A key drawback for many of the methods used to investigate cytoplasmic HTLV-1 and HIV-1 Gag behavior is that the techniques must disrupt the cellular environment in order to isolate Gag molecules for investigation. This is problematic because the isolation of Gag molecules necessitates the removal of cellular host factors and machinery that might play a prominent role in Gag trafficking. For example, it is known the HIV-1 Gag interacts with an array of host factors on its pathway to assembly (reviewed in [14
]). For this reason, it is vital to develop minimally-invasive techniques which are capable of investigating HTLV-1 Gag cytoplasmic behavior. Model systems that have used codon-optimized gag genes and carboxy-terminal tags such as GFP, have provided new tractable approaches for studying Gag trafficking and homo-interactions as well as membrane interactions. The impact of such perturbations on Gag trafficking needs further investigation.
To date, the most promising advances for studying retroviral Gag trafficking and assembly in vivo
involve fluorescence-based techniques. These techniques typically rely on Gag constructs which are tagged with one of a number of fluorescent proteins. Fluorescence-based assays are minimally invasive, as labeled Gag proteins can be monitored in living cells which are incubated on the microscope stage. The lone perturbation to the Gag expression system is the introduction of the fluorescent tag. The introduction of a tag to the protein could have unforeseen consequences on the Gag trafficking pathway. While this is a legitimate concern, it has been shown previously that HIV-1 Gag tagged at the C-terminus with yellow fluorescent protein (EYFP) and untagged wild-type constructs are incorporated non-preferentially into virus-like particles (VLPs) [39
]. VLPs are non-infectious analogues of infectious virions which contain the structural elements of the virus, but lack viral components which are required for viral propagation. They have been shown to closely mimic the structure and morphology of infectious virions (reviewed by [40
]). In addition, it has been shown that HIV-1 Gag, C-terminally labeled with GFP, is incorporated into infectious virions when coexpressed with wild-type Gag [41
]. Finally, a Gag construct with GFP inserted between the matrix and capsid domains of HIV-1 Gag has been shown to produce infectious viral particles with similar single-round infectivity to that of wild-type particles [42
]. These observations support the hypothesis that the fluorescence tag may have minimal impact on Gag trafficking and assembly.
One example of a fluorescence-based technique relevant to HTLV-1 Gag behavior is imaging localization of Gag molecules in fixed cells. Studies have revealed important insights into the role of the late domain in viral assembly [44
]. Fluorescence localization also provided evidence that HTLV-1 Gag preferentially targets tetraspanin-enriched microdomains associated with the plasma membrane for assembly [30
], avoiding interactions with intracellular membranes [23
Another promising technique for the in vivo
study of retroviral Gag trafficking and assembly is fluorescence resonance energy transfer (FRET) (reviewed in [47
]) (). This technique relies on energy transfer between two different fluorophores (an energy donor and an energy acceptor) to probe intra- and intermolecular interactions. The efficiency of energy transfer depends on the inverse sixth power of the distance between the fluorophores, and is used to monitor interactions at the nanometer scale. FRET has been used to probe Gag-Gag interactions for HIV-1 [49
] and RSV [51
], as well as HIV-1 Gag-nucleic acid interactions [36
] and HTLV-1 NC-nucleic acid interactions [52
Figure 1. Fluorescence resonance energy transfer (FRET). FRET is used for monitoring intra- and inter-molecular interactions occurring on the nanometer scale. The transfer of excitation energy from a “donor” fluorophore (Gag-CFP) to an “acceptor’ (more ...)
The fluorescence techniques described so far provide a suitable technology platform for the investigation of retroviral Gag trafficking and assembly in living cells. In recent years, the development of a suite of biophysical fluorescence techniques has further augmented the analytical power of fluorescence-based investigations. In particular, total internal reflection fluorescence (TIRF) (reviewed by [53
]) and two-photon fluorescence fluctuation spectroscopy (FFS) (reviewed by [54
]) have been used to reveal intriguing insights into Gag behavior.
In TIRF microscopy, excitation light is transmitted to the cover slip-sample interface at an angle greater than the critical angle (). This results in the incident light being totally reflected at the interface. Despite the total reflection of the excitation light, an evanescent field is established that penetrates into the sample to a depth of about 100 nm. In retroviral Gag in vivo studies, this excitation penetration depth corresponds to the cell plasma membrane and the cytoplasm-membrane interface. The limited penetration of the excitation light only excites fluorophores at or proximal to the membrane, and therefore greatly increases sensitivity and signal to noise due to the elimination of cytoplasmic background fluorescence. Thus TIRF microscopy is ideal for studying the steps of Gag trafficking and assembly which occur at the plasma membrane.
Figure 2. Total Internal Reflection Fluorescence (TIRF) microscopy. In objective-based TIRF microscopy, the excitation laser is positioned to enter at the edge of the back aperture of a high-NA objective (NA ≥ 1.45). The beam emerges from the objective (more ...)
To date, TIRF studies have focused on HIV-1 Gag behavior. Recent studies monitored the dynamics of VLP assembly at the membrane, determining the timescale from the beginning of detectable assembly at the membrane (~10 Gag molecules) to eventual budding. Jouvenet et al.
determined that 5–6 minutes passed from the first detection of an assembling VLP to a steady-state assumed to be the assembly endpoint [56
]. In another study, Ivanchenko et al.
demonstrated that the steady-state observed in the previous study was likely the budding of the viral particle and is followed by VLP release [57
]. They found that the average timescale for HIV-1 VLP assembly, budding, and release was closer to 25 minutes. In a more recent study, the role that the viral genome plays in Gag assembly at the membrane was investigated [35
]. It was found that RNA is detectable at a membrane assembly site before Gag is present at detectable levels, supporting the idea that viral RNA plays a structural role in VLP assembly. Finally, a recent study examined the role of the cellular ESCRT machinery in HIV-1 budding and release (For the role of ESCRT complexes in HIV-1 budding and release, see [41
]). The study focused on the ESCRT-associated protein VPS4A, which is thought to play a role in membrane scission, and thus particle release. The study found that VPS4A proteins indeed arrive at sites of fully-assembled VLPs, promoting VLP membrane scission and release on a ~35 s timescale [58
Two-photon fluorescence fluctuation spectroscopy (FFS) is a quantitative method which is ideally suited to explore protein interactions in living cells, and has been used to study RSV [51
], HIV-1 [39
], and HTLV-1 [59
] Gag behavior (). In two-photon excitation, a pulsed infrared laser, approximately double the one-photon excitation wavelength of the fluorescent protein, is focused within the sample of interest. If the laser intensity is high enough at the focal position, the fluorophore can be excited by the near simultaneous absorption of two infrared photons, each supplying half of the total excitation energy [61
]. Two-photon excitation is restricted to the close proximity of the laser focus, because it is only in this region that the intensity is high enough to support efficient two-photon absorption. Knowledge of the optics allows the calculation of a focal volume ([62
]), which provides a convenient measure of the effective volume observed by two-photon excitation. This results in fluorophore excitation which is limited to a spatially defined volume on the order of less than one femtoliter (10−15
L, ~1/1000th of a cell’s volume). This spatially defined excitation method has distinct advantages over one-photon excitation: it minimizes photodamage to the small two-photon excitation volume and thus minimizes photobleaching of biological samples. In addition, it avoids potential contamination of single molecule fluorescence from Raman scatter of the solvent [61
Figure 3. Fluorescence Fluctuation Spectroscopy (FFS). FFS monitors the fluorescence of single molecules moving through a laser excitation region. In FFS using two-photon excitation (2PE) a pulsed infrared laser is used to create a small excitation region (~1/1000 (more ...)
FFS monitors the fluorescence fluctuations caused by single-labeled proteins that migrate in and out of the diffraction-limited two-photon observation volume. Analysis of the fluctuations provides information on the concentration, mobility, and brightness of fluorescent proteins [62
]. The most common application of FFS analysis is the investigation of protein diffusion and transport through investigation of fluorescence burst duration, or correlation analysis. FFS restricted to correlation analysis is also known as fluorescence correlation spectroscopy (FCS), and was introduced by Webb and coworkers [66
]. Determination of protein stoichiometry through brightness analysis, or investigation of fluorescence fluctuation amplitude, is another analysis technique used in FFS. This brightness-based method uses the photon counting histogram (PCH) to determine the number of molecules in the observation volume and their brightness from the shape of the photon count distribution [68
]. Brightness is a measure of the average fluorescence intensity of a single molecule passing through the observation volume, and is reported in counts per second per molecule (cpsm). Access to the brightness of fluorescent proteins provides a unique way to determine the stoichiometry of proteins in cells. A monomer of Gag, labeled with a fluorescent protein, has a certain characteristic brightness, based on the nature of the fluorophore. A dimer of labeled Gag molecules has two, identical fluorophores, and thus would be twice as bright, with higher-order complexes being higher integer multiples of monomer brightness. Brightness has been shown to be a robust technique for the determination of protein stoichiometry over a wide range of oligomerization levels [39
The single molecule sensitivity and high spatial-temporal resolution of FFS allows for the comprehensive study of the Gag trafficking and assembly pathway. This is evident in recent FFS studies used to characterize RSV, HIV-1, and HTLV-1 Gag behavior.