Your Saturn-V example, it could send about 100,000-lb to the Moon.
Um, what? Thats' pretty much what the Apollo Saturn V stack sent to Luna
from the ground! 45 tons, no? That's the Apollo Lunar stack, CSM and LM.
Or are you saying that it could land 45 tons on the Lunar surface? That's a bit more like it, but let's see:
What follows is mainly a reply to the thread author, not brovane--perhaps brovane meant using the three stages of the Saturn V exclusively, not just to achieve TLI but then Lunar landing? Using remaining fuel in the S-2 as a crasher stage plus using the S-4B as lander?
In that scenario, I think the limiting factor is the thrust of the single J-2 engine on that stage, which governs how much mass could be landed, considering gravity losses which are complicated to say the least. Even then it ought to be feasible to land some 300 tons, which, doing a quick backwards estimate assuming the gravity loss of the descent is such that total delta-V from the translunar trajectory to the surface is 3500 m/sec, I figure the Saturn V can launch that 300 ton payload "atop" the three stages with 800 tons of propellant in the first stage to spare! (We might use it to boost an even bigger payload then, but I am dubious the final stage can land it--perhaps using that too as a crasher, we can still land over 300 tons on the Lunar surface). Interestingly, assuming it then can only use rockets of exhaust velocity of 3 km/sec to escape the Moon onto a direct Trans-Earth trajectory and then has to brake into LEO again, I do come up with about 40 tons returned to Earth orbit.
Clearly it all depends on the exact premise; below I assume we mean merely what can be launched on an Apollo-like fast TLI trajectory--3050 m/sec delta-V, considerably higher than the minimum needed for a Hohmann transfer.
As a rough rule of thumb, any launcher rocket to LEO typically achieves a total delta-V of about 10,000 m/sec. (brovane's references suggest 9000, which applies to high thrust launches I believe. Or to Skylon which proposes to use aerodynamic lift to trade off gravity loss for aerodynamic drag and comes out ahead doing so, because it is designed to fly aerodynamically). The orbital velocity is somewhat under 8000 m/sec, but we lose a lot to gravity loss and air drag. The air drag is relatively small, since the rocket gets above most atmospheric density pretty quick and still is not going at most of the speed it achieves yet when it does--the atmosphere still exists at LEO heights of course. But has a very low density there.
Gravity drag on the other hand is a lot more than 2000 m/sec, but it is at a right angle to the desired vector of motion, parallel to Earth's surface, so the two vectors add per the Pythagorean Theorem, the square root of the sum of the squares. Thus the gravity drag is really some 6000 m/sec, which is the accumulation of the net gravity force over the time of the burn. If we could boost at ten times the thrust, ten times the acceleration, we'd cut that down by a factor of ten and thus its Pythagorean addition would be pretty tiny then--of course first of all the high acceleration would tend to crush most cargoes, second of all engines that can realize those thrusts would add a significant amount of dry mass, and thirdly the air drag would be more significant.
So anyway a typical rocket, from Mercury or for that matter Explorer 1, to the STS, needs about 10,000 m/sec delta-V realistically.
Note that if we had magic antigravity this may or may not apply; if we can neutralize the force of gravity at no cost (or a one-time investment but no ongoing power cost) gravity loss doesn't exist; we can take our time reaching orbital velocity if we want to. We still need to somehow invest the energy it takes to reach orbital velocity, plus climbing up the potential well to clear the atmosphere--well if we have antigrav, arguably we don't need to be in orbit at all. I was going to respond to the "magic" approach but everyone has pointed out we don't need magic. To address that intelligently we need to specify how the "magic" works and most intuitive suggestions very quickly run into violations of very fundamental physical principles we don't want to abandon lightly. I'm not so sure how unphysical the suggestion I was going to make would be--it wouldn't be antigrav as such, but let that go for now. My notion would not encourage simply suspending masses hovering above the Earth anyway. We'd definitely want orbit.
Now then--the Saturn V achieved orbit in three stages. Looking at the mass breakdown given
here, we can see there are two versions of the rocket, the standard-issue three-stage one used to push the combination of the third stage (named S-4B--I can never remember these arbitrary stage names myself

) with about 60 percent of its fuel still not burned, plus the 45+ ton Apollo Lunar stack, to LEO parking orbit, and the modified version for Skylab (which again has a special name I don't recall, I'm sure many of the experts who have replied to you already do

) which just used two stages to lift the Skylab as a payload to orbit. I suspect the reason the Lunar mission version was designed that way was so that the second stage would not be in full orbit on burnout; firing the third stage briefly, instead of making it a bit smaller and the second a bit larger, resulted in the second stage being left far behind and burning up in the atmosphere, eliminating clutter in orbit. I suppose the Skylab version required extra retro-rockets on the second, "S-2" stage to get it out of the way once the Skylab was in orbit.
Skylab massed almost 103 metric tons, and the empty S-2 an additional 35, so overall almost 140 tons in orbit. The Apollo plus partially depleted S-4B (why the heck "S-4B?" Some historical reason I guess; it was never slated to be the fourth stage of anything, except maybe the shelved "Nova" rocket; could that be why? S-4 for Nova, S-4A for the multiple RL-10 version for the Saturn 1, S-4B when they switched to one J-2 engine?) massed about 45 tons (less for early missions, more for the last ones) plus about 60-65 for the third stage that would push this stack on to the Moon.
Let's look at the Skylab version then! You've had Scotty beam it up into orbit with a Transporter; never mind if the stages will fire correctly in free fall, assume they will, or you've attached small ullage engines to make it work, whatever.
In order to ask how much it could propel, you've got to specify what delta-V you want to reach. Let's say 3050, because you want to send a package of some mass to TLI. Since it is a two-stage rocket with the two stages having very different ISP engines it's a bit tricky to estimate....and it was! I went through many iterations, but I think you'll find that if the all-up mass of the stack in LEO is 4350 tons, the successive thrusts of the two stages will boost 1630 of it to just about that extra speed. The Skylab modified S-2 massed 36.25 tons dry so your payload toward the Moon is just under 1594.
Now I don't know what you'd want to do with this load or how you could land it on the Moon if that is the goal; if you could by whatever means put the 4000 plus tons into LEO in the first place (it would take a bit under 43 Saturn V launches to do it in pieces!

Energia could do about the same) you may have some really nifty means of landing it. If you had some kind of drive that could react against the aggregate mass of Earth or the Moon, setting it down might not involve expending energy at all--to the contrary, you'd want to stop it, and that means drawing power out of the kinetic energy. Quite a lot of energy; it would approach the Lunar surface at something like 3 km/sec (close to TLI delta-V in fact but that is a coincidence, depending on the mass of the Moon which has little to do with the TL trajectory)-we'd have to dispose of or store some 7 terajoules of energy. If we could store it you could use it later to launch the whole damn mass right back at Earth, then presumably brake it to orbit the same way (or store more energy, twice as much in fact, and land it on Earth, then bounce the same mass back directly to the Moon, and so on indefinitely using up hardly any energy at all).
Let's presume you have only rockets to land with; with such a ginormous mass of a craft we can presumably store liquid hydrogen for the half week it takes to get to Luna with little difficulty so suppose we land it with engines that expel propellant at 4400 m/sec and we have exactly 3200 m/sec to achieve, adding in a couple hundred m/sec for gravity loss and/or margin. We can then land a bit under half of it, or 770 tons, on the Lunar surface. If more than half of that is fuel for return we can by the same assumption launch 372 tons back to Earth and return 177 into LEO, using the same rockets in succession.
(Note I'm not allowing for moving into parking orbit first; with such a big payload presumably we are well past the initial exploratory stage and are moving it all with a purpose, which I suppose is on the Lunar surface. So we just go directly there, to a presumably known landing site. If the payload is meant for orbit instead, it can be a lot bigger since delta-V for that is much lower).
This by the way is why you don't want to use a Saturn V for the initial boost; the first stage, massing some 2/3 of the whole thing, is designed to get the stack off the pad and up out of the atmosphere (mostly); for this we want high thrust and can settle for a mediocre ISP if it saves us tank mass and engine mass, as it does. For operations once we reach orbit (or indeed, for achieving orbit, once we've bought time to burn in vacuum with the initial thrust of the first stage) we can afford to take more time to burn, and thus save on engine mass, and the question of ISP becomes much more pointed. What we'd want, to move your 1594 tons to Luna--let's call it 1600 tons even--is to use the best ISP hydrogen-oxygen engines we can, and we aren't so worried about thrusting quickly. I judge that if you can achieve an orbital transfer in less time than the original orbit would have taken to traverse two radians, or about 1/3 of a circular orbit, the outcome is very similar to doing it instantaneously, which is most efficient. So we have some 2000 seconds to achieve 3050 m/sec velocity change; if the average push is just 2 m/sec or about 1/5 of a G acceleration we're good. Thrusts like that are worse than useless on the ground of course!
The J-2s are not the best possible engines then; I went with 4400 m/sec above because that is the ballpark of the
Centaur RL-10 engines which were already on the shelf by 1969--they produce a lot less thrust of course, being much smaller. But we can use clusters of them for a project like this! Modern RL-10s get a thrust/weight ratio of 60 and the tankage and other stage masses of Centaurs come to about 1/10 the propellant mass. The ISP of the modern ones is even a bit higher than I thought; 4550 m/sec effective exhaust speed; actually when I compare the thrust to the propellant consumption rate it seems a bit higher still. closer to 4570 m/sec! Going with 4560 then for a 3050 delta-V, the gross mass ratio is 1.952; with a propellant consumption rate of 24.1 kg/sec per engine, we'd need 33 of them to use 1600 tons of propellant within 2000 seconds, so thrust would be over 3.6 meganewtons (more than 3 J-2, less than 4--so see that even the second Saturn stage with its 5 engines is overpowered for this application); the engines would mass 5.6 tons altogether. The tankage would be about 160 tons.
A single stage massing 175 tons dry, filled with 1690 tons of propellant, could move itself and a 1600 ton payload to just over the specified delta-V in 2125 seconds of burn time; we can bring it down to 2000 using 35 engines with minimal penalty. All up the thing is 3465 tons in LEO, a considerable savings over the Saturn stack version! Staging might bring it down more, but the mass fraction of dry stages might rise to offset the benefits so I wouldn't. Attaching those 35 engines to the payload stage would make them available for the Moon landing; that's a bit marginal since the thrust would initially allow accelerations only marginally above Lunar gravity, but by the time landing approaches I'd think the mass would be down quite a lot and their thrust would therefore probably be adequate--if not, use even more engines.
What you are talking about is Delta-V budget. So for example the Delta-V required to go from LEO to Lunar Orbit is 4.04 km/sec. Go from Lunar Orbit to lunar surface is 1.87.
I think you need a bit more for LLO to the surface or back again--one absolutely has to have some safety margin. Well anyway the Apollo missions, landing at unknown sites, did-- I think the spec was for at least 2000 m/sec, possibly 2200, each way.
http://en.wikipedia.org/wiki/Delta-v_budget
The easiest I have found is to Download this Delta-V calculator here
http://home.arcor.de/francisdrakex/download/
You can then use this to build a spacecraft add propellant etc and calculate what it takes to move Mass around the Solar System.
One should note, the table in the Wikipedia entry is for minimal velocity changes--TLI from LEO to Luna is I presume for a Hohmann transfer to Lunar orbit. The lower velocity change would allow even more mass than I estimated above to be moved!

The price we'd pay would be a slower trip than Apollo, some nine days to Luna instead of three and a bit. Moreover Apollo chose the faster, more expensive transfer not only to save time, but to enable free-return orbits in case something went wrong (as with Apollo 13). A Hohmann orbit would be free-return if the Moon's mass were negligible, but since it isn't the close encounter with Luna would throw the orbit out of whack if for any reason the craft just flew on by; the new orbit it would be in would not return it to the same low altitude perigee it launched from. The more energetic transfers Apollo used did allow for the path to whip around in just such a fashion as to send the capsule back to an altitude from which it could land on Earth.
That's probably not important for a 1600+ ton mission though!

Redundant systems should guarantee that the landing mission goes forward. Nor is time saving as critical; on such a craft any crew would have ample supplies and many opportunities for sheltering from radiation, which are the concerns governing our desire to make the transits faster.