline in viral load shortly after HIV infection, and a viral load set point is usually established during the first few months of HIV infection. In one study, higher HIV diversity was seen in women with higher viral load set points (Delwart EL et al., 1997, J Virol, 71:7498-7508). In contrast, viral loads usually remain very high in infants during the first year of life, and then decline slowly over the next few years of HIV infection, usually remaining at levels higher than those seen in adults (Huang S et al., 2008, J Immunol, 181:8103-8111; Shearer WT et al., 1997, N Engl J Med, 336:1337-1342). These differences in viral dynamics may reflect the significant differences in the immune systems of newborn infants and adults. Also, infection of adults occurs in the context of a mature immune system. Antibody responses to HIV infection are typically detected in the first few weeks following infection. In contrast, the immune system in newborn infants is immature. Production of anti-HIV antibodies occurs much later in infants, which is why antibody-based assays are not recommended for HIV diagnosis in children under 18 months of age. Cellular responses to HIV infection are also different in children and adults. Most HIV-infected infants have deficient cytoxic T lymphocyte (CTL) responses to HIV (Buseyne F et al., 1993, J Immunol, 150:3569-358; Luzuriaga K et al., J

Pediatr, 119:230-236), and often have inadequate CD4 cell count help (Huang S et al., 2008, J Immunol, 181:8103-8111; Thobakgale CF et al., 2007, J Virol, 81:12775-12784). These and other factors are likely associated with observed differences in cellular populations during HIV infection in adults vs. infants. In adults, a decline in CD4 cell count is associated with disease progression; in contrast, in children under 5 years of age, a decline in CD4 cell % is a more reliable biomarker of disease progression. Furthermore, infection of infants with variants that were able to escape the mother's CTL response may further hinder the infant's ability to contain the virus (Shalekoff S et al., 2009, AIDS, 23:789-798). For these reasons, the changes in HIV diversity overtime, and the relationship of HIV diversity to disease progression, are likely to be different in pediatric and adult populations. In this study, HIV diversity in infants measured using the HRM assay tended to increase during the first 12-18 months of life. It is not known if this increase in HIV diversity over time reflects direct selective pressure for evolution (e.g., a response to cytotoxic lymphocytes targeting gag or pol epitopes (Frahm N et al., 2007, Curr Infect Dis Rep, 9:161-166), linkage to another region that is the target of selective pressure (e.g., antibody-induced selection of env epitopes), or natural accumulation of mutations in the HIV genome without selective pressure.

The major finding in this study is that higher levels of HIV diversity near the time of birth (higher HRM scores) were significantly associated with decreased infant survival. This association was seen both for the HIV gag region (averaging results obtained for the Gagl and Gag2 amplicons) and the HIV pol region. This study differs from previous reports that examined the association between changes in HIV diversity over time in HIV-infected children and disease progression (Zhang H et al., 2006, Retrovirology, 3:73; Ganeshan S et al., 1997, J Virol 71:663- 677; Strunnikova N et al., 1995, J Virol, 69:7548-7558; Halapi E et al., 1997, AIDS, 11:1709- 1717). Those studies included smaller numbers of children (five to seven in each study), used sequence-based measures for diversity analysis, analyzed the env region (rather than gag and pol), and focused on changes in diversity overtime, rather than the level of HIV diversity early in infection. Further studies are needed to identify factors that influence the genetic diversity of HIV in young infants. In theory, maternal factors, such as high HIV viral load, high viral diversity, advanced HIV disease, or a complicated delivery, could be associated with exposure of the infant to a higher and/or more diverse viral inoculum, which could lead to establishment of infant infection with a greater number of distinct HIV variants. The genetics of the infant (e.g., HLA type, co-receptor expression) could also potentially influence the type and complexity of the viral population early in infection. If the viral population in an infant were more diverse, it might be more likely to escape immune and other selective pressures, leading to more rapid HIV disease progression. Alternatively, high levels of HIV diversity in young infants may be a surrogate marker for infection with viral variants that have more error-prone reverse transcriptase enzymes or higher rates of HIV replication; viruses with those properties might be more likely to escape immune or other selective pressures, or might cause more immune destruction over time because of increased viral replication.

The mortality in the present cohort was lower than what is usually seen among HIV- infected infants in sub-Saharan Africa. In the HIVNET 012 trial (source of the samples used in this study), the 5-year mortality was 55%, similar to the mortality seen in the subset of infants analyzed in this report. The lower mortality of infants in the HIVNET 012 trial could have reflected an effect of antiretro viral drug prophylaxis or other factors, such as enrollment into a clinical trial with access to free treatment for acute illnesses, prophylaxis for other infections, immunization, and other care that may have impacted their outcome.

Example 3: Higher HRM scores were associated with immunologic status and reduced immunologic response to HAART.

The HRM scores were measured using samples collected from children in an

observational study prior to HAART initiation (pre-HAART samples). Characteristics of the children, including age at HAART initiation, pre-HAART CD4 cell count, pre-HAART CD4 cell %, pre-HAART baseline HIV viral load, HIV subtype, and prior sdNVP exposure are shown in Table 7. In this cohort, the median age of the sdNVP-exposed children was lower than that of the sdNVP-unexposed children (median: 1.7 years, range 0.6-6.3 years, vs. median: 7.7 years, range 2.9-12.4 years, p<0.0001, Wilcoxon rank sum test), reflecting the fact that many of the older children were born prior to widespread availability of sdNVP prophylaxis. The median pre-HAART HRM score for all 79 children was 5.9 (range: 3.8-11.9). In univariate models, higher pre-HAART HRM scores were significantly associated with older age, lower pre-HAART CD4 cell count, lower pre-HAART CD4 cell %, and lack of sdNVP exposure (Table 8). In a multivariate model including age, pre-HAART CD4 cell % and pre-HAART HIV viral load as covariates, both a lower CD4 cell % and older age remained highly associated with higher HRM score (adjusted odds ratio (OR) for CD4 cell %: 5.4, 95% confidence intervals (CI): 1.7, 17.2), p=0.004; OR for age: 1.3 (1.1, 1.6), p=0.005, Table 8). As age and sdNVP exposure were highly confounded in this observational study, sdNVP was excluded from the multivariate model, as an association of increased HIV diversity with older age (reflecting a longer duration of untreated HIV infection) was more biologically plausible than an association of increased HIV diversity without prior sdNVP exposure.

The association of pre-HAART HRM score with immunologic response to HAART was examined. Due to the small size of the study cohort, there was not enough power to perform this analysis separately for children who started HAART before age 5 (using CD4% as the outcome variable) and children who started HAART at or after age 5 (using CD4 cell count as the outcome variable). Instead, the immunologic response to HAART was examined in the entire cohort using separate GEE models for CD4% and CD4 cell count as outcome variables.

Seventy-one of the 79 children had CD4 data obtained at the time of HAART initiation. Ten of the 71 had severe immune compromise prior to HAART (CD4 cell % < 5%); those children had a high rate of treatment failure and were excluded from the analysis described below. The association of pre-HAART HRM score with immunologic outcome in the remaining subset of 56 children who had CD4 data obtained after HAART initiation was determined (141 observations, all available data, collected every 12 weeks after 24 weeks on HAART).

Using CD4 cell % as the outcome variable, both lower pre-HAART HRM score and higher pre-HAART CD4 cell % were associated with an improved immunologic response to HAART in univariate models. However, in a multivariate model including both variables, only pre-HAART CD4 cell % was associated with an improved immunologic response to HAART (Table 9). In the multivariate model, for every 1% increase in pre-HAART CD4 cell %, there was an increase in CD4 cell % of 0.95% on HAART (p<0.0001). In contrast, when CD4 cell count was used as the outcome variable, both lower pre-HAART HRM score and higher CD4 cell count were independently associated with an improved immunologic response to HAART (Table 9). In the multivariate model, for every unit increase in pre-HAART HRM score, there was a decrease in CD4 cell count of 63.6 cells/mm on HAART (p=0.016). In the same multivariate model, for every 100 cells/mm increase in pre-HAART CD4 cell count, there was an increase in CD4 cell count of 87.7 cells/mm³ on HAART (p<0.0001).

The relationship between the pre-HAART HRM score and virologic suppression after 24 weeks on HAART was analyzed, defined as a binary outcome of HIV viral load <400 copies/ml, allowing one "blip" not higher than 1,000 copies/ml. During follow-up, 50 (66.7%) of the children achieved virologic suppression while 25 (33.3%) of the children did not. The two groups had a similar number of HIV viral load results obtained after 24 weeks [suppressed: median 4 results (range 2-6) vs. non-suppressed: median 4 results, range 2-7)] and were followed for a similar length of time on HAART [suppressed: median 96 weeks (range 47-101) vs. non- suppressed median 96 weeks (range 47-109)]. No association between pre-HAART HRM score and longitudinal virologic suppression was found (OR: 1.1, 95% CI: 0.4, 2.9, p=0.82).

Pre-HAART HRM scores were compared to HRM scores obtained in a subset of children who were maintained on HAART without virologic suppression. This analysis included only the subset of 15 children who had both a pre-HAART sample and sample collected at 48- or 96- weeks with a viral load >1,000 HIV RNA copies/ml (Table 10). In 12 of those 15 children, the HRM score obtained while on non-suppressive HAART was lower than the pre-HAART HRM score. The mean change between pre-HAART HRM score and HRM score on non-suppressive HAART was -1.08 (two-sided p=0.001). The possibility that lower HRM scores obtained while children were on non-suppressive HAART might reflect lower viral loads in those samples (i.e., that sampling error may have biased the analysis of HIV diversity) was considered. The mean logio HIV viral loads in the pre-HAART samples and the samples obtained on non-suppressive HAART were 5.6 and 4.5, respectively (p<0.0001). However, there was no association between HRM score and HIV viral load in these samples (R=0.19, p=0.50, Pearson correlation).

TABLE 7

Characteristics of children in the observational study (enrollment 2004-2006)

Variable N Result

Median age at HAART initiation in years (range) 79 4.7 (0.6-12.4)

Median pre-HAART CD4 cell count (range)

All children (n=71) 71 486 (2-3637)

Children > 5 years of age 35 250 (2-832)

Median pre-HAART CD4 cell % (range)

All children 72 11.1 (0.2-22.7)

Children < 5 years of age 36 13.6 (4.7-22.7)

Median pre-HAART logio HIV viral load (range) 75 5.6 (3.0-5.9)

HIV subtype 79

A 38 (48.1%)

C 1 (1.3%)

D 22 (27.8%)

R 13 (16.5%)

Unknown 5 (6.3%)

Prior single dose nevirapine exposure 79 36 (45.6%)

Median pre-HAART HRM score (range) 79 5.9 (3.8-11.9)

HAART: highly active antiretroviral therapy; N: number of children with data available for analysis.

TABLE 8

Association of pre-HAART HRM score with clinical and laboratory variables for children in the observational study

Variable Category Univariate Model⁹ Multivariate Model'^{1 0 c} Multivariate

Model^{a b d} OR (95% CI) p value OR (95% CI) p OR (95% p value CI) value

Age 1 .3 (1 .1 , 1 .6) 0.005 1 .3 (1 .0, 0.045

1 .4 (1 .2, 1 .7) <0.0001

(per year increase) 1.7)

Pre-HAART CD4 < Median vs. > median 1.7 (0.4, 0.49

5.7 (2.0, 15.8) 0.0008

cell count 7.8)

Pre-HAART CD4 < Median vs. > median 5.4 (1 .7, 0.004

6.8 (2.4, 19.0) 0.0003

cell % 17.2)

< Median vs. > median 0.8 (0.2, 2.9) 0.78 0.8 (0.3, 0.77

Pre-HAART HIV

1.6 (0.7, 4.0) 0.30 2.7) viral load

Prior sdNVP Exposed vs. non- 0.07 (0.02,

<0.0001

exposure^b exposed 0.21 )

HIV subtype⁶ A vs. Non-A 1.3 (0.5, 3.1 ) 0.63

a Odds ratios (OR), 95% confidence intervals (CI), and p-values are shown. The logistic regression model of pre- HAART HRM score was based on categories < or > the median of 6. The probability of having a higher HRM score was modeled.

b Prior sdNVP exposure was not included in the multivariate model to avoid confounding.

c Covariates of this multivariate logistic regression model include age, pre-HAART CD4 cell %, and pre-HAART HIV viral load.

d Covariates of this multivariate logistic regression model include age, pre-HAART CD4 cell count, and pre- HAART HIV viral load.

e Similar results were obtained when comparing subtype D vs. non-D (univariate model, P=0.84). TABLE 9: Association of pre-HAART HRM score with immunologic response to HAART (GEE models)*

Pre-HAART CD4 cell % as outcome CD4 cell count as outcome

Variable

Univariate Model Multivariate Univariate Model Multivariate

Model Model

Estimate p value Estimate p value Estimate p value Estimate p value

HRM score -1.48 0.01 -0.76 0.15 -107.5 0.01 -63.6 0.016

CD4 cell % 1.03 <0.0001 0.95 <0.0001

CD4 cell count 90.7 <0.0001 87.7 <0.0001

*The analysis was based on 56 children followed for up to 96 weeks; children with severe immunoloi compromise at the time of HAART initiation (CD4 cell % < 5%) were excluded from the analysis.

TABLE 10: Comparison of pre-HAART HRM scores to HRM scores at the 48- or 96-week visit among children who were maintained on antiretro viral therapy without virologic suppression.*

Child HRM HRM I I KM

VL

Score VL VL VL Score VL VL VL Score Pre- Pre- 24 wks 36 wks 48 wks 48 60 wks 72 wks 96 wks 96 HAART

HAART wks

715,877 ft Ί <400 1,246 10,704 4 0:

580,717 4 ft 98,719 168,610 58,828 5.5

>750,000 s si 58,244 39,452 16,403 5.0:

>750,000 41 233,684 457,203 225,2 I S 4 '

>750,000 4.7: 494 762 38,937 4.::

>750,000 4 115,654 20,090 2,398 144,182 100,227

622,414 - I 1,020 >750,000 302,282 4.2 442,454 337,903 156,809

47,095 10 4 22,375 34,429 17,400 ^«). |; 12,687 14,733

468,974 ft '⁾ i 39 218,324 67,150 5.8 56,428 237,244

10 477,812 fi 4! 15,971 20,400 196,81s 5.'): 128,982 30,349

11 421,825 ft 4 58,692 73,536 47,478 ft 4: 35,110 36,279

12 >750,000 5.4 27,020 36,190 185,10ft 4.:: 101,992 120,010 4:

13 76,131 ft 4 <400 <400 22,696 1,345

14 183,871 4.'⁾ <400 <400 3,546 57,457

15 646,450 5.4 <400 <400 <400 2,622

*Children in the observational study were followed for up to 96 weeks on antiretroviral treatment (see Methods).

High resolution melting (HRM) scores were measured for children who had >1,000 HIV RNA copies/ml at the 48- and/or 96-week study visit. Note that the median HRM score for control plasmids is 3.4 (range 3.2-3.8). Pre- HAART samples were collected on the day of treatment initiation. All available viral load results (VL, HIV RNA

copies/ml) are shown. In a previous study, it was demonstrated that children #1-12 had resistance to both nucleoside reverse transcriptase inhibitors (NRTIs) and non-nucleoside reverse transcriptase inhibitors (NNRTIs) at 48 weeks

(Towler WI et al., 2010, AIDS Res Hum Retroviruses 2010, 26: 563-568). At 96 weeks, four of those children (#7, 9,

11, and 12) acquired additional NRTI resistance mutations (Towler WI et al., 2010, AIDS Res Hum Retroviruses

2010, 26: 563-568). Child #13 had both NRTI and NNRTI resistance at 96 weeks due to the presence of the M184V

and G190A mutations (unpublished data); resistance results could not be obtained for children #14 and #15 at 96

weeks. Using the HRM assay, a strong association was found between a high level of HIV diversity in

the gag region and low CD4 cell % in HIV-infected children in both univariate and adjusted

models. An association was also found between older age and lack of sdNVP exposure with

higher HRM scores in univariate analysis. However this likely reflected the demographic characteristics of the children enrolled in the observational study. In this study, children who were older and more likely to have had a low CD4 cell count and low CD4 cell % were also less likely to have received sdNVP for prevention of mother-to-child transmission of HIV (pMTCT), since sdNVP was not widely available for pMTCT in Uganda when they were born. The HRM assay used in this study measures genetic diversity in the p6 region of HIV gag. It is not clear whether the higher diversity it was observed in children with low CD4 cell % reflects selective pressures acting on epitopes in the region analyzed (e.g., cyotoxic T lymphocyte (CTL) responses directed toward p6 epitopes), or selective pressures acting on other HIV proteins (e.g., antibody or CTL responses directed toward env epitopes), with convergent evolution in HIV gag due to genetic linkage. Some studies in adults using more traditional measures of viral diversity have found convergent evolution in the env and gag regions, while others observed that these regions diversified at different rates . (Liu S. et al., 2008, China. Arch Virol 2008, 153:1233- 1240; Piantadosi A. et al., 2009, AIDS 2009, 23:579-587). The HRM assay could be easily adapted for analysis of HIV diversity in other regions of the HIV genome. Analysis of the degree of diversity in different regions (e.g., HIV env, other portions of HIV gag) in children with different levels of immune compromise may help define the relationship between viral diversification and immune status in HIV-infected children.

It was also found that high HRM scores were associated with more advanced HIV disease in adults (acute infection < recent infection < chronic infection and AIDS). However, in that study, there was no association between HRM score and CD4 cell count (Towler WI, et al., 2010, In 17th Confon Retroviruses and Opportunistic Infections; February 16-19; San

Francisco, CA. 2010. Abstract #267). Other studies have shown a relationship between HIV env diversity and disease progression in adults (Gottlieb GS et al., 2008, J Infect Dis 2008, 197:1011- 1015; Sagar M et al., 2003, J Virol, 77:12921-12926). However, it is important to note that there are differences in viral dynamics and immune response in HIV-infected children and adults. In HIV-infected children, viral load declines more slowly than in adults and CTL responses occur later ( Leal E et al., 2007, Infect Genet Evol, 7:694-707). In adults and children 5 years of age or older, CD4 cell counts are used to monitor immunologic status ( Dunn D et al., 2008, J Infect Dis , 197:398-404). In contrast, CD4 cell % is often used to assess immunologic status in younger children because there is less fluctuation in CD4 cell % compared to CD4 cell count in younger children (Lancet 2003, 362:1605-1611; AIDS 2006, 20:1289-1294). In this cohort, an association between pre-HAART HRM scores and virologic response to HAART was not seen. However, among children who had a CD4 cell % > 5% at the time of HAART initiation, higher pre-HAART HRM scores were associated with a reduced immunologic response to HAART, when CD4 cell count was used to evaluate treatment response.

In adults an association between nucleoside reverse transcriptase inhibitor (NRTI) use and selection of mutations in the HIV p6 gag region was found (Peters S et al., 2001, J Virol 2001, 75:9644-9653; Ojesina AI et al., 2008, AIDS Res Hum Retroviruses, 24:1167-1174). In this study, a reduction in genetic diversity in the p6 region in clinically and immunologically stable children who were maintained on non- suppressive HAART was found. This observation was based on analysis of a subset of 15 children who had an HIV viral load >1,000, 48 and/or 96-weeks after HAART initiation; most of these children had high viral loads for 6 months or more prior to those study visits. The reduction in HIV gag diversity that it was observed in these children most likely reflects bottlenecking of the viral population due to selection of drug- resistant HIV variants. In a previous study, it was demonstrated that all 12 of the children in this study who had >1,000 copies of HIV RNA after 48 weeks on non-suppressive HAART had resistance to both NRTIs and non-nucleoside reverse transcriptase inhibitors (NNRTIs). Some children who still had >1,000 copies of HIV RNA at 96 weeks acquired additional resistance mutations (Towler WI et al., 2010, AIDS Res Hum Retroviruses 2010, 26: 563-568). The reduction in HIV diversity that was observed in this cohort in children maintained on a non- suppressive HAART regimen may be relevant to use of non-suppressive antiretroviral regimens in resource-limited settings that have limited access to second-line treatment regimens, or in other clinical settings, such as management of patients with multi-class antiretroviral drug resistance for whom treatment options are limited, or use as a bridging strategy in patients failing HAART due to non-adherence (Abadi J et al., 2006, JAcquir Immune Defic Syndr, 41:298-303; Castagna A et al., 2006, Aids, 20:795-803).

An association between high pre-HAART HRM scores (reflecting a high level of genetic diversity in the HIV gag region) and low CD4 cell % in HIV-infected children was identified. In children who did not have severe immune compromise prior to treatment, a higher pre-HAART HRM score was associated with a reduced immunologic response to HAART. Among children who did not achieve virologic suppression on HAART, maintenance of those children on non- suppressive HAART was associated with a reduction in HRM score, reflecting bottlenecking of the viral population.

Example 4: HRM can discriminate between recent and non-recent HIV infections.

HIV populations diversify after infection due to a high mutation frequency (lack of proofreading by HIV enzymes) and diverse selective pressures. HIV transmission from one individual to another creates a bottleneck in viral diversity and new infections are typically established with one or a few viral variants. Shortly after infection, numerous factors act as selective pressures, increasing the diversity of the viral population. Viral diversity allows the virus to escape immune and other selective pressures. High diversity early in infection has been associated with more rapid disease progression.

Most studies of HIV diversity have focused on changes in the env and/or gag genes over time. Over the course of infection, gag and env diversity may occur concurrently or discordantly. Diversity studies are often limited by the effort needed to analyze individual HIV variants. Most studies have examined relatively small numbers of subjects and singular regions of the HIV genome. Most studies of HIV diversity and evolution have been based on sequence-based analyses of individual HIV variants in an infected individual (e.g., by cloning and limiting dilution or single genome sequencing). Methods such as pyrosequencing have also been used in some studies. Heteroduplex mobility or tracking assays have also been used, but provide more limited information and can be difficult to interpret. The HRM assay provides a simple, rapid method for quantifying the level of HIV diversity by generating a single numeric score that reflects the level of genetic diversity in the region analyzed. Multiple platforms can be used for HRM analysis, including the LightScanner (Idaho Technology, Salt Lake City, UT) platform which has very good temperature accuracy and data density and is provided with useful software applications.

HIV genetic diversity was compared in four regions in HIV in samples collected from adults with different stages of HIV disease. Source of samples used for analysis: EXPLORE STUDY: Samples from adults with acute (pre- seroconversion) and recent HIV infection were obtained from the EXPLORE (HPTN 015) study, which compared the effectiveness of behavioral intervention with standard risk reduction counseling for prevention of HIV infection. This study enrolled >4,000 HIV-uninfected men who have sex with men (MSM) in six cities in the U.S. (1999-2001) and tested them every 6 months for HIV infection; 259 men acquired HIV infection during the study. Samples collected at the time of HIV seroconversion or shortly thereafter were available from 102 men (recent samples, collected a median of 187 days after the last negative HIV test [range 14-540 days]). Samples collected from the visit prior to HIV seroconversion that tested HIV negative in the EXPLORE study were screened for HIV RNA and re-tested using second and third generation enzyme immunoassays in a retrospective sub- study. Twenty HIV RNA positive, antibody negative samples were available for this study (acute samples). Antiretroviral drug resistance, HIV tropism, and HIV subtype were determined in a previous study, and all infections were caused by be subtype B virus.

JHH Serosurvey samples were collected from patients in The Johns Hopkins Hospital

(JHH) Emergency Department (ED) as part of an on-going volunteer, risk-targeted HIV serosurvey study. Plasma aliquots collected from 11 patients in 2001 and 2003 were available for analysis and were included in this study. The JHH ED serves a local inner-city population made up primarily of socioeconomically disadvantaged minorities, approximately 10-12% of whom are infected with HIV.

Templates for HRM analysis of a region of gag were prepared using the ViroSeq system (ViroSeq, Celera, Alameda, CA) as previously noted. Templates for HRM analysis of three regions of env were prepared using the method that follows. RNA extraction was carried out using the ViroSeq system. A region of env (gpl60) was reverse transcribed and amplified using the Qiagen OneStep RT-PCR Kit (QIAGEN Inc., Valencia, CA). RT-PCR reactions contained 10 μΐ of viral RNA extraction, lx Qiagen OneStep RT-PCR Buffer, 400 μΜ dNTP Mix, 2 μΐ Qiagen OneStep RT-PCR Enzyme Mix, 20 μΜ forward and reverse primers, and RNase-free water for a total reaction volume of 50 μΐ. Reverse transcription and amplification were carried out on a 9700 Thermal Cycler (Applied Biosystems, Foster City, CA). Initially, the 10 μΐ RNA extract aliquots were subjected to a 5 minute hold at 65° C to denature RNA secondary structure. The RT-PCR mastermix, containing all other reaction components, was subsequently added to the RNA extract, and the completed reaction mix was returned to the thermocycler. The RT-PCR began with a 45 min. 50° C hold followed by a 15 min. 95° C hold. The method continued with 20 cycles of 94° C for 15 sec, 60° C for 30 sec, and 68° C for 60 sec followed by 30 cycles of 94° C for 15 sec, 60° C for 30 sec, and 68° C for 90 sec. Cycling concluded with a 7 min 68° C hold and an indefinite hold at 4° C. Primer sequences were as follows: Forward, JH35F (5'- TGARGGAC AATTGG AGAARTGA-3 ' ) and Reverse, JH38R (5 -

GGTG ART ATCCCTKCCT A AC- 3 ') (Mani I et al., 2002, / Virol, 76:10745-10755). Templates were diluted and purified according to the ViroSeq system using a gel-based dilution method and ExoSAP-ΓΓ™ (UBS, Cleveland, OH) purification. Purified templates were diluted 1:10 prior to HRM analysis.

PCR amplicons generated utilizing the methods described above were diluted and used as template DNA for HRM analysis. A region of each PCR product was amplified in the presence of a fluorescent dye (LCGreen® Plus, Idaho Technology Inc., Salt Lake City, UT), a DNA binding dye that fluoresces in duplex-dependent fashion. Primer sequences used to amplify each genomic region prior to melting are given in Table 14. Relative amplicon sizes and locations are shown in Figure 1. Using a high-resolution melting instrument (LightScanner® Instrument Model HR 96, Idaho Technology Inc., Salt Lake City, UT), PCR products were melted, and fluorescence loss was tracked as the dye was released during DNA denaturation. Melt range for gag amplicons was 68-98° C with a 65° C hold while the melt range for HRl, HR2 and the IDR amplicons was 60-98° C with a 57° C hold. Data plots of fluorescence (y-axis) vs. temperature (x-axis) were used to create melt curves, which were processed to calculate the HRM Score (the difference between the temperature when amplicon melting began and the temperature when amplicon melting was complete). Analysis of each DNA template was conducted in duplicate. Where a deviation of >0.5 was observed between duplicate analyses, data were rejected, and samples were reanalyzed.

This study included samples collected before seroconversion (acute, n=20), at or near the time of seroconversion (recent, n=102), and more than two years of infection where patients a CD4 cell count greater than 50 cells/ul (chronic, n=35), or less than 50 cells/ul (AIDS, n=33). General participant characteristics and the median (range) HRM scores for the different regions and infection stages are provided in Table 12. For the region of gag and two regions of env (HRl and HR2), HRM scores in adults with acute HIV infection were lower than in adults with recent HIV infection (p=0.0095, 0.0176, and 0.0007 for gag, HRl, and HR2 respectively), and the median HRM score in adults with recent HIV infection was lower than in adults with chronic HIV infection or AIDS (p<0.0001 for all three regions). Table 11. P values obtained by comparing HRM scores in the gag, HRl, and HR2 regions ts with different stages of HIV disease.

Table 12. Association of HRM and stage of HIV disease.

Abbreviations: NA: not available, HRM: high resolution melting.

a Samples were obtained from the EXPLORE study (1), the Johns Hopkins University Moore Clinic (2), and the Johns Hopkins Hospital Serosurvey (3).

b For adults with chronic HIV infection, age was only known for 27 individuals, CD4 cell count data was only available for 24 individuals, and HIV viral load data was only available for 22 individuals.

10 ^c Data for gender and race were not available for some individuals with chronic infection or

AIDS.

The data below (Table 13) show the ability of the HRM assay to discriminate between recent vs. non-recent HIV infections, by combining HRM scores from different regions and by

15 using a variety of assay cut-offs. Data are for the HIV gag regions and two regions in HIV gp41 (HR1 and HR2). These data show that the discrimination between recent vs. non-recent infection is improved when multiple regions are analyzed to generate a "diversity signature" of the sample. The relationship of results obtained for each of the three regions in adults with different stages of HIV disease is shown in Figure 6. Analysis of the HRM score from adults

20 with different stages of HIV disease shows that the HRM score increases with progressive stages of the disease (Figure 7). TABLE 13

Cutoff = Recent Mean+2SD

Cutoff = Acute (n=20) Mean +3SD

The work described in Examples 1-5 was performed using the following materials and methods. HRM assay - analysis of HIV gag and pol regions

PCR products generated in the ViroSeq system were used as template DNA. In each reaction, the region of interest was amplified in a nested PCR reaction that included a fluorescent dye that was incorporated into the amplified DNA duplexes. The primer sequences and regions amplified are shown in Table 2 and Figure 3. The Gagl amplicon includes a portion of the coding regions for gag p7 and gag pi. The Gag2 amplicon includes a portion of the coding regions for gag p7, gag pi, and gag p6. The Pol amplicon includes a portion of the coding regions for protease and reverse transcriptase. A high-resolution melting instrument

(LightScanner) was used to warm the samples over a melt range 68°C to 98°C, and to monitor the change in fluorescence that resulted as the DNA duplexes melted and the fluorescent dye was released. Those data were used to produce a melting curve for each sample that displayed the change in fluorescence as a function of temperature. Those curves were used to determine the temperatures over which melting occurred (defined as the HRM score, number of degrees of temperature between the left and right margins). Each DNA sample was analyzed in duplicate; results were deemed acceptable if the HRM scores from the duplicate assays differed by <0.5.

HRM assay - analysis of HIV env regions

Templates for HRM analysis of regions of env were prepared as follows. RNA extraction was carried out using the ViroSeq system. A region of env (gpl60) was reverse transcribed and amplified using the Qiagen OneStep RT-PCR Kit (QIAGEN Inc., Valencia, CA). RT-PCR reactions contained 10 μΐ of viral RNA extraction, lx Qiagen OneStep RT-PCR Buffer, 400 μΜ dNTP Mix, 2 μΐ Qiagen OneStep RT-PCR Enzyme Mix, 20 μΜ forward and reverse primers, and RNase-free water for a total reaction volume of 50 μΐ. Reverse transcription and

amplification were carried out on a 9700 Thermal Cycler (Applied Biosystems, Foster City, CA). Initially, the 10 μΐ RNA extract aliquots were subjected to a 5 minute hold at 65° C to denature RNA secondary structure. The RT-PCR mastermix, containing all other reaction components, was subsequently added to the RNA extract, and the completed reaction mix was returned to the thermocycler. The RT-PCR began with a 45 min. 50° C hold followed by a 15 min. 95° C hold. The method continued with 20 cycles of 94° C for 15 sec, 60° C for 30 sec, and 68° C for 60 sec followed by 30 cycles of 94° C for 15 sec, 60° C for 30 sec, and 68° C for 90 sec. Cycling concluded with a 7 min 68° C hold and an indefinite hold at 4° C. Primer sequences were as follows: Forward, JH35F (5'-TGARGGACAATTGGAGAARTGA-3 ' ) and Reverse, JH38R (5'- GGTGARTATCCCTKCCTAAC-3') (Mani I et al., 2002, / Virol, 76:10745-10755). Templates were diluted and purified according to the ViroSeq system using a gel-based dilution method and ExoSAP-IT™ (UBS, Cleveland, OH) purification. Purified templates were diluted 1:10 prior to HRM analysis.

PCR amplicons generated utilizing the methods described above were diluted and used as template DNA for HRM analysis. A region of each PCR product was amplified in the presence of a DNA binding dye that fluoresces in duplex-dependent fashion (LCGreen® Plus; Idaho Technology Inc., Salt Lake City, UT). Primer sequences used to amplify each genomic region prior to melting are given in Table 14. Table 14. Regions of the HIV genome and primer sequences for analysis of HIV gag and HIV

Control reagents: region of gag - 2 subtype A plasmids and 2 subtype D plasmids; regions of env - 13 subtype B plasmids.

b Referred to as Gag2 in table 3.

c Primer Sequences: Prl -Forward: 5'- ACTGAGAGACAGGCTAATTTTTTAG; PR2-Reverse: 5'- GGTCGTTGCCAAAGAGTGATTTG; PR3-Forward: 5'- CAGCAGGWAGCACKATGGG; PR4-Reverse: 5'- GC AR ATGWG YTTTCC AG AGC ADCC ; PR5-Forward: 5'- CT YC AGRC A AG ARTC YTGGC ; PR6-Reverse: 5'- TCCC A YTS C AKCC ARGTC ; PR7- Forward: 5'- TGCTCTGG A A ARCWC AT YTGC ; PR8-Reverse: 5'-

AARCCTCCTACTATCATTATRA.

Relative amplicon sizes and locations are shown in Figure 3B. Using a high-resolution melting instrument (LightScanner® Instrument Model HR 96, Idaho Technology Inc., Salt Lake City, UT), PCR products were melted, and fluorescence loss was tracked as the dye was released during DNA denaturation. Melt range for gag amplicons was 68-98° C with a 65° C hold while the melt range for HRl, HR2 and the IDR amplicons was 60-98° C with a 57° C hold. Data plots of fluorescence (y-axis) vs. temperature (x-axis) were used to create melt curves, which were processed to calculate the HRM Score (the difference between the temperature when amplicon melting began and the temperature when amplicon melting was complete). Analysis of each DNA template was conducted in duplicate. Where a deviation of >0.5 was observed between duplicate analyses, data were rejected, and samples were reanalyzed.

Implementation in Software and/or Hardware

As will be appreciated by one of ordinary skill in the art, the methods described herein can be implemented in hardware and/or software. For example, a general purpose computer can execute a software program that implements the methods described herein, thereby becoming a specially-programmed computer configured to implement the methods herein. Such a computer can include a processor for performing the methods described herein and communication interfaces for receiving data from an data source (e.g. , memory, an imaging device, another computer, and the like), receiving instructions (e.g., a keyboard, a mouse, a touch screen, and the like), and communication results (e.g. , a monitor, a printer, and the like).

Software implementing the methods herein can be stored in computer-readable media for execution on a computer. Such computer-readable media can be tangible or intangible and can be transitory or non-transitory.

An aspect of the invention which implements computer readable software is a computer program product containing computer-usable medium having control logic stored therein for causing a computer to perform a method of measuring genetic diversity, the control logic involving: (a) first computer readable medium code means for causing the computer to receive a melting curve for a polynucleotide corresponding to a region of the genome of an organism, (b) second computer readable medium code means for causing the computer to ascertain a left margin temperature and a right margin temperature for the melting curve; (c) third computer readable medium code means for causing the computer to calculate a high-resolution melting (HRM) score for the sample by subtracting the left margin temperature from the right margin temperature; and (d) fourth computer readable medium

Other Embodiments

From the foregoing description, it will be apparent that variations and modifications may be made to the invention described herein to adopt it to various usages and conditions. Such embodiments are also within the scope of the following claims.

The recitation of a listing of elements in any definition of a variable herein includes definitions of that variable as any single element or combination (or subcombination) of listed elements. The recitation of an embodiment herein includes that embodiment as any single embodiment or in combination with any other embodiments or portions thereof.

All patents and publications mentioned in this specification are herein incorporated by reference to the same extent as if each independent patent and publication was specifically and individually indicated to be incorporated by reference.

Claims

What is claimed is:

1. A method of measuring genetic diversity comprising:

(a) receiving a melting curve for a polynucleotide;

(b) ascertaining a left margin temperature and a right margin temperature for the melting curve;

(c) calculating a high-resolution melting (HRM) score for the sample by subtracting the left margin temperature from the right margin temperature; and

(d) correlating the HRM score with genetic diversity.

2. The method of claim 1, further comprising the step of generating a melting curve for a polynucleotide.

3. The method of claim 1, wherein the polynucleotide is an amplicon comprising a detectable moiety incorporated into the amplicon in a polymerase chain reaction.

4. The method of claim 1, wherein the polynucleotide is selected from the group consisting of dsDNA, dsRNA, and DNA/RNA hybrid.

5. The method of claim 1, wherein the melting curve is ascertained using a fluorescent assay.

6. A method for measuring genetic diversity, the method comprising

(a) amplifying a region of the genome of an organism in the presence of a detectable moiety, such that the amplicon comprises the detectable moiety;

(b) heating the polynucleotide comprising the detectable moiety;

(c) detecting an alteration in the signal generated by the detectable moiety in response to heating, wherein the detected alteration when plotted as a function of temperature defines a melting curve for said polynucleotide, and the difference in temperature between a left and a right margins of the melting curve is a HRM score; and

(d) correlating the HRM score with genetic diversity.

7. The method of claim 1, wherein the polynucleotide is isolated from an organism.

8. The method of claim 7, wherein the organism is selected from a human, an animal, a plant, a virus, a bacterium, a fungus, and a protozoa.

9. The method of claim 8, wherein the virus is human immunodeficiency virus (HIV).

10. The method of claim 4, wherein the detectable moiety is a fluorescent dye.

11. The method of claim 4, wherein the detected alteration is fluorescence which changes as a function of temperature.

12. The method of claim 4, wherein the amplicon is heated over a melt range of 68°C to 98°C.

13. The method of claim 1 or 4, wherein the melting curve displays a change in fluorescence as a function of temperature.

14. The method of claim 1 or 4, wherein the HRM score reflects the temperatures over which melting occurred.

15. The method of claim 4, wherein the amplicon comprises at least a portion of a viral gag coding region.

16. The method of claim 15, wherein the amplicon comprises at least a portion of HIV gag p7 and gag pi .

17. The method of claim 16, wherein the amplicon comprises at least a portion of the coding regions for gag p7, gag pi, and/or gag p6.

18. The method of claim 4, wherein the amplicon comprises at least a portion of a viral env coding region.

19. The method of claim 18, wherein the amplicon comprises at least a portion of a viral gp41 coding region.

20. The method of claim 19, wherein the gp41 amplicon comprises at least a portion of the coding regions for gp41 HR1, gp41 HR2, and/or gp41 IDR.

21. The method of claim 17, wherein the HRM score for a region in HIV gag varies with the stage of HIV disease.

22. The method of claim 20, wherein the HRM score for HIV gp41 HR1 or gp41 HR2 region varies with the stage of HIV disease.

23. The method of any one of claim 8-22, wherein an increased HRM score in an adult relative to a control is indicative of an increased severity of HIV.

24. The method of claim 23, wherein an increased HRM score is correlated with acute HIV, recent acquisition of HIV, chronic HIV, and AIDS.

25. The method of claim 4, wherein the amplicon comprises at least a portion of a viral pol coding region.

26. The method of claim 25, wherein the pol coding region comprises a portion of the coding regions for protease and reverse transcriptase.

27. The method of claim 1, wherein the polynucleotide is purified from a source selected from the group comprising viruses, bacteria, fungi, cancer cells, tissue, and bodily fluids.

28. The method of claim 1 or 4, wherein the range of melting temperatures of DNA duplexes provides a measure of HIV diversity.

29. The method of claim 4, wherein the amplicon is generated using primers designed to amplify HIV subtypes A, B, C, and D.

30. A method of determining the severity of a viral infection comprising, determining an HRM score for a virus, wherein the HRM score is determined according to the method of claim 1 or 4, and correlating the HRM score with the severity of the infection.

31. A method of determining the length of time a subject has had a viral infection comprising; determining the HRM score for a virus isolated from the subject, wherein the HRM score is determined according to the method of claim 1 or 4, and correlating the HRM with the length of infection.

32. The method of claim 31, wherein HRM score increases in acute HIV, recent acquisition of HIV, chronic HIV, and AIDS.

33. A method of discriminating between recent and non-recent HIV infections comprising; determining an HRM score for an HIV virus isolated from a subject, wherein the HRM score correlates with the length of infection.

34. A method of determining prognosis in a subject having a viral infection comprising; determining the HRM score for a virus isolated from the subject, wherein the HRM score is determined according to the method of claim 1 or 4, and correlating the HRM with prognosis.

35. A method of determining the efficacy of a treatment for a viral infection comprising; determining the HRM score for a virus isolated from the subject before and after treatment, wherein the HRM score is determined according to the method of claim 1 or 3, and correlating the HRM with the efficacy of treatment.

36. A method of identifying virus from a subject having a viral infection comprising;

determining a melting curve for a virus isolated from the subject, and correlating the melting curve with a particular viral fingerprint.

37. A method of determining the duration of viral infection in a subject comprising;

determining the HRM score for a virus isolated from the subject, wherein the HRM score is determined according to the method of claim 1 or 4, and correlating the HRM with the duration of infection.

38. A method of determining a cross-sectional incidence of a viral infection comprising; determining the HRM scores for virus from samples isolated from a population of subjects, wherein the HRM scores are determined according to the method of claim 1 or 4, and correlating the HRM with the incidence of infection.

39. The method of claim 1, wherein the melting curve is generated for a mixture of at least two polynucleotides.

40. The method of claim 1, wherein the HRM score of at least two polynucleotides are determined when the polynucleotides are analyzed as a mixture.

41. The method of claim 2, wherein the melting curve is generated by plotting the negative derivative of fluorescence/temperature [-d(fluorescence/dT)] against temperature.

42. The method of claim 1, wherein the left and right margin temperatures are determined by measuring where the slope of the melting curve achieves a 30 degree angle.