HIV circulating recombinant form (CRF) describes a recombinant lineage that plays an important role in the HIV pandemic. As we know, a designation of CRF requires three representative strains to be identified in at least three HIV-infected persons without direct epidemiological linkage, and three near full-length genomic (NFLG) sequences are preferred . However, two complete genomes in conjunction with partial sequences of a third strain are sufficient to designate a CRF (the partial sequences must also confirm the CRF's mosaic structure) .
Over time, this CRF nomenclature contributes to the investigation of both regional and global HIV pandemics. For example, CRF01_AE strains are predominant in Southeast Asia, CRF02_AG are found primarily in Africa, CRF07_BC and CRF08_BC are only found in China, and so on [2,3]. Thus far, there have been 58 HIV-1 CRFs and one HIV-2 CRF defined worldwide. We have extracted the information of all the HIV-1 CRFs from the Los Alamos HIV database (http://www.hiv.lanl.gov/content/index) and published articles are listed in Table 1. Within these CRFs, there are some problems. First, information about two CRFs (CRF50_A1D and CRF56) was absent. Second, data supporting the definition of three CRFs (CRF30_0206, CRF32_06A1 and CRF41_CD) was insufficient: for CRF30_0206, recombinant breakpoints for the three NFLGs were different; for CRF32_06A1, only two NFLGs were found; and for CRF41_CD, the three NFLGs were unavailable because the correspondence articles remained unpublished. More importantly, no additional sequences had been submitted to support these designations. Third, some CRFs had not been found in any other follow-up study after their first identification (Table 1). The majority of the recently reported CRFs have been isolated from Asia as a result of the extensive unique recombinant forms (URFs) present in this region [4–6]. These CRFs have only been recently identified. It is possible that given additional time, follow-up reports will emerge to support their designation. However, some CRFs are still not detected after they were identified more than 3 years (Table 1), an alternate explanation for the dearth of follow-up reports is that these CRFs have not caused a pandemic.
Taking five CRF_01Bs representing the recombinants of CRF01_AE and subtype B in Malaysia for example, new strains of CRF48_01B had not been detected in this country after it was identified in the year of 2010. However, other CRFs, such as CRF52_01B, CRF53_01B and CRF54_01B, had continued to be reported throughout 2012. Furthermore, the most recent CRF58_01B from Malaysia was defined in 2013. The latest report showed that there were massive expansions of CRF33_01B strains and demonstrated the emergence of three small unique recombinant clusters in Malaysia, and these unique recombinant clusters still belong to the recombinants of CRF01_AE and subtype B . Together, these phenomena raise a question: are these newly-defined CRFs and URFs still in rapid and continuous evolution? If these new recombinant strains remain in a transitional stage, their variations may not be stable, so they are hardly to form a stable spread with the same mosaic structure. Only when the structure of recombinant strains is stable will they have the opportunity to cause an epidemic as one truly CRF. With improvements in sequencing technology, NFLGs for HIV can now be obtained with relative ease. However, the definition of a new CRF requires only that three NFLGs without direct epidemiological linkage be identified. This low standard results in the overestimation of CRFs worldwide.
CRFs should be limited to recombinant strains, which really caused a new epidemic. It will be of no use if CRF just describes a small number of recombinant strains. As the current nomenclature stands, the definition of CRF is of limited use and creates confusion. To improve the CRF nomenclature and ensure that it plays a meaningful role in the description of transmission dynamics for HIV circulating recombinant strains, we suggest that newly-identified recombinant forms with three NFLGs first be termed recombinant form. Then, if they are confirmed to have caused an epidemic in the future, the recombinant form can be defined as a new CRF. In addition, perhaps existing CRFs may be downgraded to recombinant form if they are not detected after 5–10 years? We hope that specialists in HIV epidemiology will turn their attention to this issue in CRF nomenclature and draw up a new standard together.
All the authors contributed to the concept, design and writing of this article.
Conflicts of interest
There are no conflicts of interest.
1. Robertson DL, Anderson JP, Bradac JA, Carr JK, Foley B, Funkhouser RK, et al. HIV-1 nomenclature proposal
2. Lihana RW, Ssemwanga D, Abimiku A, Ndembi N. Update on HIV-1 diversity in Africa: a decade in review
. AIDS Rev
3. Lau KA, Wang B, Saksena NK. Emerging trends of HIV epidemiology in Asia
. AIDS Rev
4. Pang W, Zhang CY, Duo L, Zhou YH, Yao ZH, Liu FL, et al. Extensive and complex HIV-1 recombination between B ’, C and CRF01_AE among IDUs in south-east Asia
5. Chen M, Yang L, Ma YL, Su YZ, Yang CJ, Luo HB, et al. Emerging variability in HIV-1 genetics among recently infected individuals in Yunnan, China
. PLoS One
6. Lau KA, Wang B, Kamarulzaman A, Ngb KP, Saksena NK. Continuous crossover(s) events of HIV-1 CRF01_AE and B subtype strains in Malaysia: evidence of rapid and extensive HIV-1 evolution in the region
. Curr HIV Res
7. Chow WZ, Ong LY, Razak SH, Lee YM, Ng KT, Yong YK, et al. Molecular diversity of HIV-1 among people who inject drugs in Kuala Lumpur, Malaysia: massive expansion of circulating recombinant form (CRF) 33_01B and emergence of multiple unique recombinant clusters
. PLoS One