Easlick, Juliet BS*; Szubin, Richard BS†; Lantz, Samantha BS‡; Baumgarth, Nicole DVM, PhD§; Abel, Kristina PhD‖
Type I interferons (IFN) have been known for more than 50 years when it was first discovered that virally infected cells secrete a substance that prevents the infection of other cells with virus.1 We now know that this “substance,” called interferon, in fact is comprised of a mixture of multiple subtypes that include alpha, beta, omega, kappa, epsilon, tau, and zeta interferon. In humans, IFN-α represents a family of 21 IFN-α subtypes. Although some IFN-α genes are pseudogenes, the majority of them are expressed at the protein level, and all bind to the same receptor.
The antiviral activity of IFN-α and IFN-β has been extensively studied, but many open questions remain. Why do multiple IFN-α subtypes exist? What regulates the biologic activity of all the diverse Type I IFNs when they all bind and signal through the same receptor? Different pathogens can induce distinct IFN-α subtypes in vitro,2-5 and IFN-α subtypes can differ in their antiviral activity. However, only few in vivo studies have assessed the functional activity of individual IFN-α subtypes and their roles during infection. Mice expressing an IFN-α1 transgene had significantly lower virus titers after murine cytomegalovirus infection than mice transfected with the IFN-α4 or IFN-α9 gene.6. In another murine study, acute cytomegalovirus-induced myocarditis was most efficiently reduced with IFN-α9 or IFN-β.7 Murine IFN-α1 was also highly effective in reducing herpes simplex virus-induced ocular infection.8 In contrast, IFN-α1 exerted less antiviral activity in a murine model of influenza infection when compared with IFN-α5 and IFN-α6.9 Recently, it was demonstrated that IFN-α1, 4, and 9, but not IFN-α6, were able to reduce virus replication in the murine model of Friends retrovirus infection.10 How these results translate to human diseases is unknown.
Current IFN-α-based clinical treatments like pegylated interferon (eg, Pegasys; Genetech, San Francisco, CA) and RoferonA (Hoffmann-LaRoche, Inc., Nutley, NJ), often used to combat hepatitis C infection, are based on the application of a single IFN-α subtype, IFN-α2. An exception is the use of Multiferon (Viragen, Inc., Plantation, FL). This is despite the fact that in human patients infected with hepatitis C, IFN-α5 appears to be the most prominent IFN-α subtype expressed in the liver.4 Based on the few but striking results demonstrating differences in antiviral activity by various IFN-α subtypes, and considering the central role of Type I IFNs in the early immune defense and in directing T and B cell differentiation, a more comprehensive understanding of IFN-α subtypes in different disease settings is needed to determine whether distinct IFN-α subtype treatments might be beneficial in targeting distinct types of infections.
We previously showed that IFN-α mRNA is rapidly induced after oral SIV infection in infant rhesus macaques.11 Despite similar high levels of virus replication in lymphoid and mucosal tissues, we observed significantly higher increases in IFN-α mRNA levels in lymphoid compared with mucosal tissues.11 Because the early antiviral response at entry sites of the virus is critically important in the early control of virus replication, a thorough analysis of the various IFN-α subtypes and their expression patterns in mucosal compared with lymphoid tissues is warranted.
Therefore, we conducted the study presented here, in which we developed the tools to distinguish and measure multiple IFN-α subtypes at the mRNA level in rhesus macaques, an animal species used as a model system for many human diseases. We then determined which IFN-α subtypes were induced in response to pathogenic SIV infection and/or whether distinct tissues show unique IFN-α expression patterns.
The results show that after SIV infection of infant rhesus macaques, several IFN-α subtypes are rapidly induced in lymphoid but not at oral and gastrointestinal mucosal surfaces. Although each IFN-α subtype was induced at distinct levels, their relative expression patterns were identical in all lymphoid tissues examined.
MATERIALS AND METHODS
Animals and SIV Infection
Newborn rhesus macaques (Macaca mulatta) were housed and hand-reared in a primate nursery in accordance with the regulations of the American Association for Accreditation of Laboratory Animal Care Standards at the California National Primate Research Center (CNPRC). Animal procedures were approved by the University of California-Davis Institutional Animal Use and Care Committee. All animals were born to rhesus macaques of the CNPRC colony and were negative for HIV-2, SIV, Type D retrovirus, and simian T-cell lymphotropic virus type 1. Six infant macaques were exposed to SIVmac251 (obtained from the CNPRC Analytical Core) using a previously described repeated oral exposure model.11,12 The infected animals were euthanized 3 days after the last SIVmac251 exposure, ie, Day 8 after the first virus exposure. The time of euthanasia is subsequently referred to as 1 week postinfection. Six age-matched SIV-naive macaques served as controls.
Tissue Collection and Cell Isolation
At the time of euthanasia, blood, gingiva, tonsil, retropharyngeal and submandibular lymph nodes (LNs), jejunum, colon, mesenteric LN, and axillary LN samples were collected. Tissue samples were stored in RNAlater (Ambion, Austin, TX) at -20°C until RNA preparation. In addition, cell suspensions for functional assays were prepared as described.11 Cell isolation from intestinal tissues was performed according to previously described methods.13 Briefly, approximately 2-inch pieces of the ileum and colon were rinsed with phosphate-buffered saline (Invitrogen, Grand Island, NY) and then minced using sterile scalpels. The tissue suspensions were placed in a shaking water bath in RPMI 1640 (Invitrogen) containing 7.5% fetal bovine serum and collagenase Type II (0.5 mg/mL collagenase Type II; Sigma, St. Louis, MO) for 30 minutes at 37°C. After the digestion, the single-cell suspension was passed through a 100-μm filter, spun down, and resuspended in 10% fetal bovine serum in RPMI 1640. The remaining undigested tissue was resuspended in collagenase media and the digestion step was repeated a total of three times. Mucosal lymphocytes were then isolated from the obtained single-cell suspension by performing a 35%/60% Percoll (Sigma) gradient centrifugation. Intestinal lymphocytes were collected from the 35%/60% interface and washed twice with phosphate-buffered saline before being resuspended in 10% fetal bovine serum in RPMI 1640.
RNA ISOLATION AND cDNA PREPARATION
Before total RNA isolation, using Trizol (Invitrogen) according to the manufacturer's protocol, the tissue samples were homogenized using a Power Homogenizer (PowerGen 7 mm × 195 mm; Fisher Scientific, Pittsburgh, PA). RNA was used to determine tissue viral RNA levels and for gene expression analysis. RNA samples were DNase-treated with DNA-free (Ambion) for 1 hour at 37°C. Complementary DNA was prepared using random hexamer primers (Amersham-Pharmacia Biotech, Inc, Piscataway, NJ) and M-MLV-Reverse Transcriptase (Invitrogen). As a result of the low expression levels of some IFN-α subtype mRNA levels in tissues of SIV-naïve control animals, complementary DNA preamplification was performed using the ABI TaqMan Pre-Amp Master Mix Kit (Applied Biosystems, Foster City, CA) according to the manufacturer's instructions.
Interferon-α Subtype Cloning and Primer/Probe Design
At the initiation of the study, the rhesus macaque genome was not fully sequenced. However, we recently developed a nested polymerase chain reaction (PCR) strategy to clone the human IFN-α subtypes.14 Based on the strategy applied to human IFN-α subtype cloning and the available sequence information listed on the Baylor College of Medicine Rhesus Macaque Sequencing Project web site, rhesus IFN-α subtype primers located in the 5′ and 3′ UTR regions were designed (see “Results”). The PCR was carried out using the Stratagene Easy A Hi Fidelity PCR cloning enzyme (Stratagene, La Jolla, CA) under the following conditions: 1 minute at 94°C, 30 cycles at 94°C for 15 seconds, 55°C for 15 seconds, 70°C for 45 seconds, followed by 70°C for 5 minutes. The obtained PCR products were cloned using Invitrogen's TOPO TA cloning kit (TOPO TA Cloning Kit with One Shot® MAX Efficiency DH5α-T1R Escherichia coli). For each IFN-α subtype, a minimum of three clones were sequenced and compared for sequence homology using BioEdit Software.15 Once the sequence was confirmed to be IFN-α subtype-specific, inner primer/probe sets were designed using the previous described criteria for the amplification of human IFN-α subtypes.14 The specificity of IFN-α subtype-specific primer-probe sets was confirmed by using 0.01 ng DNA of each cloned IFN-α subtype (plasmids) for amplification with matched and all nonmatched primer/probe sets.
Real-Time (RT) PCR was performed as previously described.11,16 Briefly, samples were tested in duplicate, and the RT-PCR for the housekeeping gene GAPDH and the target gene from each sample were run in parallel on the same 96-well Optical Plate (Applied Biosystems, Foster City, CA) in a 25-μL reaction volume containing 5 μL cDNA + 20 μL Mastermix (Applied Biosystems). All sequences were amplified using the 7900 default amplification program: 2 minutes at 50°C, 10 minutes at 95°C, followed by 45 cycles of 15 seconds at 95°C and 1 minute at 60°C. Results were analyzed with the SDS 7900 system software, Version 2.1 (Applied Biosystems). In this analysis, the Ct value for the housekeeping gene (GAPDH) was subtracted from the Ct value of the target gene (delta, ΔCt) to normalize for mRNA input. The average ΔCt value for the tissue sample from all uninfected animals was then subtracted from the delta Ct value of the corresponding tissue sample from each of the SIV infected monkey (ΔΔCt). Because the target gene (cytokine) and the reference gene (GAPDH) are amplified with the same efficiency, and GAPDH values did not differ depending on the infection status of animals (data not shown), the increase in cytokine mRNA levels in tissue samples of SIV-infected monkeys compared with tissue samples of uninfected animals could then be calculated as: increase = 2−ΔΔCt (User Bulletin #2, ABI Prism 7700 Sequence Detection System; Applied Biosystems).
Immunochistochemistry for Interferon-α and MxA Protein Detection
Tissue section slides were stained using the DAKO Autostainer (Dako North America, Inc., Carpinteria, CA) as follows: formalin-fixed, paraffin-embedded tissue sections were cleared and rehydrated with xylene (3×) followed by a graded series of ethanol incubations (100%, 95%, 80%, and 50%), brief rinsing with water, and then with Tris-buffered saline containing Tween 20 (TBST; Dako). For IFN-α detection, slides underwent antigen retrieval with AR10 (Biogenex, San Ramon, CA; diluted 1:10) in a pressure cooker followed by a rinse with water and TBST. Slides for both IFN-α and MxA detection were blocked with Protein Block (Dako) for 10 minutes and then incubated in a humidified chamber with rabbit antihuman IFN-α antibody (US Biological, Swampscott, MA) or mouse antihuman MxA antibody (kindly provided by Dr O. Haller, University of Freiburg, Department of Virology, Germany) for 1 hour. All incubations were done at RT. Slides were washed with TBST (2×), incubated for 20 minutes with Ready-to-Use Peroxidase Quenching solution (DAKO), and rinsed with TBST (2×). Next, slides for IFN-α detection were treated with Advance Link (Dako), and slides for MxA staining with anti-mouse horseradish peroxidase polymer (Dako) for 15 and 30 minutes, respectively, followed by another wash with TBST (2×). An additional incubation (15 minutes) with Advance Enzyme (Dako) was performed for IFN-α detection. Then Substrate-DAB+ (Dako) was added for 10 minutes, slides were washed with running water for 5 minutes, and counterstained with hematoxylin (Dako) for 5 minutes. Finally, slides were dehydrated with 90% EtOH (three × 1 minute) and 100% xylene (three × 2 minutes).
Tissue sections from age-matched SIV-naïve animals were included as negative controls and tissue sections from SIV-infected animals that had been previously confirmed to be IFN-α- and/or MxA-positive served as positive controls.
Slides were examined with a Zeiss Axioskop microscope using Axiovision Software (Carl Zeiss MicroImaging, Inc, Thornwood, NY). Images were acquired at 200× magnification. Relative quantitation of IFN-α or MxA-positive cells per tissue was performed using two different methods. First, tissue sections were analyzed using Image ProPlus software (Media Cybernetics, Bethesda, MD). Briefly, for each tissue, three to five images per slides were acquired and analyzed. IFN-α or MxA-positive cells showed a dark brown color compared with IFN-α or MxA-negative cells only showing the blue/purple hematoxylin staining. Then, the T cell area within a lymph node section was outlined, and based on the assigned color parameters for positive and negative staining, the software generated area values for positive and negative staining within the determined T cell area. Thus, a relative assessment of positive and negative staining patterns was obtained, but an exact quantitation of positive cells/tissue mm2 was not performed. To confirm the results, IFN-α or MxA-positive cells were manually counted in each of the T cell areas examined (three to five fields per tissue). Depending on the number of positive cells present per field, a score was assigned. A score of “1” was assigned when a particular field contained less than 50 positive IFN-α-positive cells, a score of “2” for greater than 50 or less than 75, and a score of “3” for fields with greater than 75 IFN-α-positive cells. Microscopic fields with MxA-positive cells were assigned a score of “1” for greater than 3 or less than 10 positive cells/field, a score of “2” for greater than 10 or less than 25, and a score of “3” if greater than 25 MxA-positive cells were counted. Fields negative for IFN-α or MxA-positive cells were assigned a score of “0.” Both methods resulted in a similar ranking of tissues with respect to the number of IFN-α and/or MxA-positive cells, and therefore only the data from the latter scoring analysis are reported.
Analysis of Plasmacytoid Dendritic Cell Frequencies and Function
The frequencies and ability of plasmacytoid dendritic cell (pDC) to produce IFN-α were determined essentially as described previously.17 Briefly, peripheral blood mononuclear cells (PBMCs) or tissue cell suspensions were prepared and stimulated with herpes simplex virus-2 for 6 hours at 37°C, 5%CO2. Brefeldin A (10 μg/mL, Sigma) was added for the last 5 hours of incubation. After in vitro stimulation, pDC were analyzed by flow cytometry. pDCs were identified as lineage (CD3, CD20, CD14, CD16)-negative, HLA-DR, and CD123-positive. Generally, 300,000 events were acquired on a FACS ARIA (Becton-Dickinson, San Jose, CA). Data were analyzed with FlowJo (TreeStar, Ashland, OR). Frequencies of pDC were calculated as percent of mononuclear cells, and frequencies of IFN-α and tumor necrosis factor-α−producing cells as percentage of pDC.
In Vitro Peripheral Blood Mononuclear Cell Stimulation to Determine Interferon-α Subtype mRNA Expression
Rhesus PBMC were purified and stimulated at 1 × 106 cells/mL in RPMI1640 (supplemented with 10% fetal bovine serum and penicillin/streptomycin) with herpes simplex virus-2 (multiplicity of infection = 1), 50 μg of CpG ODN 2336 (Invitrogen, San Diego, CA), or 10 μg ssRNA40/LyoVec (InVivoGen) for 5 hours. RNA was purified as described11,16,18 to determine mRNA levels of the various IFN-α subtypes by RT- PCR.
Virus Load Measurement
Plasma and tissue RNA samples were analyzed for viral RNA as described previously.11
Results were analyzed using one-way analysis of variance or the Student t test using GraphPad Prism and InStat software programs, Version 4 (Graph Pad Software, Inc, San Diego, CA). Correlations between two parameters were determined using regression analysis tools by GraphPad Prism software.
Design of Rhesus Macaque-Specific Interferon-α Subtype Primer-Probe Sets
Our initial studies were based primarily on the available human IFN-α subtype sequence information. Assuming that sequences would be highly homologous between humans and rhesus macaques, we searched the Rhesus Macaque Sequencing Project web site at the Baylor College of Medicine for sequences with homology to human IFN-α subtypes. Based on these results, rhesus macaque-specific IFN-α subtype sequences were amplified from rhesus genomic DNA using primer sequences located in the 5′ and 3′ UTR regions (Table 114). The obtained PCR products were cloned, and for each expected rhesus IFN-α subtype, three clones were sequenced, aligned to each other, and compared with the relevant human IFN-α subtype sequence. Using this strategy, we generated rhesus macaque-specific clones for eight of the 13 known human IFN-α subtypes (IFN-α1/13, IFN-α2, IFN-α4, IFN-α6, IFN-α8, IFN-α14, and IFN-α21). Rhesus-specific clones with sequence homology to human IFN-α 5, 7, 10, and 16 could not be generated with this initial approach. The clone obtained for rhesus IFN-α17 showed a similar degree of nucleotide homology to four human IFN-α subtype genes (IFN-α4, 7, 10, and 17) and was therefore not considered to represent a specific rhesus IFN-α subtype. All of the obtained rhesus clones span the full coding sequence of the relative IFN-α subtype gene they encode when compared with the rhesus IFN-α subtype gene listed in GenBank (Table 2), except for the IFN-α8 clone that misses the last 25 base pairs. The sequence homology of each IFN-α subtype clone (beginning from the start codon) to the coding sequence of the rhesus IFN-α subtype gene in GenBank ranges from 98% to 100%. Note that, like in humans, the rhesus macaque sequences for IFN-α1 and IFN-α13 are almost identical and therefore were amplified by the same primer sets and are represented by a single clone.
Coincidentally, while we were pursuing the human14 and rhesus studies, the rhesus macaque genome was published.19 A search of all human IFN-α subtype sequences against the now annotated rhesus macaque genome database revealed the predicted rhesus IFN-α17 sequence. This rhesus IFN-α17 sequence was used in the subsequent real-time PCR assay design. Thus, although there are 13 expressed human IFN-α subtypes, we could identify and design rhesus-specific molecular assays to distinguish among nine IFN-a subtypes (IFN-α1/13, IFN-α2, IFN-α4, IFN-α6, IFN-α8, IFN-α14, IFN-α17, and IFN-α21).
Consistent with our failure to obtain rhesus IFN-α7, IFN-α10, and IFN-α16 clones or sequences, the human IFN-α7, IFN-α10, and IFN-α16 sequences were most similar to the predicted sequence for rhesus IFN-α4. In addition, a rhesus counterpart for the human IFN-α5 gene was not found. Instead, the human IFN-α5 sequence appeared to be most closely related to the predicted sequence for rhesus IFN-α6.
Table 2 presents the summary of the Genbank annotation numbers for the human IFN-α subtypes, their rhesus counterparts, and their nucleotide sequence homology. Sequence differences for human and rhesus macaque IFN-α subtypes extended to the amino acid level (Fig. 1), thus demonstrating the divergent evolution of the rhesus macaque and human IFN-α subtype genes.20
Next, primer-probe sets for single-round RT PCR amplification for all confirmed rhesus IFN-α subtype sequences were designed. Toward this goal, previously developed RT PCR primer/probe sets for a subset of human IFN-α subtypes (IFN-α1/13, IFN-α2, IFN-α6, and IFN-α8)5 were compared with rhesus IFN-α subtype sequences and mismatched nucleotides between the human and rhesus sequence were optimized for the rhesus specific sequence (Fig. 2; Table 1). All other primer-probe sets were designed using the ABI Primer Design Software. Despite the extensive nucleotide homology among all rhesus IFN-α subtypes (Fig. 2), each subtype-specific primer-probe set preferentially amplified the unique IFN-α subtype it was matched to (Fig. 3A). After confirming the specificity of the designed rhesus-specific IFN-α subtype primer-probe sets, PCR amplification of all cloned rhesus IFN-α subtypes was performed over a range from 10 to 1 × 106 copies plasmid DNA to ensure that all IFN-α subtype sequences were amplified with comparable efficiency. The similar slopes and the partial overlap of the amplification curves of the various IFN-α subtypes (Fig. 3B) suggest similar efficiency in the PCR amplification.
Interferon-α mRNA Subtype Measurement After In Vitro Stimulation
To determine the extent to which the various identified rhesus IFN-α subtypes can be measured with the developed assay, rhesus PBMCs were stimulated for 6 hours with herpes simplex virus, a virus that induces strong IFN-α production in human and rhesus PBMC,17,21 the TLR9 ligand CpG ODN 2236, or the TLR7/8 ligand ssRNA40/LyoVec. The results demonstrate that multiple IFN-α subtypes can be induced in vitro in rhesus macaque PBMC (Fig. 3C). The magnitude of individual IFN-α subtype mRNA levels was dependent on the stimulus type. Herpes simplex virus induced at least 10-fold higher IFN-α subtype mRNA levels compared with CpG ODN and ssRNA. Generally, IFN-α14, IFN-α17, and IFN-α21 were less induced than IFN-α1/13, IFN-α2, IFN-α4, IFN-α6, and IFN-α8. Consistent with earlier studies in humans,14 whereas the magnitude of induction of the various IFN-α subtypes differed dependent on the stimulus used, there was no apparent preferential induction of any IFN-α subtype by any one treatment.
Taken together, these results demonstrated that this highly sensitive and specific assay could be used to measure the induction of individual IFN-α subtypes in rhesus macaques during the course of in vivo infection with SIV or other pathogens.
Interferon-α Subtype Expression Patterns in Lymphoid Tissues After Oral SIVmac251 Infection in Infant Macaques
To determine the levels and IFN-α subtype responses to SIV infection in infants, we used available tissue samples from six infant macaques that were infected orally at 4 weeks of age by repeated SIVmac251 inoculation. Of those six macaques, all but one animal tested positive for plasma viral RNA 1 week after the first SIV exposure.11 The one that tested negative (No. 36444) had detectable viral RNA in the submandibular lymph node and the colon (76 and 139 SIV RNA copies per μg tissue RNA, respectively), indicating infection of all six SIV-exposed macaques.
SIV infection resulted in the increase of multiple IFN-α subtypes in various tissues of all animals (Fig. 4 and data not shown). Similar to the in vitro studies (Fig. 3C), the magnitude of mRNA induction differed between the individual IFN-α subtypes with the highest increases seen for IFN-α1/13, IFN-α2, IFN-α6, and IFN-α8. The relative expression levels between the various IFN-α subtypes in the different tissues of an individual animal, however, were remarkably similar. Thus, IFN-α subtypes did not show any tissue-specific expression patterns in orally SIV-infected infant macaques.
Generally, the overall magnitude of IFN-α subtype induction seemed to be correlated with the level of virus replication in individual animals. The monkey with the highest plasma SIV RNA levels (36438) also showed the highest increases in IFN-−α mRNA levels for all the IFN-a subtypes tested (Fig. 4A and data not shown). Furthermore, in the tonsil, but not in the axillary or mesenteric LN, the mRNA expression levels of the most strongly upregulated IFN-α subtypes, IFN-α 1/13, IFN-α2, IFN-α6, and IFN-α8, positively correlated with the levels of virus replication in the same tissue (Fig. 4B). Consistent with the pattern of virus dissemination after oral SIV exposure,11 IFN-α subtype mRNA expression levels were generally highest in the sites closest to the exposure site, ie, the tonsil, and lowest in the distant mesenteric LN. In fact, IFN-α mRNA levels of animal 36438 were significantly higher (P < 0.05) in the tonsil compared with the retropharyngeal, the axillary and the mesenteric LN, and the retropharyngeal LN had higher gene expression of all IFN-α subtypes than the mesenteric LN (P < 0.05).
Interferon-α Protein Production in Lymphoid Tissues After Oral SIV Infection
To assess whether IFN-α subtype mRNA expression increases correlated with enhanced IFN-α protein production, we performed immunohistochemistry in various LNs. The majority of IFN-α-positive cells were localized in the T cell areas of the LN (Fig. 5A). Relative frequencies of IFN-α-positive cells per T cell area for each LN correlated with the IFN-α subtype mRNA expression data. The highest frequencies of IFN-α-positive cells were found predominantly in LNs draining the oral cavity (retropharyngeal LN) and the tonsil, but less so in the more distal mesenteric LN 1 week after oral SIV exposure (Fig. 5A-B). The colocalization of an interferon-inducible protein, MxA, with IFN-α in the same tissues was evidence that the IFN-α was biologically active (Fig. 5A-B). Consistent with low frequencies of IFN-α-positive cells in mesenteric LNs, MxA-positive cells were not or only at low frequencies detectable in the mesenteric lymph nodes (Fig. 5). Thus, at the earliest time points post-SIV infection, the expression of MxA is likely related to the total frequencies of IFN-α-secreting cells and to the time IFN-α is present in the tissue to result in the induction of IFN-α-inducible genes like MxA.
Plasmacytoid Dendritic Cells as a Source of Interferon-α
pDCs have been identified as the main source of IFN-α during acute viral infections. To determine whether the increases in IFN-α mRNA and protein were the result of increases in pDC frequencies and/or their activation state, we compared frequencies in peripheral blood and lymphoid tissues of 4-week-old SIV-naïve with SIV-infected infant macaques at 1 week after SIV exposure using flow cytometry (Fig. 6). The submandibular LN is the main draining LN in the oral cavity,22 and indeed, pDC frequencies were significantly increased in the submandibular and cervical LN of SIV-infected compared with SIV-naïve animals. For the retropharyngeal LN, a similar trend was observed but did not reach statistical significance, likely as a result of the relatively high animal-to-animal variation in pDC frequencies. The low cell yields from infant tonsils did not allow the pDC analysis for tonsils. Importantly, however, the tissues with the highest numbers of pDC at 1 week after oral SIV infection, eg, the submandibular and retropharyngeal LN, also showed the highest increases in IFN-α mRNA levels (Fig. 4).
Thus, oral SIV exposure in infant macaques results in 1) the rapid and simultaneous induction of multiple IFN-α subtypes in various tissues; 2) the appearance of IFN-α-producing cells in lymph nodes; and 3) increased pDC frequencies in various LNs draining the oral cavity. The magnitude of the Type I IFN response was highest in the tonsil and in LNs draining the oral entry site of SIV, thus following the pattern of virus dissemination.
Mucosal Interferon-α Responses
In contrast to the strong induction in lymphoid tissues, individual IFN-α subtype mRNA levels increased only slightly (two- to maximal six-fold) in the gingiva and the colon (Fig. 7A). This is consistent with our previous data generated with a primer/probe set that preferentially amplified IFN-α2.11,18 This was not the result of the induction of Type I IFN genes other than IFN-α. In fact, elevated IFN-β mRNA levels were only observed in the tonsil and only in the two animals (36438, 36414) with the highest virus replication overall (data not shown). Similarly, IFN-κ mRNA levels in the gingiva of SIV-infected animals were not increased compared with SIV-naïve animals (data not shown) despite the fact that IFN-kappa is preferentially produced by epidermal keratinocytes and can be induced in the gingiva.23 Despite our failure to detect a strong induction of any Type I IFN in the mucosa, IRF-7 mRNA levels were strongly expressed in the gingiva and the colon compared with the controls and positively correlated with the level of SIV replication in the same tissue (Fig. 7B).
To determine whether the differences in IFN-α induction between LNs and mucosal sites were the result of quantitative or qualitative differences in pDCs, we compared IFN-α responses by pDC in cell suspensions from lymphoid and mucosal tissues of 6-month-old macaques after in vitro stimulation with Imiquimod, a TLR7/8 agonist. Consistent with the ex vivo IFN-α expression analysis, few pDCs from the intestinal tissues produced IFN-α (Fig. 7C). Tumor necrosis factor-α production by mucosal pDC was also reduced, albeit to a lesser extent.
Thus, despite the increase of IRF-7 mRNA levels in the gingiva and the colon after SIV infection, and in contrast to the strong induction of IFN-α subtypes in lymphoid tissues, mucosal tissues are relatively devoid of IFN-α-producing cells during acute SIV infection. The in vitro TLR7/8 stimulation studies indicated a lower and/or altered responsiveness of pDC from intestinal tissues compared to LN pDC within an individual animal.
Innate responses are critically important in the early control of pathogen replication and thereby prevention of dissemination from the local entry to distal sites. Innate responses also determine the nature of the subsequent adaptive response and are involved in the maintenance of the pathogen-specific response. Therefore, a more detailed knowledge of the breadth of innate responses will be beneficial in our overall understanding of pathogenesis and in the design of more efficacious intervention strategies.
Type I IFNs have long been known as a first line defense mechanism in viral infections. In fact, IFN-α is used clinically in the treatment of patients infected with hepatitis or HIV. However, although IFN-α treatment is successful in some patients, it shows no effect in others. Multiple factors could contribute to treatment failure.24,25 Knowing that IFN-α comprises a family of multiple subtypes, it is imperative that we gain insight into the regulation and function of these various IFN-α subtypes in human diseases.
The current study developed IFN-α subtype-specific primer/probe sets as a means to dissect the IFN-α response in rhesus macaques (Table 1; Figs. 2 and 3). As expected, the rhesus IFN-α subtype clones revealed a high degree of homology to the human IFN-α subtype genes (Table 2). Interestingly, however, nucleotide alignments between the human and macaque sequences demonstrated that not all human IFN-α subtypes necessarily correspond to the same numeric IFN-α subtype in rhesus macaques and vice versa. The comparative analysis demonstrated that the human sequences for IFN-α7, IFN-α10, and IFN-α16 had no apparent rhesus gene homolog but showed a high degree of nucleotide homology to rhesus IFN-α4. Similarly, we were unable to clone rhesus IFN-α5 and the alignment of the human IFN-α5 sequence against predicted rhesus IFN-α subtype sequences identified rhesus IFN-α6 as having the highest homology to human IFN-α5. In this context, it is noteworthy that human IFN-α6 has been proposed to be a pseudogene.26,27 Although we recently showed that IFN-α6 could be induced at low levels after in vitro TLR stimulation in human pDC,14 other human IFN-α subtypes, including IFN-α5, were induced at much higher levels. In contrast, in rhesus macaques, IFN-α6, but not IFN-α5, was strongly induced after in vitro TLR stimulation and in vivo SIV infection. Although one could therefore speculate that the functional equivalent of human IFN-α5 is rhesus IFN-α6, studies analyzing the genetic relatedness or divergence of human and rhesus IFN-α subtype sequences were beyond the scope of this study.
Genetic studies by Woelk et al have suggested that IFN-α subtype genes in humans and chimpanzees are closely related.20 In fact, an alignment of all genic and intergenic regions of each of the IFN-α genes in humans shows the exact same location of the family locus compared with the chimpanzee IFN-α genes.20 In contrast, only IFN-α1, IFN-α2, IFN-α8, IFN-α13, and IFN-α6 are preserved in the locus between humans and rhesus macaques. Thus, the authors proposed that the most recent common ancestor contained only a subset of the IFN-α subtype genes, but then the human/chimpanzee and the rhesus IFN-α subtype genes diverged separately.20 Both gene duplication and gene conversion contributed to the diversity of the IFN-α subytpe genes observed in the different species.20
Because rhesus macaques are used as model systems for multiple human diseases, it is important to identify the similarities and differences in immune responses between rhesus monkeys and humans. In HIV-1-infected patients, increased serum levels of IFN-α have been associated with disease progression.28,29 Recently, it was demonstrated that, dependent on the stage of infection, the expression of IFN-α subtypes in PBMC of HIV-1 infected patients varies,30 a finding that has potential implications for improved diagnostics and treatment. The current study in the rhesus macaque model of SIV infection focuses on the early IFN-α subtype responses as they relate to virus dissemination and anatomic compartment.
The data demonstrate that multiple IFN-α subtypes are induced in acutely SIV-infected infant macaques (Fig. 4). Consistent with IFN-α1 as the first IFN-α subtype induced after stimulation and necessary for the induction of all other IFN-α subtypes in humans, we observed a strong increase of IFN-α1/13 mRNA levels in lymphoid tissue of rhesus macaques 1 week after SIV infection. A significant increase of IFN-α mRNA levels was also observed for IFN-α2, IFN-α4, IFN-α6, and IFN-α8. Why gene expression levels were less elevated for IFN-α14, IFN-α17, and IFN-α21 could not be determined. A recent study showed that the induction of distinct IFN-α subtypes is regulated by the differential recruitment of the transcription factors IRF-3 and IRF-7 to the promoter regions of the individual IFN-α subtype genes.31 Thus, in future studies, it would be interesting to determine how IRF-3 and IRF-7 expression and activation levels are altered by SIV infection. Importantly, if it could be demonstrated that the various IFN-α subtypes differ in their anti-SIV activity, the understanding of regulation of these subtypes would be critical. Similarly, it would be important to determine whether SIV proteins could induce or inhibit IFN-α directly. In vitro studies have shown that some HIV-1 proteins, eg, gp120, can directly induce IFN-α in human pDC.32 At the same time, through the recognition of a common epitope described for HIV-1 gp41 and IFN-α, antibodies against IFN-α could interfere with HIV-1 gp41 binding to cells, and vice versa, HIV gp41-specific antibodies could inhibit IFN-α function.33,34
Interestingly, however, although the various IFN-α subtypes were induced with different magnitudes, the overall expression pattern was similar in each of the lymphoid tissues tested in any individual animal. Thus, there was no tissue-specific expression of distinct IFN-α subtypes and their induction occurred with similar kinetics. Furthermore, we provide evidence that increased mRNA levels of IFN-α translated into the production of IFN-α protein and also the production of the IFN-α-inducible protein MxA (Fig. 5).
The increased pDC frequencies observed in several lymph nodes close to the virus entry site at 1 week after oral SIV exposure, and the established positive correlation between IFN-α mRNA levels and virus replication in the tonsil, suggested that this early IFN-α response in infant macaques was predominantly exerted by pDC (Fig. 6). This conclusion is supported by recent studies in adult macaques showing the fast influx of pDC into LNs in acute SIV infection35-37 and by demonstrating the role of pDC in the early IFN-α response.38 We cannot exclude the possibility, however, that differences in IFN-α production between various tissues could be the result of different frequencies of functionally impaired and/or dying pDC. Studies in SIV-infected macaques have shown that pDCs undergo rapid apoptotic cell death after migration into lymph nodes.35
It is noteworthy that infant macaques had significantly higher frequencies of pDC in LNs draining the oral cavity compared with pDC frequencies observed in the same tissues of six adult macaques (data not shown). One could speculate that the exploration of new things by mouth in infants might result in increased innate effector mechanisms at sites of oral exposure. In general, in both infant and adult macaques, pDC frequencies were highest in the spleen followed by LNs draining the oral cavity and lowest in the mesenteric LNs and in PBMC (data not shown).
In contrast to lymphoid tissues, mucosal tissues appeared less able to induce potent Type I IFN responses (Fig. 7). Although SIV infection induced slightly elevated mRNA levels for a number of IFN-α subtypes, within an individual animal, the magnitude was considerably lower compared with the increase of IFN-α subtype mRNA levels observed in lymphoid tissues. Consistent with the low IFN-α subtype mRNA induction in the gingiva and colon, IFN-α-positive cells could rarely be detected by immunohistochemistry (data not shown).
Assuming that pDC are responsible for the early IFN-α response in SIV infection, pDC frequencies and/or function must have differed between lymphoid and mucosal tissues of these animals. In fact, we present in vitro data demonstrating that TLR stimulation resulted in lower frequencies of cytokine-producing pDC in intestinal compared with PBMC and LN cell suspensions. This finding is consistent with data obtained in mouse studies.39-43 Considering that mucosal sites present the main entry sites for SIV and HIV, future studies need to address this question more thoroughly by comparing the functional ability of dendritic cells isolated from effector versus inductive sites within mucosal tissues and comparing them with dendritic cell function in lymphoid tissues of the same animal.
We thank Dr. Otto Hahn (University of Freiburg, Freiburg, Germany) for providing the MxA antibody and appreciate the technical assistance by Mathieu Lemieux and Joyce Lee and Kathy Lantz of the CNPRC Analytical Core.
1. Isaacs A, Lindenmann J. Virus interference. I. The interferon. Proc R Soc Ser B, London. 1957;147:258-267.
2. Brandt ER, Linnane AW, Devenish RJ. Expression of IFN A genes in subpopulations of peripheral blood cells. Br J Haematol. 1994;86:717-725.
3. Foster GR, Finter NB. Are all type I human interferons equivalent? J Viral Hepat. 1998;5:143-152.
4. Larrea E, Alberdi A, Castelruiz Y, et al. Expression of interferon-alpha subtypes in peripheral mononuclear cells from patients with chronic hepatitis C: a role for interferon-alpha5. J Viral Hepat. 2001;8:103-110.
5. Loseke S, Grage-Griebenow E, Wagner A, et al. Differential expression of IFN-alpha subtypes in human PBMC: evaluation of novel real-time PCR assays. J Immunol Methods. 2003;276:207-222.
6. Yeow WS, Lawson CM, Beilharz MW. Antiviral activities of individual murine IFN-alpha subtypes in vivo: intramuscular injection of IFN expression constructs reduces cytomegalovirus replication. J Immunol. 1998;160:2932-2939.
7. Cull VS, Bartlett EJ, James CM. Type I interferon gene therapy protects against cytomegalovirus-induced myocarditis. Immunology. 2002;106:428-437.
8. Carr DJ, Noisakran S. The antiviral efficacy of the murine alpha-1 interferon transgene against ocular herpes simplex virus type 1 requires the presence of CD4(+), alpha/beta T-cell receptor-positive T lymphocytes with the capacity to produce gamma interferon. J Virol. 2002;76:9398-9406.
9. James CM, Abdad MY, Mansfield JP, et al. Differential activities of alpha/beta IFN subtypes against influenza virus in vivo and enhancement of specific immune responses in DNA vaccinated mice expressing haemagglutinin and nucleoprotein. Vaccine. 2007;25:1856-1867.
10. Gerlach N, Gibbert K, Alter C, et al. Anti-retroviral effects of type I IFN subtypes in vivo. Eur J Immunol. 2009;39:136-146.
11. Abel K, Pahar B, Van Rompay KK, et al. Rapid virus dissemination in infant macaques after oral simian immunodeficiency virus exposure in the presence of local innate immune responses. J Virol. 2006;80:6357-6367.
12. Van Rompay KK, Schmidt KA, Lawson JR, et al. Topical administration of low-dose tenofovir disoproxil fumarate to protect infant macaques against multiple oral exposures of low doses of simian immunodeficiency virus. J Infect Dis. 2002;186:1508-1513.
13. Shacklett BL, Yang O, Hausner MA, et al. Optimization of methods to assess human mucosal T-cell responses to HIV infection. J Immunol Methods. 2003;279:17-31.
14. Szubin R, Chang WL, Greasby T, et al. Rigid interferon-alpha subtype responses of human plasmacytoid dendritic cells. J Interferon Cytokine Res. 2008;28:749-763.
15. Hall TA. BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucelic Acids Symposium Series. 1999;41:95-98.
16. Abel K, Rocke DM, Chohan B, et al. Temporal and anatomic relationship between virus replication and cytokine gene expression after vaginal simian immunodeficiency virus infection. J Virol. 2005;79:12164-12172.
17. Abel K, Wang Y, Fritts L, et al. Deoxycytidyl-deoxyguanosine oligonucleotide classes A, B, and C induce distinct cytokine gene expression patterns in rhesus monkey peripheral blood mononuclear cells and distinct alpha interferon responses in TLR9-expressing rhesus monkey plasmacytoid dendritic cells. Clin Diagn Lab Immunol. 2005;12:606-621.
18. Abel K, Alegria-Hartman MJ, Rothaeusler K, et al. The relationship between simian immunodeficiency virus RNA levels and the mRNA levels of alpha/beta interferons (IFN-alpha/beta) and IFN-alpha/beta-inducible Mx in lymphoid tissues of rhesus macaques during acute and chronic infection. J Virol. 2002;76:8433-8445.
19. Gibbs RA, Rogers J, Katze MG, et al. Evolutionary and biomedical insights from the rhesus macaque genome. Science. 2007;316:222-234.
20. Woelk CH, Frost SD, Richman DD, et al. Evolution of the interferon alpha gene family in eutherian mammals. Gene. 2007;397:38-50.
21. Chung E, Amrute SB, Abel K, et al. Characterization of virus-responsive plasmacytoid dendritic cells in the rhesus macaque. Clin Diagn Lab Immunol. 2005;12:426-435.
22. Werner JA. The lymph vessel system of the mouth cavity and pharynx [in German]. Laryngorhinootologie. 1995;74:622-628.
23. LaFleur DW, Nardelli B, Tsareva T, et al. Interferon-kappa, a novel type I interferon expressed in human keratinocytes. J Biol Chem. 2001;276:39765-39771.
24. Jorns C, Holzinger D, Thimme R, et al. Rapid and simple detection of IFN-neutralizing antibodies in chronic hepatitis C non-responsive to IFN-alpha. J Med Virol. 2006;78:74-82.
25. Sarasin-Filipowicz M, Oakeley EJ, Duong FH, et al. Interferon signaling and treatment outcome in chronic hepatitis C. Proc Natl Acad Sci U S A. 2008;105:7034-7039.
26. Hiscott J, Cantell K, Weissmann C. Differential expression of human interferon genes. Nucleic Acids Res. 1984;12:3727-3746.
27. Lopez S, Reeves R, Island ML, et al. Silencer activity in the interferon-A gene promoters. J Biol Chem. 1997;272:22788-22799.
28. Fuchs D, Shearer GM, Boswell RN, et al. Negative correlation between blood cell counts and serum neopterin concentration in patients with HIV-1 infection. AIDS. 1991;5:209-212.
29. von Sydow M, Sonnerborg A, Gaines H, et al. Interferon-alpha and tumor necrosis factor-alpha in serum of patients in various stages of HIV-1 infection. AIDS Res Hum Retroviruses. 1991;7:375-380.
30. Lehmann C, Taubert D, Jung N, et al. Preferential upregulation of interferon-alpha subtype 2 expression in HIV-1 patients. AIDS Res Hum Retroviruses. 2009;25:577-581.
31. Genin P, Vaccaro A, Civas A. The role of differential expression of human interferon--a genes in antiviral immunity. Cytokine Growth Factor Rev. 2009;20:283-295.
32. Del Corno M, Gauzzi MC, Penna G, et al. Human immunodeficiency virus type 1 gp120 and other activation stimuli are highly effective in triggering alpha interferon and CC chemokine production in circulating plasmacytoid but not myeloid dendritic cells. J Virol. 2005;79:12597-12601.
33. Chen YH, Xiao Y, Dierich MP. HIV-1 gp41 and type I interferon: sequence homology and biological as well as clinical implications. Immunol Res. 2000;22:61-66.
34. Chen YH, Wu W, Yang J, et al. Antibodies against human IFN-alpha and -beta recognized the immunosuppressive domain of HIV-1 gp41 and inhibit gp41-binding to the putative cellular receptor protein p45. Immunol Lett. 1999;69:253-257.
35. Brown KN, Wijewardana V, Liu X, et al. Rapid influx and death of plasmacytoid dendritic cells in lymph nodes mediate depletion in acute simian immunodeficiency virus infection. PLoS Pathog. 2009;5:e1000413.
36. Malleret B, Karlsson I, Maneglier B, et al. Effect of SIVmac infection on plasmacytoid and CD1c+ myeloid dendritic cells in cynomolgus macaques. Immunology. 2008;124:223-233.
37. Malleret B, Maneglier B, Karlsson I, et al. Primary infection with simian immunodeficiency virus: plasmacytoid dendritic cell homing to lymph nodes, type I interferon, and immune suppression. Blood. 2008;112:4598-4608.
38. Mandl JN, Barry AP, Vanderford TH, et al. Divergent TLR7 and TLR9 signaling and type I interferon production distinguish pathogenic and nonpathogenic AIDS virus infections. Nat Med. 2008;14:1077-1087.
39. Alpan O, Rudomen G, Matzinger P. The role of dendritic cells, B cells, and M cells in gut-oriented immune responses. J Immunol. 2001;166:4843-4852.
40. Chirdo FG, Millington OR, Beacock-Sharp H, et al. Immunomodulatory dendritic cells in intestinal lamina propria. Eur J Immunol. 2005;35:1831-1840.
41. Iwasaki A, Kelsall BL. Freshly isolated Peyer's patch, but not spleen, dendritic cells produce interleukin 10 and induce the differentiation of T helper type 2 cells. J Exp Med. 1999;190:229-239.
42. Iwasaki A, Kelsall BL. Mucosal immunity and inflammation. I. Mucosal dendritic cells: their specialized role in initiating T cell responses. Am J Physiol. 1999;276:G1074-G1078.
43. Iwasaki A, Kelsall BL. Localization of distinct Peyer's patch dendritic cell subsets and their recruitment by chemokines macrophage inflammatory protein (MIP)-3alpha, MIP-3beta, and secondary lymphoid organ chemokine. J Exp Med. 2000;191:1381-1394.
© 2010 Lippincott Williams & Wilkins, Inc.